Comparing Pixel and Object-Based Approaches to Map an Understorey Invasive Shrub in Tropical Mixed Forests

PubMed Central

Niphadkar, Madhura; Nagendra, Harini; Tarantino, Cristina; Adamo, Maria; Blonda, Palma

2017-01-01

The establishment of invasive alien species in varied habitats across the world is now recognized as a genuine threat to the preservation of biodiversity. Specifically, plant invasions in understory tropical forests are detrimental to the persistence of healthy ecosystems. Monitoring such invasions using Very High Resolution (VHR) satellite remote sensing has been shown to be valuable in designing management interventions for conservation of native habitats. Object-based classification methods are very helpful in identifying invasive plants in various habitats, by their inherent nature of imitating the ability of the human brain in pattern recognition. However, these methods have not been tested adequately in dense tropical mixed forests where invasion occurs in the understorey. This study compares a pixel-based and object-based classification method for mapping the understorey invasive shrub Lantana camara (Lantana) in a tropical mixed forest habitat in the Western Ghats biodiversity hotspot in India. Overall, a hierarchical approach of mapping top canopy at first, and then further processing for the understorey shrub, using measures such as texture and vegetation indices proved effective in separating out Lantana from other cover types. In the first method, we implement a simple parametric supervised classification for mapping cover types, and then process within these types for Lantana delineation. In the second method, we use an object-based segmentation algorithm to map cover types, and then perform further processing for separating Lantana. The improved ability of the object-based approach to delineate structurally distinct objects with characteristic spectral and spatial characteristics of their own, as well as with reference to their surroundings, allows for much flexibility in identifying invasive understorey shrubs among the complex vegetation of the tropical forest than that provided by the parametric classifier. Conservation practices in tropical mixed forests can benefit greatly by adopting methods which use high resolution remotely sensed data and advanced techniques to monitor the patterns and effective functioning of native ecosystems by periodically mapping disturbances such as invasion. PMID:28620400

a Data Field Method for Urban Remotely Sensed Imagery Classification Considering Spatial Correlation

NASA Astrophysics Data System (ADS)

Zhang, Y.; Qin, K.; Zeng, C.; Zhang, E. B.; Yue, M. X.; Tong, X.

2016-06-01

Spatial correlation between pixels is important information for remotely sensed imagery classification. Data field method and spatial autocorrelation statistics have been utilized to describe and model spatial information of local pixels. The original data field method can represent the spatial interactions of neighbourhood pixels effectively. However, its focus on measuring the grey level change between the central pixel and the neighbourhood pixels results in exaggerating the contribution of the central pixel to the whole local window. Besides, Geary's C has also been proven to well characterise and qualify the spatial correlation between each pixel and its neighbourhood pixels. But the extracted object is badly delineated with the distracting salt-and-pepper effect of isolated misclassified pixels. To correct this defect, we introduce the data field method for filtering and noise limitation. Moreover, the original data field method is enhanced by considering each pixel in the window as the central pixel to compute statistical characteristics between it and its neighbourhood pixels. The last step employs a support vector machine (SVM) for the classification of multi-features (e.g. the spectral feature and spatial correlation feature). In order to validate the effectiveness of the developed method, experiments are conducted on different remotely sensed images containing multiple complex object classes inside. The results show that the developed method outperforms the traditional method in terms of classification accuracies.

Corn and soybean Landsat MSS classification performance as a function of scene characteristics

NASA Technical Reports Server (NTRS)

Batista, G. T.; Hixson, M. M.; Bauer, M. E.

1982-01-01

In order to fully utilize remote sensing to inventory crop production, it is important to identify the factors that affect the accuracy of Landsat classifications. The objective of this study was to investigate the effect of scene characteristics involving crop, soil, and weather variables on the accuracy of Landsat classifications of corn and soybeans. Segments sampling the U.S. Corn Belt were classified using a Gaussian maximum likelihood classifier on multitemporally registered data from two key acquisition periods. Field size had a strong effect on classification accuracy with small fields tending to have low accuracies even when the effect of mixed pixels was eliminated. Other scene characteristics accounting for variability in classification accuracy included proportions of corn and soybeans, crop diversity index, proportion of all field crops, soil drainage, slope, soil order, long-term average soybean yield, maximum yield, relative position of the segment in the Corn Belt, weather, and crop development stage.

Estimation of proportions in mixed pixels through their region characterization

NASA Technical Reports Server (NTRS)

Chittineni, C. B. (Principal Investigator)

1981-01-01

A region of mixed pixels can be characterized through the probability density function of proportions of classes in the pixels. Using information from the spectral vectors of a given set of pixels from the mixed pixel region, expressions are developed for obtaining the maximum likelihood estimates of the parameters of probability density functions of proportions. The proportions of classes in the mixed pixels can then be estimated. If the mixed pixels contain objects of two classes, the computation can be reduced by transforming the spectral vectors using a transformation matrix that simultaneously diagonalizes the covariance matrices of the two classes. If the proportions of the classes of a set of mixed pixels from the region are given, then expressions are developed for obtaining the estmates of the parameters of the probability density function of the proportions of mixed pixels. Development of these expressions is based on the criterion of the minimum sum of squares of errors. Experimental results from the processing of remotely sensed agricultural multispectral imagery data are presented.

Classification of multispectral image data by the Binary Diamond neural network and by nonparametric, pixel-by-pixel methods

NASA Technical Reports Server (NTRS)

Salu, Yehuda; Tilton, James

1993-01-01

The classification of multispectral image data obtained from satellites has become an important tool for generating ground cover maps. This study deals with the application of nonparametric pixel-by-pixel classification methods in the classification of pixels, based on their multispectral data. A new neural network, the Binary Diamond, is introduced, and its performance is compared with a nearest neighbor algorithm and a back-propagation network. The Binary Diamond is a multilayer, feed-forward neural network, which learns from examples in unsupervised, 'one-shot' mode. It recruits its neurons according to the actual training set, as it learns. The comparisons of the algorithms were done by using a realistic data base, consisting of approximately 90,000 Landsat 4 Thematic Mapper pixels. The Binary Diamond and the nearest neighbor performances were close, with some advantages to the Binary Diamond. The performance of the back-propagation network lagged behind. An efficient nearest neighbor algorithm, the binned nearest neighbor, is described. Ways for improving the performances, such as merging categories, and analyzing nonboundary pixels, are addressed and evaluated.

Object based image analysis for the classification of the growth stages of Avocado crop, in Michoacán State, Mexico

NASA Astrophysics Data System (ADS)

Gao, Yan; Marpu, Prashanth; Morales Manila, Luis M.

2014-11-01

This paper assesses the suitability of 8-band Worldview-2 (WV2) satellite data and object-based random forest algorithm for the classification of avocado growth stages in Mexico. We tested both pixel-based with minimum distance (MD) and maximum likelihood (MLC) and object-based with Random Forest (RF) algorithm for this task. Training samples and verification data were selected by visual interpreting the WV2 images for seven thematic classes: fully grown, middle stage, and early stage of avocado crops, bare land, two types of natural forests, and water body. To examine the contribution of the four new spectral bands of WV2 sensor, all the tested classifications were carried out with and without the four new spectral bands. Classification accuracy assessment results show that object-based classification with RF algorithm obtained higher overall higher accuracy (93.06%) than pixel-based MD (69.37%) and MLC (64.03%) method. For both pixel-based and object-based methods, the classifications with the four new spectral bands (overall accuracy obtained higher accuracy than those without: overall accuracy of object-based RF classification with vs without: 93.06% vs 83.59%, pixel-based MD: 69.37% vs 67.2%, pixel-based MLC: 64.03% vs 36.05%, suggesting that the four new spectral bands in WV2 sensor contributed to the increase of the classification accuracy.

Land Cover Analysis by Using Pixel-Based and Object-Based Image Classification Method in Bogor

NASA Astrophysics Data System (ADS)

Amalisana, Birohmatin; Rokhmatullah; Hernina, Revi

2017-12-01

The advantage of image classification is to provide earth’s surface information like landcover and time-series changes. Nowadays, pixel-based image classification technique is commonly performed with variety of algorithm such as minimum distance, parallelepiped, maximum likelihood, mahalanobis distance. On the other hand, landcover classification can also be acquired by using object-based image classification technique. In addition, object-based classification uses image segmentation from parameter such as scale, form, colour, smoothness and compactness. This research is aimed to compare the result of landcover classification and its change detection between parallelepiped pixel-based and object-based classification method. Location of this research is Bogor with 20 years range of observation from 1996 until 2016. This region is famous as urban areas which continuously change due to its rapid development, so that time-series landcover information of this region will be interesting.

Scattering property based contextual PolSAR speckle filter

NASA Astrophysics Data System (ADS)

Mullissa, Adugna G.; Tolpekin, Valentyn; Stein, Alfred

2017-12-01

Reliability of the scattering model based polarimetric SAR (PolSAR) speckle filter depends upon the accurate decomposition and classification of the scattering mechanisms. This paper presents an improved scattering property based contextual speckle filter based upon an iterative classification of the scattering mechanisms. It applies a Cloude-Pottier eigenvalue-eigenvector decomposition and a fuzzy H/α classification to determine the scattering mechanisms on a pre-estimate of the coherency matrix. The H/α classification identifies pixels with homogeneous scattering properties. A coarse pixel selection rule groups pixels that are either single bounce, double bounce or volume scatterers. A fine pixel selection rule is applied to pixels within each canonical scattering mechanism. We filter the PolSAR data and depending on the type of image scene (urban or rural) use either the coarse or fine pixel selection rule. Iterative refinement of the Wishart H/α classification reduces the speckle in the PolSAR data. Effectiveness of this new filter is demonstrated by using both simulated and real PolSAR data. It is compared with the refined Lee filter, the scattering model based filter and the non-local means filter. The study concludes that the proposed filter compares favorably with other polarimetric speckle filters in preserving polarimetric information, point scatterers and subtle features in PolSAR data.

Mixed deep learning and natural language processing method for fake-food image recognition and standardization to help automated dietary assessment.

PubMed

Mezgec, Simon; Eftimov, Tome; Bucher, Tamara; Koroušić Seljak, Barbara

2018-04-06

The present study tested the combination of an established and a validated food-choice research method (the 'fake food buffet') with a new food-matching technology to automate the data collection and analysis. The methodology combines fake-food image recognition using deep learning and food matching and standardization based on natural language processing. The former is specific because it uses a single deep learning network to perform both the segmentation and the classification at the pixel level of the image. To assess its performance, measures based on the standard pixel accuracy and Intersection over Union were applied. Food matching firstly describes each of the recognized food items in the image and then matches the food items with their compositional data, considering both their food names and their descriptors. The final accuracy of the deep learning model trained on fake-food images acquired by 124 study participants and providing fifty-five food classes was 92·18 %, while the food matching was performed with a classification accuracy of 93 %. The present findings are a step towards automating dietary assessment and food-choice research. The methodology outperforms other approaches in pixel accuracy, and since it is the first automatic solution for recognizing the images of fake foods, the results could be used as a baseline for possible future studies. As the approach enables a semi-automatic description of recognized food items (e.g. with respect to FoodEx2), these can be linked to any food composition database that applies the same classification and description system.

GENIE: a hybrid genetic algorithm for feature classification in multispectral images

NASA Astrophysics Data System (ADS)

Perkins, Simon J.; Theiler, James P.; Brumby, Steven P.; Harvey, Neal R.; Porter, Reid B.; Szymanski, John J.; Bloch, Jeffrey J.

2000-10-01

We consider the problem of pixel-by-pixel classification of a multi- spectral image using supervised learning. Conventional spuervised classification techniques such as maximum likelihood classification and less conventional ones s uch as neural networks, typically base such classifications solely on the spectral components of each pixel. It is easy to see why: the color of a pixel provides a nice, bounded, fixed dimensional space in which these classifiers work well. It is often the case however, that spectral information alone is not sufficient to correctly classify a pixel. Maybe spatial neighborhood information is required as well. Or maybe the raw spectral components do not themselves make for easy classification, but some arithmetic combination of them would. In either of these cases we have the problem of selecting suitable spatial, spectral or spatio-spectral features that allow the classifier to do its job well. The number of all possible such features is extremely large. How can we select a suitable subset? We have developed GENIE, a hybrid learning system that combines a genetic algorithm that searches a space of image processing operations for a set that can produce suitable feature planes, and a more conventional classifier which uses those feature planes to output a final classification. In this paper we show that the use of a hybrid GA provides significant advantages over using either a GA alone or more conventional classification methods alone. We present results using high-resolution IKONOS data, looking for regions of burned forest and for roads.

A comparative analysis of pixel- and object-based detection of landslides from very high-resolution images

NASA Astrophysics Data System (ADS)

Keyport, Ren N.; Oommen, Thomas; Martha, Tapas R.; Sajinkumar, K. S.; Gierke, John S.

2018-02-01

A comparative analysis of landslides detected by pixel-based and object-oriented analysis (OOA) methods was performed using very high-resolution (VHR) remotely sensed aerial images for the San Juan La Laguna, Guatemala, which witnessed widespread devastation during the 2005 Hurricane Stan. A 3-band orthophoto of 0.5 m spatial resolution together with a 115 field-based landslide inventory were used for the analysis. A binary reference was assigned with a zero value for landslide and unity for non-landslide pixels. The pixel-based analysis was performed using unsupervised classification, which resulted in 11 different trial classes. Detection of landslides using OOA includes 2-step K-means clustering to eliminate regions based on brightness; elimination of false positives using object properties such as rectangular fit, compactness, length/width ratio, mean difference of objects, and slope angle. Both overall accuracy and F-score for OOA methods outperformed pixel-based unsupervised classification methods in both landslide and non-landslide classes. The overall accuracy for OOA and pixel-based unsupervised classification was 96.5% and 94.3%, respectively, whereas the best F-score for landslide identification for OOA and pixel-based unsupervised methods: were 84.3% and 77.9%, respectively.Results indicate that the OOA is able to identify the majority of landslides with a few false positive when compared to pixel-based unsupervised classification.

Mapping shorelines to subpixel accuracy using Landsat imagery

NASA Astrophysics Data System (ADS)

Abileah, Ron; Vignudelli, Stefano; Scozzari, Andrea

2013-04-01

A promising method to accurately map the shoreline of oceans, lakes, reservoirs, and rivers is proposed and verified in this work. The method is applied to multispectral satellite imagery in two stages. The first stage is a classification of each image pixel into land/water categories using the conventional 'dark pixel' method. The approach presented here, makes use of a single shortwave IR image band (SWIR), if available. It is well known that SWIR has the least water leaving radiance and relatively little sensitivity to water pollutants and suspended sediments. It is generally the darkest (over water) and most reliable single band for land-water discrimination. The boundary of the water cover map determined in stage 1 underestimates the water cover and often misses the true shoreline by a quantity up to one pixel. A more accurate shoreline would be obtained by connecting the center point of pixels with exactly 50-50 mix of water and land. Then, stage 2 finds the 50-50 mix points. According to the method proposed, image data is interpolated and up-sampled to ten times the original resolution. The local gradient in radiance is used to find the direction to the shore, thus searching along that path for the interpolated pixel closest to a 50-50 mix. Landsat images with 30m resolution, processed by this method, may thus provide the shoreline accurate to 3m. Compared to similar approaches available in the literature, the method proposed discriminates sub-pixels crossed by the shoreline by using a criteria based on the absolute value of radiance, rather than its gradient. Preliminary experimentation of the algorithm shows that 10m resolution accuracy is easily achieved and in some cases is often better than 5m. The proposed method can be used to study long term shoreline changes by exploiting the 30 years of archived world-wide coverage Landsat imagery. Landsat imagery is free and easily accessible for downloading. Some applications that exploit the Landsat dataset and the new method are discussed in the companion poster: "Case-studies of potential applications for highly resolved shorelines."

Multi-Pixel Simultaneous Classification of PolSAR Image Using Convolutional Neural Networks

PubMed Central

Xu, Xin; Gui, Rong; Pu, Fangling

2018-01-01

Convolutional neural networks (CNN) have achieved great success in the optical image processing field. Because of the excellent performance of CNN, more and more methods based on CNN are applied to polarimetric synthetic aperture radar (PolSAR) image classification. Most CNN-based PolSAR image classification methods can only classify one pixel each time. Because all the pixels of a PolSAR image are classified independently, the inherent interrelation of different land covers is ignored. We use a fixed-feature-size CNN (FFS-CNN) to classify all pixels in a patch simultaneously. The proposed method has several advantages. First, FFS-CNN can classify all the pixels in a small patch simultaneously. When classifying a whole PolSAR image, it is faster than common CNNs. Second, FFS-CNN is trained to learn the interrelation of different land covers in a patch, so it can use the interrelation of land covers to improve the classification results. The experiments of FFS-CNN are evaluated on a Chinese Gaofen-3 PolSAR image and other two real PolSAR images. Experiment results show that FFS-CNN is comparable with the state-of-the-art PolSAR image classification methods. PMID:29510499

Multi-Pixel Simultaneous Classification of PolSAR Image Using Convolutional Neural Networks.

PubMed

Wang, Lei; Xu, Xin; Dong, Hao; Gui, Rong; Pu, Fangling

2018-03-03

Convolutional neural networks (CNN) have achieved great success in the optical image processing field. Because of the excellent performance of CNN, more and more methods based on CNN are applied to polarimetric synthetic aperture radar (PolSAR) image classification. Most CNN-based PolSAR image classification methods can only classify one pixel each time. Because all the pixels of a PolSAR image are classified independently, the inherent interrelation of different land covers is ignored. We use a fixed-feature-size CNN (FFS-CNN) to classify all pixels in a patch simultaneously. The proposed method has several advantages. First, FFS-CNN can classify all the pixels in a small patch simultaneously. When classifying a whole PolSAR image, it is faster than common CNNs. Second, FFS-CNN is trained to learn the interrelation of different land covers in a patch, so it can use the interrelation of land covers to improve the classification results. The experiments of FFS-CNN are evaluated on a Chinese Gaofen-3 PolSAR image and other two real PolSAR images. Experiment results show that FFS-CNN is comparable with the state-of-the-art PolSAR image classification methods.

Methods for Monitoring the Detection of Multi-Temporal Land Use Change Through the Classification of Urban Areas

NASA Astrophysics Data System (ADS)

Alhaddad, B. I.; Burns, M. C.; Roca, J.

2011-08-01

Urban areas are complicated due to the mix of man-made features and natural features. A higher level of structural information plays an important role in land cover/use classification of urban areas. Additional spatial indicators have to be extracted based on structural analysis in order to understand and identify spatial patterns or the spatial organization of features, especially for man-made feature. It's very difficult to extract such spatial patterns by using only classification approaches. Clusters of urban patterns which are integral parts of other uses may be difficult to identify. A lot of public resources have been directed towards seeking to develop a standardized classification system and to provide as much compatibility as possible to ensure the widespread use of such categorized data obtained from remote sensor sources. In this paper different methods applied to understand the change in the land use areas by understanding and monitoring the change in urban areas and as its hard to apply those methods to classification results for high elements quantities, dusts and scratches (Roca and Alhaddad, 2005). This paper focuses on a methodology developed based relation between urban elements and how to join this elements in zones or clusters have commune behaviours such as form, pattern, size. The main objective is to convert urban class category in to various structure densities depend on conjunction of pixel and shortest distance between them, Delaunay triangulation has been widely used in spatial analysis and spatial modelling. To identify these different zones, a spatial density-based clustering technique was adopted. In highly urban zones, the spatial density of the pixels is high, while in sparsely areas the density of points is much lower. Once the groups of pixels are identified, the calculation of the boundaries of the areas containing each group of pixels defines the new regions indicate the different contains inside such as high or low urban areas. Multi-temporal datasets from 1986, 1995 and 2004 used to urban region centroid to be our reference in this study which allow us to follow the urban movement, increase and decrease by the time. Kernel Density function used to Calculates urban magnitude, Voronoi algorithm is proposed for deriving explicit boundaries between objects units. To test the approach, we selected a site in a suburban area in Barcelona Municipality, the Spain.

Defect detection and classification of galvanized stamping parts based on fully convolution neural network

NASA Astrophysics Data System (ADS)

Xiao, Zhitao; Leng, Yanyi; Geng, Lei; Xi, Jiangtao

2018-04-01

In this paper, a new convolution neural network method is proposed for the inspection and classification of galvanized stamping parts. Firstly, all workpieces are divided into normal and defective by image processing, and then the defective workpieces extracted from the region of interest (ROI) area are input to the trained fully convolutional networks (FCN). The network utilizes an end-to-end and pixel-to-pixel training convolution network that is currently the most advanced technology in semantic segmentation, predicts result of each pixel. Secondly, we mark the different pixel values of the workpiece, defect and background for the training image, and use the pixel value and the number of pixels to realize the recognition of the defects of the output picture. Finally, the defect area's threshold depended on the needs of the project is set to achieve the specific classification of the workpiece. The experiment results show that the proposed method can successfully achieve defect detection and classification of galvanized stamping parts under ordinary camera and illumination conditions, and its accuracy can reach 99.6%. Moreover, it overcomes the problem of complex image preprocessing and difficult feature extraction and performs better adaptability.

Supervised pixel classification using a feature space derived from an artificial visual system

NASA Technical Reports Server (NTRS)

Baxter, Lisa C.; Coggins, James M.

1991-01-01

Image segmentation involves labelling pixels according to their membership in image regions. This requires the understanding of what a region is. Using supervised pixel classification, the paper investigates how groups of pixels labelled manually according to perceived image semantics map onto the feature space created by an Artificial Visual System. Multiscale structure of regions are investigated and it is shown that pixels form clusters based on their geometric roles in the image intensity function, not by image semantics. A tentative abstract definition of a 'region' is proposed based on this behavior.

Digital mammography: observer performance study of the effects of pixel size on radiologists' characterization of malignant and benign microcalcifications

NASA Astrophysics Data System (ADS)

Chan, Heang-Ping; Helvie, Mark A.; Petrick, Nicholas; Sahiner, Berkman; Adler, Dorit D.; Blane, Caroline E.; Joynt, Lynn K.; Paramagul, Chintana; Roubidoux, Marilyn A.; Wilson, Todd E.; Hadjiiski, Lubomir M.; Goodsitt, Mitchell M.

1999-05-01

A receiver operating characteristic (ROC) experiment was conducted to evaluate the effects of pixel size on the characterization of mammographic microcalcifications. Digital mammograms were obtained by digitizing screen-film mammograms with a laser film scanner. One hundred twelve two-view mammograms with biopsy-proven microcalcifications were digitized at a pixel size of 35 micrometer X 35 micrometer. A region of interest (ROI) containing the microcalcifications was extracted from each image. ROI images with pixel sizes of 70 micrometers, 105 micrometers, and 140 micrometers were derived from the ROI of 35 micrometer pixel size by averaging 2 X 2, 3 X 3, and 4 X 4 neighboring pixels, respectively. The ROI images were printed on film with a laser imager. Seven MQSA-approved radiologists participated as observers. The likelihood of malignancy of the microcalcifications was rated on a 10-point confidence rating scale and analyzed with ROC methodology. The classification accuracy was quantified by the area, Az, under the ROC curve. The statistical significance of the differences in the Az values for different pixel sizes was estimated with the Dorfman-Berbaum-Metz (DBM) method for multi-reader, multi-case ROC data. It was found that five of the seven radiologists demonstrated a higher classification accuracy with the 70 micrometer or 105 micrometer images. The average Az also showed a higher classification accuracy in the range of 70 to 105 micrometer pixel size. However, the differences in A(subscript z/ between different pixel sizes did not achieve statistical significance. The low specificity of image features of microcalcifications an the large interobserver and intraobserver variabilities may have contributed to the relatively weak dependence of classification accuracy on pixel size.

Identification of coffee bean varieties using hyperspectral imaging: influence of preprocessing methods and pixel-wise spectra analysis.

PubMed

Zhang, Chu; Liu, Fei; He, Yong

2018-02-01

Hyperspectral imaging was used to identify and to visualize the coffee bean varieties. Spectral preprocessing of pixel-wise spectra was conducted by different methods, including moving average smoothing (MA), wavelet transform (WT) and empirical mode decomposition (EMD). Meanwhile, spatial preprocessing of the gray-scale image at each wavelength was conducted by median filter (MF). Support vector machine (SVM) models using full sample average spectra and pixel-wise spectra, and the selected optimal wavelengths by second derivative spectra all achieved classification accuracy over 80%. Primarily, the SVM models using pixel-wise spectra were used to predict the sample average spectra, and these models obtained over 80% of the classification accuracy. Secondly, the SVM models using sample average spectra were used to predict pixel-wise spectra, but achieved with lower than 50% of classification accuracy. The results indicated that WT and EMD were suitable for pixel-wise spectra preprocessing. The use of pixel-wise spectra could extend the calibration set, and resulted in the good prediction results for pixel-wise spectra and sample average spectra. The overall results indicated the effectiveness of using spectral preprocessing and the adoption of pixel-wise spectra. The results provided an alternative way of data processing for applications of hyperspectral imaging in food industry.

Sub-pixel mapping of hyperspectral imagery using super-resolution

NASA Astrophysics Data System (ADS)

Sharma, Shreya; Sharma, Shakti; Buddhiraju, Krishna M.

2016-04-01

With the development of remote sensing technologies, it has become possible to obtain an overview of landscape elements which helps in studying the changes on earth's surface due to climate, geological, geomorphological and human activities. Remote sensing measures the electromagnetic radiations from the earth's surface and match the spectral similarity between the observed signature and the known standard signatures of the various targets. However, problem lies when image classification techniques assume pixels to be pure. In hyperspectral imagery, images have high spectral resolution but poor spatial resolution. Therefore, the spectra obtained is often contaminated due to the presence of mixed pixels and causes misclassification. To utilise this high spectral information, spatial resolution has to be enhanced. Many factors make the spatial resolution one of the most expensive and hardest to improve in imaging systems. To solve this problem, post-processing of hyperspectral images is done to retrieve more information from the already acquired images. The algorithm to enhance spatial resolution of the images by dividing them into sub-pixels is known as super-resolution and several researches have been done in this domain.In this paper, we propose a new method for super-resolution based on ant colony optimization and review the popular methods of sub-pixel mapping of hyperspectral images along with their comparative analysis.

Investigation of correlation classification techniques

NASA Technical Reports Server (NTRS)

Haskell, R. E.

1975-01-01

A two-step classification algorithm for processing multispectral scanner data was developed and tested. The first step is a single pass clustering algorithm that assigns each pixel, based on its spectral signature, to a particular cluster. The output of that step is a cluster tape in which a single integer is associated with each pixel. The cluster tape is used as the input to the second step, where ground truth information is used to classify each cluster using an iterative method of potentials. Once the clusters have been assigned to classes the cluster tape is read pixel-by-pixel and an output tape is produced in which each pixel is assigned to its proper class. In addition to the digital classification programs, a method of using correlation clustering to process multispectral scanner data in real time by means of an interactive color video display is also described.

Where can pixel counting area estimates meet user-defined accuracy requirements?

NASA Astrophysics Data System (ADS)

Waldner, François; Defourny, Pierre

2017-08-01

Pixel counting is probably the most popular way to estimate class areas from satellite-derived maps. It involves determining the number of pixels allocated to a specific thematic class and multiplying it by the pixel area. In the presence of asymmetric classification errors, the pixel counting estimator is biased. The overarching objective of this article is to define the applicability conditions of pixel counting so that the estimates are below a user-defined accuracy target. By reasoning in terms of landscape fragmentation and spatial resolution, the proposed framework decouples the resolution bias and the classifier bias from the overall classification bias. The consequence is that prior to any classification, part of the tolerated bias is already committed due to the choice of the spatial resolution of the imagery. How much classification bias is affordable depends on the joint interaction of spatial resolution and fragmentation. The method was implemented over South Africa for cropland mapping, demonstrating its operational applicability. Particular attention was paid to modeling a realistic sensor's spatial response by explicitly accounting for the effect of its point spread function. The diagnostic capabilities offered by this framework have multiple potential domains of application such as guiding users in their choice of imagery and providing guidelines for space agencies to elaborate the design specifications of future instruments.

Modeling misregistration and related effects on multispectral classification

NASA Technical Reports Server (NTRS)

Billingsley, F. C.

1981-01-01

The effects of misregistration on the multispectral classification accuracy when the scene registration accuracy is relaxed from 0.3 to 0.5 pixel are investigated. Noise, class separability, spatial transient response, and field size are considered simultaneously with misregistration in their effects on accuracy. Any noise due to the scene, sensor, or to the analog/digital conversion, causes a finite fraction of the measurements to fall outside of the classification limits, even within nominally uniform fields. Misregistration causes field borders in a given band or set of bands to be closer than expected to a given pixel, causing additional pixels to be misclassified due to the mixture of materials in the pixel. Simplified first order models of the various effects are presented, and are used to estimate the performance to be expected.

Evaluation of Aster Images for Characterization and Mapping of Amethyst Mining Residues

NASA Astrophysics Data System (ADS)

Markoski, P. R.; Rolim, S. B. A.

2012-07-01

The objective of this work was to evaluate the potential of Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER), subsystems VNIR (Visible and Near Infrared) and SWIR (Short Wave Infrared) images, for discrimination and mapping of amethyst mining residues (basalt) in the Ametista do Sul Region, Rio Grande do Sul State, Brazil. This region provides the most part of amethyst mining of the World. The basalt is extracted during the mining process and deposited outside the mine. As a result, mounts of residues (basalt) rise up. These mounts are many times smaller than ASTER pixel size (VNIR - 15 meters and SWIR - 30 meters). Thus, the pixel composition becomes a mixing of various materials, hampering its identification and mapping. Trying to solve this problem, multispectral algorithm Maximum Likelihood (MaxVer) and the hyperspectral technique SAM (Spectral Angle Mapper) were used in this work. Images from ASTER subsystems VNIR and SWIR were used to perform the classifications. SAM technique produced better results than MaxVer algorithm. The main error found by the techniques was the mixing between "shadow" and "mining residues/basalt" classes. With the SAM technique the confusion decreased because it employed the basalt spectral curve as a reference, while the multispectral techniques employed pixels groups that could have spectral mixture with other targets. The results showed that in tropical terrains as the study area, ASTER data can be efficacious for the characterization of mining residues.

Point spread function based classification of regions for linear digital tomosynthesis

NASA Astrophysics Data System (ADS)

Israni, Kenny; Avinash, Gopal; Li, Baojun

2007-03-01

In digital tomosynthesis, one of the limitations is the presence of out-of-plane blur due to the limited angle acquisition. The point spread function (PSF) characterizes blur in the imaging volume, and is shift-variant in tomosynthesis. The purpose of this research is to classify the tomosynthesis imaging volume into four different categories based on PSF-driven focus criteria. We considered linear tomosynthesis geometry and simple back projection algorithm for reconstruction. The three-dimensional PSF at every pixel in the imaging volume was determined. Intensity profiles were computed for every pixel by integrating the PSF-weighted intensities contained within the line segment defined by the PSF, at each slice. Classification rules based on these intensity profiles were used to categorize image regions. At background and low-frequency pixels, the derived intensity profiles were flat curves with relatively low and high maximum intensities respectively. At in-focus pixels, the maximum intensity of the profiles coincided with the PSF-weighted intensity of the pixel. At out-of-focus pixels, the PSF-weighted intensity of the pixel was always less than the maximum intensity of the profile. We validated our method using human observer classified regions as gold standard. Based on the computed and manual classifications, the mean sensitivity and specificity of the algorithm were 77+/-8.44% and 91+/-4.13% respectively (t=-0.64, p=0.56, DF=4). Such a classification algorithm may assist in mitigating out-of-focus blur from tomosynthesis image slices.

The effect of lossy image compression on image classification

NASA Technical Reports Server (NTRS)

Paola, Justin D.; Schowengerdt, Robert A.

1995-01-01

We have classified four different images, under various levels of JPEG compression, using the following classification algorithms: minimum-distance, maximum-likelihood, and neural network. The training site accuracy and percent difference from the original classification were tabulated for each image compression level, with maximum-likelihood showing the poorest results. In general, as compression ratio increased, the classification retained its overall appearance, but much of the pixel-to-pixel detail was eliminated. We also examined the effect of compression on spatial pattern detection using a neural network.

Combination of support vector machine, artificial neural network and random forest for improving the classification of convective and stratiform rain using spectral features of SEVIRI data

NASA Astrophysics Data System (ADS)

Lazri, Mourad; Ameur, Soltane

2018-05-01

A model combining three classifiers, namely Support vector machine, Artificial neural network and Random forest (SAR) is designed for improving the classification of convective and stratiform rain. This model (SAR model) has been trained and then tested on a datasets derived from MSG-SEVIRI (Meteosat Second Generation-Spinning Enhanced Visible and Infrared Imager). Well-classified, mid-classified and misclassified pixels are determined from the combination of three classifiers. Mid-classified and misclassified pixels that are considered unreliable pixels are reclassified by using a novel training of the developed scheme. In this novel training, only the input data corresponding to the pixels in question to are used. This whole process is repeated a second time and applied to mid-classified and misclassified pixels separately. Learning and validation of the developed scheme are realized against co-located data observed by ground radar. The developed scheme outperformed different classifiers used separately and reached 97.40% of overall accuracy of classification.

Ground truth management system to support multispectral scanner /MSS/ digital analysis

NASA Technical Reports Server (NTRS)

Coiner, J. C.; Ungar, S. G.

1977-01-01

A computerized geographic information system for management of ground truth has been designed and implemented to relate MSS classification results to in situ observations. The ground truth system transforms, generalizes and rectifies ground observations to conform to the pixel size and shape of high resolution MSS aircraft data. These observations can then be aggregated for comparison to lower resolution sensor data. Construction of a digital ground truth array allows direct pixel by pixel comparison between classification results of MSS data and ground truth. By making comparisons, analysts can identify spatial distribution of error within the MSS data as well as usual figures of merit for the classifications. Use of the ground truth system permits investigators to compare a variety of environmental or anthropogenic data, such as soil color or tillage patterns, with classification results and allows direct inclusion of such data into classification operations. To illustrate the system, examples from classification of simulated Thematic Mapper data for agricultural test sites in North Dakota and Kansas are provided.

Detection of Olea europaea subsp. cuspidata and Juniperus procera in the dry Afromontane forest of northern Ethiopia using subpixel analysis of Landsat imagery

NASA Astrophysics Data System (ADS)

Hishe, Hadgu; Giday, Kidane; Neka, Mulugeta; Soromessa, Teshome; Van Orshoven, Jos; Muys, Bart

2015-01-01

Comprehensive and less costly forest inventory approaches are required to monitor the spatiotemporal dynamics of key species in forest ecosystems. Subpixel analysis using the earth resources data analysis system imagine subpixel classification procedure was tested to extract Olea europaea subsp. cuspidata and Juniperus procera canopies from Landsat 7 enhanced thematic mapper plus imagery. Control points with various canopy area fractions of the target species were collected to develop signatures for each of the species. With these signatures, the imagine subpixel classification procedure was run for each species independently. The subpixel process enabled the detection of O. europaea subsp. cuspidata and J. procera trees in pure and mixed pixels. Total of 100 pixels each were field verified for both species. An overall accuracy of 85% was achieved for O. europaea subsp. cuspidata and 89% for J. procera. A high overall accuracy level of detecting species at a natural forest was achieved, which encourages using the algorithm for future species monitoring activities. We recommend that the algorithm has to be validated in similar environment to enrich the knowledge on its capability to ensure its wider usage.

Enhancing spatial resolution of (18)F positron imaging with the Timepix detector by classification of primary fired pixels using support vector machine.

PubMed

Wang, Qian; Liu, Zhen; Ziegler, Sibylle I; Shi, Kuangyu

2015-07-07

Position-sensitive positron cameras using silicon pixel detectors have been applied for some preclinical and intraoperative clinical applications. However, the spatial resolution of a positron camera is limited by positron multiple scattering in the detector. An incident positron may fire a number of successive pixels on the imaging plane. It is still impossible to capture the primary fired pixel along a particle trajectory by hardware or to perceive the pixel firing sequence by direct observation. Here, we propose a novel data-driven method to improve the spatial resolution by classifying the primary pixels within the detector using support vector machine. A classification model is constructed by learning the features of positron trajectories based on Monte-Carlo simulations using Geant4. Topological and energy features of pixels fired by (18)F positrons were considered for the training and classification. After applying the classification model on measurements, the primary fired pixels of the positron tracks in the silicon detector were estimated. The method was tested and assessed for [(18)F]FDG imaging of an absorbing edge protocol and a leaf sample. The proposed method improved the spatial resolution from 154.6   ±   4.2 µm (energy weighted centroid approximation) to 132.3   ±   3.5 µm in the absorbing edge measurements. For the positron imaging of a leaf sample, the proposed method achieved lower root mean square error relative to phosphor plate imaging, and higher similarity with the reference optical image. The improvements of the preliminary results support further investigation of the proposed algorithm for the enhancement of positron imaging in clinical and preclinical applications.

Enhancing spatial resolution of 18F positron imaging with the Timepix detector by classification of primary fired pixels using support vector machine

NASA Astrophysics Data System (ADS)

Wang, Qian; Liu, Zhen; Ziegler, Sibylle I.; Shi, Kuangyu

2015-07-01

Position-sensitive positron cameras using silicon pixel detectors have been applied for some preclinical and intraoperative clinical applications. However, the spatial resolution of a positron camera is limited by positron multiple scattering in the detector. An incident positron may fire a number of successive pixels on the imaging plane. It is still impossible to capture the primary fired pixel along a particle trajectory by hardware or to perceive the pixel firing sequence by direct observation. Here, we propose a novel data-driven method to improve the spatial resolution by classifying the primary pixels within the detector using support vector machine. A classification model is constructed by learning the features of positron trajectories based on Monte-Carlo simulations using Geant4. Topological and energy features of pixels fired by 18F positrons were considered for the training and classification. After applying the classification model on measurements, the primary fired pixels of the positron tracks in the silicon detector were estimated. The method was tested and assessed for [18F]FDG imaging of an absorbing edge protocol and a leaf sample. The proposed method improved the spatial resolution from 154.6   ±   4.2 µm (energy weighted centroid approximation) to 132.3   ±   3.5 µm in the absorbing edge measurements. For the positron imaging of a leaf sample, the proposed method achieved lower root mean square error relative to phosphor plate imaging, and higher similarity with the reference optical image. The improvements of the preliminary results support further investigation of the proposed algorithm for the enhancement of positron imaging in clinical and preclinical applications.

A Method of Spatial Mapping and Reclassification for High-Spatial-Resolution Remote Sensing Image Classification

PubMed Central

Wang, Guizhou; Liu, Jianbo; He, Guojin

2013-01-01

This paper presents a new classification method for high-spatial-resolution remote sensing images based on a strategic mechanism of spatial mapping and reclassification. The proposed method includes four steps. First, the multispectral image is classified by a traditional pixel-based classification method (support vector machine). Second, the panchromatic image is subdivided by watershed segmentation. Third, the pixel-based multispectral image classification result is mapped to the panchromatic segmentation result based on a spatial mapping mechanism and the area dominant principle. During the mapping process, an area proportion threshold is set, and the regional property is defined as unclassified if the maximum area proportion does not surpass the threshold. Finally, unclassified regions are reclassified based on spectral information using the minimum distance to mean algorithm. Experimental results show that the classification method for high-spatial-resolution remote sensing images based on the spatial mapping mechanism and reclassification strategy can make use of both panchromatic and multispectral information, integrate the pixel- and object-based classification methods, and improve classification accuracy. PMID:24453808

Mapping and monitoring changes in vegetation communities of Jasper Ridge, CA, using spectral fractions derived from AVIRIS images

NASA Technical Reports Server (NTRS)

Sabol, Donald E., Jr.; Roberts, Dar A.; Adams, John B.; Smith, Milton O.

1993-01-01

An important application of remote sensing is to map and monitor changes over large areas of the land surface. This is particularly significant with the current interest in monitoring vegetation communities. Most of traditional methods for mapping different types of plant communities are based upon statistical classification techniques (i.e., parallel piped, nearest-neighbor, etc.) applied to uncalibrated multispectral data. Classes from these techniques are typically difficult to interpret (particularly to a field ecologist/botanist). Also, classes derived for one image can be very different from those derived from another image of the same area, making interpretation of observed temporal changes nearly impossible. More recently, neural networks have been applied to classification. Neural network classification, based upon spectral matching, is weak in dealing with spectral mixtures (a condition prevalent in images of natural surfaces). Another approach to mapping vegetation communities is based on spectral mixture analysis, which can provide a consistent framework for image interpretation. Roberts et al. (1990) mapped vegetation using the band residuals from a simple mixing model (the same spectral endmembers applied to all image pixels). Sabol et al. (1992b) and Roberts et al. (1992) used different methods to apply the most appropriate spectral endmembers to each image pixel, thereby allowing mapping of vegetation based upon the the different endmember spectra. In this paper, we describe a new approach to classification of vegetation communities based upon the spectra fractions derived from spectral mixture analysis. This approach was applied to three 1992 AVIRIS images of Jasper Ridge, California to observe seasonal changes in surface composition.

Integrated pillar scatterers for speeding up classification of cell holograms.

PubMed

Lugnan, Alessio; Dambre, Joni; Bienstman, Peter

2017-11-27

The computational power required to classify cell holograms is a major limit to the throughput of label-free cell sorting based on digital holographic microscopy. In this work, a simple integrated photonic stage comprising a collection of silica pillar scatterers is proposed as an effective nonlinear mixing interface between the light scattered by a cell and an image sensor. The light processing provided by the photonic stage allows for the use of a simple linear classifier implemented in the electric domain and applied on a limited number of pixels. A proof-of-concept of the presented machine learning technique, which is based on the extreme learning machine (ELM) paradigm, is provided by the classification results on samples generated by 2D FDTD simulations of cells in a microfluidic channel.

A comparison of the accuracy of pixel based and object based classifications of integrated optical and LiDAR data

NASA Astrophysics Data System (ADS)

Gajda, Agnieszka; Wójtowicz-Nowakowska, Anna

2013-04-01

A comparison of the accuracy of pixel based and object based classifications of integrated optical and LiDAR data Land cover maps are generally produced on the basis of high resolution imagery. Recently, LiDAR (Light Detection and Ranging) data have been brought into use in diverse applications including land cover mapping. In this study we attempted to assess the accuracy of land cover classification using both high resolution aerial imagery and LiDAR data (airborne laser scanning, ALS), testing two classification approaches: a pixel-based classification and object-oriented image analysis (OBIA). The study was conducted on three test areas (3 km2 each) in the administrative area of Kraków, Poland, along the course of the Vistula River. They represent three different dominating land cover types of the Vistula River valley. Test site 1 had a semi-natural vegetation, with riparian forests and shrubs, test site 2 represented a densely built-up area, and test site 3 was an industrial site. Point clouds from ALS and ortophotomaps were both captured in November 2007. Point cloud density was on average 16 pt/m2 and it contained additional information about intensity and encoded RGB values. Ortophotomaps had a spatial resolution of 10 cm. From point clouds two raster maps were generated: intensity (1) and (2) normalised Digital Surface Model (nDSM), both with the spatial resolution of 50 cm. To classify the aerial data, a supervised classification approach was selected. Pixel based classification was carried out in ERDAS Imagine software. Ortophotomaps and intensity and nDSM rasters were used in classification. 15 homogenous training areas representing each cover class were chosen. Classified pixels were clumped to avoid salt and pepper effect. Object oriented image object classification was carried out in eCognition software, which implements both the optical and ALS data. Elevation layers (intensity, firs/last reflection, etc.) were used at segmentation stage due to proper wages usage. Thus a more precise and unambiguous boundaries of segments (objects) were received. As a results of the classification 5 classes of land cover (buildings, water, high and low vegetation and others) were extracted. Both pixel-based image analysis and OBIA were conducted with a minimum mapping unit of 10m2. Results were validated on the basis on manual classification and random points (80 per test area), reference data set was manually interpreted using ortophotomaps and expert knowledge of the test site areas.

Mapping of land cover in northern California with simulated hyperspectral satellite imagery

NASA Astrophysics Data System (ADS)

Clark, Matthew L.; Kilham, Nina E.

2016-09-01

Land-cover maps are important science products needed for natural resource and ecosystem service management, biodiversity conservation planning, and assessing human-induced and natural drivers of land change. Analysis of hyperspectral, or imaging spectrometer, imagery has shown an impressive capacity to map a wide range of natural and anthropogenic land cover. Applications have been mostly with single-date imagery from relatively small spatial extents. Future hyperspectral satellites will provide imagery at greater spatial and temporal scales, and there is a need to assess techniques for mapping land cover with these data. Here we used simulated multi-temporal HyspIRI satellite imagery over a 30,000 km2 area in the San Francisco Bay Area, California to assess its capabilities for mapping classes defined by the international Land Cover Classification System (LCCS). We employed a mapping methodology and analysis framework that is applicable to regional and global scales. We used the Random Forests classifier with three sets of predictor variables (reflectance, MNF, hyperspectral metrics), two temporal resolutions (summer, spring-summer-fall), two sample scales (pixel, polygon) and two levels of classification complexity (12, 20 classes). Hyperspectral metrics provided a 16.4-21.8% and 3.1-6.7% increase in overall accuracy relative to MNF and reflectance bands, respectively, depending on pixel or polygon scales of analysis. Multi-temporal metrics improved overall accuracy by 0.9-3.1% over summer metrics, yet increases were only significant at the pixel scale of analysis. Overall accuracy at pixel scales was 72.2% (Kappa 0.70) with three seasons of metrics. Anthropogenic and homogenous natural vegetation classes had relatively high confidence and producer and user accuracies were over 70%; in comparison, woodland and forest classes had considerable confusion. We next focused on plant functional types with relatively pure spectra by removing open-canopy shrublands, woodlands and mixed forests from the classification. This 12-class map had significantly improved accuracy of 85.1% (Kappa 0.83) and most classes had over 70% producer and user accuracies. Finally, we summarized important metrics from the multi-temporal Random Forests to infer the underlying chemical and structural properties that best discriminated our land-cover classes across seasons.

The fragmented nature of tundra landscape

NASA Astrophysics Data System (ADS)

Virtanen, Tarmo; Ek, Malin

2014-04-01

The vegetation and land cover structure of tundra areas is fragmented when compared to other biomes. Thus, satellite images of high resolution are required for producing land cover classifications, in order to reveal the actual distribution of land cover types across these large and remote areas. We produced and compared different land cover classifications using three satellite images (QuickBird, Aster and Landsat TM5) with different pixel sizes (2.4 m, 15 m and 30 m pixel size, respectively). The study area, in north-eastern European Russia, was visited in July 2007 to obtain ground reference data. The QuickBird image was classified using supervised segmentation techniques, while the Aster and Landsat TM5 images were classified using a pixel-based supervised classification method. The QuickBird classification showed the highest accuracy when tested against field data, while the Aster image was generally more problematic to classify than the Landsat TM5 image. Use of smaller pixel sized images distinguished much greater levels of landscape fragmentation. The overall mean patch sizes in the QuickBird, Aster, and Landsat TM5-classifications were 871 m2, 2141 m2 and 7433 m2, respectively. In the QuickBird classification, the mean patch size of all the tundra and peatland vegetation classes was smaller than one pixel of the Landsat TM5 image. Water bodies and fens in particular occur in the landscape in small or elongated patches, and thus cannot be realistically classified from larger pixel sized images. Land cover patterns vary considerably at such a fine-scale, so that a lot of information is lost if only medium resolution satellite images are used. It is crucial to know the amount and spatial distribution of different vegetation types in arctic landscapes, as carbon dynamics and other climate related physical, geological and biological processes are known to vary greatly between vegetation types.

Quantifying tree mortality in a mixed species woodland using multitemporal high spatial resolution satellite imagery

USGS Publications Warehouse

Garrity, Steven R.; Allen, Craig D.; Brumby, Steven P.; Gangodagamage, Chandana; McDowell, Nate G.; Cai, D. Michael

2013-01-01

Widespread tree mortality events have recently been observed in several biomes. To effectively quantify the severity and extent of these events, tools that allow for rapid assessment at the landscape scale are required. Past studies using high spatial resolution satellite imagery have primarily focused on detecting green, red, and gray tree canopies during and shortly after tree damage or mortality has occurred. However, detecting trees in various stages of death is not always possible due to limited availability of archived satellite imagery. Here we assess the capability of high spatial resolution satellite imagery for tree mortality detection in a southwestern U.S. mixed species woodland using archived satellite images acquired prior to mortality and well after dead trees had dropped their leaves. We developed a multistep classification approach that uses: supervised masking of non-tree image elements; bi-temporal (pre- and post-mortality) differencing of normalized difference vegetation index (NDVI) and red:green ratio (RGI); and unsupervised multivariate clustering of pixels into live and dead tree classes using a Gaussian mixture model. Classification accuracies were improved in a final step by tuning the rules of pixel classification using the posterior probabilities of class membership obtained from the Gaussian mixture model. Classifications were produced for two images acquired post-mortality with overall accuracies of 97.9% and 98.5%, respectively. Classified images were combined with land cover data to characterize the spatiotemporal characteristics of tree mortality across areas with differences in tree species composition. We found that 38% of tree crown area was lost during the drought period between 2002 and 2006. The majority of tree mortality during this period was concentrated in piñon-juniper (Pinus edulis-Juniperus monosperma) woodlands. An additional 20% of the tree canopy died or was removed between 2006 and 2011, primarily in areas experiencing wildfire and management activity. -Our results demonstrate that unsupervised clustering of bi-temporal NDVI and RGI differences can be used to detect tree mortality resulting from numerous causes and in several forest cover types.

Land Cover Classification in a Complex Urban-Rural Landscape with Quickbird Imagery

PubMed Central

Moran, Emilio Federico.

2010-01-01

High spatial resolution images have been increasingly used for urban land use/cover classification, but the high spectral variation within the same land cover, the spectral confusion among different land covers, and the shadow problem often lead to poor classification performance based on the traditional per-pixel spectral-based classification methods. This paper explores approaches to improve urban land cover classification with Quickbird imagery. Traditional per-pixel spectral-based supervised classification, incorporation of textural images and multispectral images, spectral-spatial classifier, and segmentation-based classification are examined in a relatively new developing urban landscape, Lucas do Rio Verde in Mato Grosso State, Brazil. This research shows that use of spatial information during the image classification procedure, either through the integrated use of textural and spectral images or through the use of segmentation-based classification method, can significantly improve land cover classification performance. PMID:21643433

Impervious surface mapping with Quickbird imagery

PubMed Central

Lu, Dengsheng; Hetrick, Scott; Moran, Emilio

2010-01-01

This research selects two study areas with different urban developments, sizes, and spatial patterns to explore the suitable methods for mapping impervious surface distribution using Quickbird imagery. The selected methods include per-pixel based supervised classification, segmentation-based classification, and a hybrid method. A comparative analysis of the results indicates that per-pixel based supervised classification produces a large number of “salt-and-pepper” pixels, and segmentation based methods can significantly reduce this problem. However, neither method can effectively solve the spectral confusion of impervious surfaces with water/wetland and bare soils and the impacts of shadows. In order to accurately map impervious surface distribution from Quickbird images, manual editing is necessary and may be the only way to extract impervious surfaces from the confused land covers and the shadow problem. This research indicates that the hybrid method consisting of thresholding techniques, unsupervised classification and limited manual editing provides the best performance. PMID:21643434

Contribution of non-negative matrix factorization to the classification of remote sensing images

NASA Astrophysics Data System (ADS)

Karoui, M. S.; Deville, Y.; Hosseini, S.; Ouamri, A.; Ducrot, D.

2008-10-01

Remote sensing has become an unavoidable tool for better managing our environment, generally by realizing maps of land cover using classification techniques. The classification process requires some pre-processing, especially for data size reduction. The most usual technique is Principal Component Analysis. Another approach consists in regarding each pixel of the multispectral image as a mixture of pure elements contained in the observed area. Using Blind Source Separation (BSS) methods, one can hope to unmix each pixel and to perform the recognition of the classes constituting the observed scene. Our contribution consists in using Non-negative Matrix Factorization (NMF) combined with sparse coding as a solution to BSS, in order to generate new images (which are at least partly separated images) using HRV SPOT images from Oran area, Algeria). These images are then used as inputs of a supervised classifier integrating textural information. The results of classifications of these "separated" images show a clear improvement (correct pixel classification rate improved by more than 20%) compared to classification of initial (i.e. non separated) images. These results show the contribution of NMF as an attractive pre-processing for classification of multispectral remote sensing imagery.

Rule-driven defect detection in CT images of hardwood logs

Treesearch

Erol Sarigul; A. Lynn Abbott; Daniel L. Schmoldt

2000-01-01

This paper deals with automated detection and identification of internal defects in hardwood logs using computed tomography (CT) images. We have developed a system that employs artificial neural networks to perform tentative classification of logs on a pixel-by-pixel basis. This approach achieves a high level of classification accuracy for several hardwood species (...

Mapping ecological states in a complex environment

NASA Astrophysics Data System (ADS)

Steele, C. M.; Bestelmeyer, B.; Burkett, L. M.; Ayers, E.; Romig, K.; Slaughter, A.

2013-12-01

The vegetation of northern Chihuahuan Desert rangelands is sparse, heterogeneous and for most of the year, consists of a large proportion of non-photosynthetic material. The soils in this area are spectrally bright and variable in their reflectance properties. Both factors provide challenges to the application of remote sensing for estimating canopy variables (e.g., leaf area index, biomass, percentage canopy cover, primary production). Additionally, with reference to current paradigms of rangeland health assessment, remotely-sensed estimates of canopy variables have limited practical use to the rangeland manager if they are not placed in the context of ecological site and ecological state. To address these challenges, we created a multifactor classification system based on the USDA-NRCS ecological site schema and associated state-and-transition models to map ecological states on desert rangelands in southern New Mexico. Applying this system using per-pixel image processing techniques and multispectral, remotely sensed imagery raised other challenges. Per-pixel image classification relies upon the spectral information in each pixel alone, there is no reference to the spatial context of the pixel and its relationship with its neighbors. Ecological state classes may have direct relevance to managers but the non-unique spectral properties of different ecological state classes in our study area means that per-pixel classification of multispectral data performs poorly in discriminating between different ecological states. We found that image interpreters who are familiar with the landscape and its associated ecological site descriptions perform better than per-pixel classification techniques in assigning ecological states. However, two important issues affect manual classification methods: subjectivity of interpretation and reproducibility of results. An alternative to per-pixel classification and manual interpretation is object-based image analysis. Object-based image analysis provides a platform for classification that more closely resembles human recognition of objects within a remotely sensed image. The analysis presented here compares multiple thematic maps created for test locations on the USDA-ARS Jornada Experimental Range ranch. Three study sites in different pastures, each 300 ha in size, were selected for comparison on the basis of their ecological site type (';Clayey', ';Sandy' and a combination of both) and the degree of complexity of vegetation cover. Thematic maps were produced for each study site using (i) manual interpretation of digital aerial photography (by five independent interpreters); (ii) object-oriented, decision-tree classification of fine and moderate spatial resolution imagery (Quickbird; Landsat Thematic Mapper) and (iii) ground survey. To identify areas of uncertainty, we compared agreement in location, areal extent and class assignation between 5 independently produced, manually-digitized ecological state maps and with the map created from ground survey. Location, areal extent and class assignation of the map produced by object-oriented classification was also assessed with reference to the ground survey map.

Object-based land cover classification based on fusion of multifrequency SAR data and THAICHOTE optical imagery

NASA Astrophysics Data System (ADS)

Sukawattanavijit, Chanika; Srestasathiern, Panu

2017-10-01

Land Use and Land Cover (LULC) information are significant to observe and evaluate environmental change. LULC classification applying remotely sensed data is a technique popularly employed on a global and local dimension particularly, in urban areas which have diverse land cover types. These are essential components of the urban terrain and ecosystem. In the present, object-based image analysis (OBIA) is becoming widely popular for land cover classification using the high-resolution image. COSMO-SkyMed SAR data was fused with THAICHOTE (namely, THEOS: Thailand Earth Observation Satellite) optical data for land cover classification using object-based. This paper indicates a comparison between object-based and pixel-based approaches in image fusion. The per-pixel method, support vector machines (SVM) was implemented to the fused image based on Principal Component Analysis (PCA). For the objectbased classification was applied to the fused images to separate land cover classes by using nearest neighbor (NN) classifier. Finally, the accuracy assessment was employed by comparing with the classification of land cover mapping generated from fused image dataset and THAICHOTE image. The object-based data fused COSMO-SkyMed with THAICHOTE images demonstrated the best classification accuracies, well over 85%. As the results, an object-based data fusion provides higher land cover classification accuracy than per-pixel data fusion.

A mutual information-Dempster-Shafer based decision ensemble system for land cover classification of hyperspectral data

NASA Astrophysics Data System (ADS)

Pahlavani, Parham; Bigdeli, Behnaz

2017-12-01

Hyperspectral images contain extremely rich spectral information that offer great potential to discriminate between various land cover classes. However, these images are usually composed of tens or hundreds of spectrally close bands, which result in high redundancy and great amount of computation time in hyperspectral classification. Furthermore, in the presence of mixed coverage pixels, crisp classifiers produced errors, omission and commission. This paper presents a mutual information-Dempster-Shafer system through an ensemble classification approach for classification of hyperspectral data. First, mutual information is applied to split data into a few independent partitions to overcome high dimensionality. Then, a fuzzy maximum likelihood classifies each band subset. Finally, Dempster-Shafer is applied to fuse the results of the fuzzy classifiers. In order to assess the proposed method, a crisp ensemble system based on a support vector machine as the crisp classifier and weighted majority voting as the crisp fusion method are applied on hyperspectral data. Furthermore, a dimension reduction system is utilized to assess the effectiveness of mutual information band splitting of the proposed method. The proposed methodology provides interesting conclusions on the effectiveness and potentiality of mutual information-Dempster-Shafer based classification of hyperspectral data.

Efficiency of the spectral-spatial classification of hyperspectral imaging data

NASA Astrophysics Data System (ADS)

Borzov, S. M.; Potaturkin, O. I.

2017-01-01

The efficiency of methods of the spectral-spatial classification of similarly looking types of vegetation on the basis of hyperspectral data of remote sensing of the Earth, which take into account local neighborhoods of analyzed image pixels, is experimentally studied. Algorithms that involve spatial pre-processing of the raw data and post-processing of pixel-based spectral classification maps are considered. Results obtained both for a large-size hyperspectral image and for its test fragment with different methods of training set construction are reported. The classification accuracy in all cases is estimated through comparisons of ground-truth data and classification maps formed by using the compared methods. The reasons for the differences in these estimates are discussed.

Testing random forest classification for identifying lava flows and mapping age groups on a single Landsat 8 image

NASA Astrophysics Data System (ADS)

Li, Long; Solana, Carmen; Canters, Frank; Kervyn, Matthieu

2017-10-01

Mapping lava flows using satellite images is an important application of remote sensing in volcanology. Several volcanoes have been mapped through remote sensing using a wide range of data, from optical to thermal infrared and radar images, using techniques such as manual mapping, supervised/unsupervised classification, and elevation subtraction. So far, spectral-based mapping applications mainly focus on the use of traditional pixel-based classifiers, without much investigation into the added value of object-based approaches and into advantages of using machine learning algorithms. In this study, Nyamuragira, characterized by a series of > 20 overlapping lava flows erupted over the last century, was used as a case study. The random forest classifier was tested to map lava flows based on pixels and objects. Image classification was conducted for the 20 individual flows and for 8 groups of flows of similar age using a Landsat 8 image and a DEM of the volcano, both at 30-meter spatial resolution. Results show that object-based classification produces maps with continuous and homogeneous lava surfaces, in agreement with the physical characteristics of lava flows, while lava flows mapped through the pixel-based classification are heterogeneous and fragmented including much "salt and pepper noise". In terms of accuracy, both pixel-based and object-based classification performs well but the former results in higher accuracies than the latter except for mapping lava flow age groups without using topographic features. It is concluded that despite spectral similarity, lava flows of contrasting age can be well discriminated and mapped by means of image classification. The classification approach demonstrated in this study only requires easily accessible image data and can be applied to other volcanoes as well if there is sufficient information to calibrate the mapping.

Hyperspectral Image Classification via Kernel Sparse Representation

DTIC Science & Technology

2013-01-01

classification algorithms. Moreover, the spatial coherency across neighboring pixels is also incorporated through a kernelized joint sparsity model , where...joint sparsity model , where all of the pixels within a small neighborhood are jointly represented in the feature space by selecting a few common training...hyperspectral imagery, joint spar- sity model , kernel methods, sparse representation. I. INTRODUCTION HYPERSPECTRAL imaging sensors capture images

Application of LANDSAT system for improving methodology for inventory and classification of wetlands

NASA Technical Reports Server (NTRS)

Gilmer, D. S. (Principal Investigator)

1976-01-01

The author has identified the following significant results. A newly developed software system for generating statistics on surface water features was tested using LANDSAT data acquired previous to 1975. This software test provided a satisfactory evaluation of the system and also allowed expansion of data base on prairie water features. The software system recognizes water on the basis of a classification algorithm. This classification is accomplished by level thresholding a single near infrared data channel. After each pixel is classified as water or nonwater, the software system then recognizes ponds or lakes as sets of contiguous pixels or as single isolated pixels in the case of very small ponds. Pixels are considered to be contiguous if they are adjacent between successive scan lines. After delineating each water feature, the software system then assigns the feature a position based upon a geographic grid system and calculates the feature's planimetric area, its perimeter, and a parameter known as the shape factor.

Comparison of Sub-Pixel Classification Approaches for Crop-Specific Mapping

EPA Science Inventory

This paper examined two non-linear models, Multilayer Perceptron (MLP) regression and Regression Tree (RT), for estimating sub-pixel crop proportions using time-series MODIS-NDVI data. The sub-pixel proportions were estimated for three major crop types including corn, soybean, a...

Effects of autocorrelation upon LANDSAT classification accuracy. [Richmond, Virginia and Denver, Colorado

NASA Technical Reports Server (NTRS)

Craig, R. G. (Principal Investigator)

1983-01-01

Richmond, Virginia and Denver, Colorado were study sites in an effort to determine the effect of autocorrelation on the accuracy of a parallelopiped classifier of LANDSAT digital data. The autocorrelation was assumed to decay to insignificant levels when sampled at distances of at least ten pixels. Spectral themes developed using blocks of adjacent pixels, and using groups of pixels spaced at least 10 pixels apart were used. Effects of geometric distortions were minimized by using only pixels from the interiors of land cover sections. Accuracy was evaluated for three classes; agriculture, residential and "all other"; both type 1 and type 2 errors were evaluated by means of overall classification accuracy. All classes give comparable results. Accuracy is approximately the same in both techniques; however, the variance in accuracy is significantly higher using the themes developed from autocorrelated data. The vectors of mean spectral response were nearly identical regardless of sampling method used. The estimated variances were much larger when using autocorrelated pixels.

Combinational pixel-by-pixel and object-level classifying, segmenting, and agglomerating in performing quantitative image analysis that distinguishes between healthy non-cancerous and cancerous cell nuclei and delineates nuclear, cytoplasm, and stromal material objects from stained biological tissue materials

DOEpatents

Boucheron, Laura E

2013-07-16

Quantitative object and spatial arrangement-level analysis of tissue are detailed using expert (pathologist) input to guide the classification process. A two-step method is disclosed for imaging tissue, by classifying one or more biological materials, e.g. nuclei, cytoplasm, and stroma, in the tissue into one or more identified classes on a pixel-by-pixel basis, and segmenting the identified classes to agglomerate one or more sets of identified pixels into segmented regions. Typically, the one or more biological materials comprises nuclear material, cytoplasm material, and stromal material. The method further allows a user to markup the image subsequent to the classification to re-classify said materials. The markup is performed via a graphic user interface to edit designated regions in the image.

Spectral-Spatial Classification of Hyperspectral Images Using Hierarchical Optimization

NASA Technical Reports Server (NTRS)

Tarabalka, Yuliya; Tilton, James C.

2011-01-01

A new spectral-spatial method for hyperspectral data classification is proposed. For a given hyperspectral image, probabilistic pixelwise classification is first applied. Then, hierarchical step-wise optimization algorithm is performed, by iteratively merging neighboring regions with the smallest Dissimilarity Criterion (DC) and recomputing class labels for new regions. The DC is computed by comparing region mean vectors, class labels and a number of pixels in the two regions under consideration. The algorithm is converged when all the pixels get involved in the region merging procedure. Experimental results are presented on two remote sensing hyperspectral images acquired by the AVIRIS and ROSIS sensors. The proposed approach improves classification accuracies and provides maps with more homogeneous regions, when compared to previously proposed classification techniques.

Hyperspectral imaging for detection of non-O157 Shiga-toxin producing Escherichia coli (STEC) serogroups on spread plates of mixed cultures

NASA Astrophysics Data System (ADS)

Yoon, Seung Chul; Windham, William R.; Ladely, Scott; Heitschmidt, Gerald W.; Lawrence, Kurt C.; Park, Bosoon; Narang, Neelam; Cray, William C.

2012-05-01

We investigated the feasibility of visible and near-infrared (VNIR) hyperspectral imaging for rapid presumptive-positive screening of six representative non-O157 Shiga-toxin producing Escherichia coli (STEC) serogroups (O26, O45, O103, O111, O121, and O145) on spread plates of mixed cultures. Although the traditional culture method is still the "gold standard" for presumptive-positive pathogen screening, it is time-consuming, labor-intensive, not effective in testing large amount of food samples, and cannot completely prevent unwanted background microflora from growing together with target microorganisms on agar media. A previous study was performed using the data obtained from pure cultures individually inoculated on spot and/or spread plates in order to develop multivariate classification models differentiating each colony of the six non-O157 STEC serogroups and to optimize the models in terms of parameters. This study dealt with the validation of the trained and optimized models with a test set of new independent samples obtained from colonies on spread plates of mixed cultures. A new validation protocol appropriate to a hyperspectral imaging study for mixed cultures was developed. One imaging experiment with colonies obtained from two serial dilutions was performed. A total of six agar plates were prepared, where O45, O111 and O121 serogroups were inoculated into all six plates and each of O45, O103 and O145 serogroups was added into the mixture of the three common bacterial cultures. The number of colonies grown after 24-h incubation was 331 and the number of pixels associated with the grown colonies was 16,379. The best model found from this validation study was based on pre-processing with standard normal variate and detrending (SNVD), first derivative, spectral smoothing, and k-nearest neighbor classification (kNN, k=3) of scores in the principal component subspace spanned by 6 principal components. The independent testing results showed 95% overall detection accuracy at pixel level and 97% at colony level. The developed model was proven to be still valid even for the independent samples although the size of a test set was small and only one experiment was performed. This study was an important first step in validating and updating multivariate classification models for rapid screening of ground beef samples contaminated by non-O157 STEC pathogens using hyperspectral imaging.

Automated artery-venous classification of retinal blood vessels based on structural mapping method

NASA Astrophysics Data System (ADS)

Joshi, Vinayak S.; Garvin, Mona K.; Reinhardt, Joseph M.; Abramoff, Michael D.

2012-03-01

Retinal blood vessels show morphologic modifications in response to various retinopathies. However, the specific responses exhibited by arteries and veins may provide a precise diagnostic information, i.e., a diabetic retinopathy may be detected more accurately with the venous dilatation instead of average vessel dilatation. In order to analyze the vessel type specific morphologic modifications, the classification of a vessel network into arteries and veins is required. We previously described a method for identification and separation of retinal vessel trees; i.e. structural mapping. Therefore, we propose the artery-venous classification based on structural mapping and identification of color properties prominent to the vessel types. The mean and standard deviation of each of green channel intensity and hue channel intensity are analyzed in a region of interest around each centerline pixel of a vessel. Using the vector of color properties extracted from each centerline pixel, it is classified into one of the two clusters (artery and vein), obtained by the fuzzy-C-means clustering. According to the proportion of clustered centerline pixels in a particular vessel, and utilizing the artery-venous crossing property of retinal vessels, each vessel is assigned a label of an artery or a vein. The classification results are compared with the manually annotated ground truth (gold standard). We applied the proposed method to a dataset of 15 retinal color fundus images resulting in an accuracy of 88.28% correctly classified vessel pixels. The automated classification results match well with the gold standard suggesting its potential in artery-venous classification and the respective morphology analysis.

A fast and efficient segmentation scheme for cell microscopic image.

PubMed

Lebrun, G; Charrier, C; Lezoray, O; Meurie, C; Cardot, H

2007-04-27

Microscopic cellular image segmentation schemes must be efficient for reliable analysis and fast to process huge quantity of images. Recent studies have focused on improving segmentation quality. Several segmentation schemes have good quality but processing time is too expensive to deal with a great number of images per day. For segmentation schemes based on pixel classification, the classifier design is crucial since it is the one which requires most of the processing time necessary to segment an image. The main contribution of this work is focused on how to reduce the complexity of decision functions produced by support vector machines (SVM) while preserving recognition rate. Vector quantization is used in order to reduce the inherent redundancy present in huge pixel databases (i.e. images with expert pixel segmentation). Hybrid color space design is also used in order to improve data set size reduction rate and recognition rate. A new decision function quality criterion is defined to select good trade-off between recognition rate and processing time of pixel decision function. The first results of this study show that fast and efficient pixel classification with SVM is possible. Moreover posterior class pixel probability estimation is easy to compute with Platt method. Then a new segmentation scheme using probabilistic pixel classification has been developed. This one has several free parameters and an automatic selection must dealt with, but criteria for evaluate segmentation quality are not well adapted for cell segmentation, especially when comparison with expert pixel segmentation must be achieved. Another important contribution in this paper is the definition of a new quality criterion for evaluation of cell segmentation. The results presented here show that the selection of free parameters of the segmentation scheme by optimisation of the new quality cell segmentation criterion produces efficient cell segmentation.

Shape classification of malignant lymphomas and leukemia by morphological watersheds and ARMA modeling

NASA Astrophysics Data System (ADS)

Celenk, Mehmet; Song, Yinglei; Ma, Limin; Zhou, Min

2003-05-01

A new algorithm that can be used to automatically recognize and classify malignant lymphomas and lukemia is proposed in this paper. The algorithm utilizes the morphological watershed to extract boundaries of cells from their grey-level images. It generates a sequence of Euclidean distances by selecting pixels in clockwise direction on the boundary of the cell and calculating the Euclidean distances of the selected pixels from the centroid of the cell. A feature vector associated with each cell is then obtained by applying the auto-regressive moving-average (ARMA) model to the generated sequence of Euclidean distances. The clustering measure J3=trace{inverse(Sw-1)Sm} involving the within (Sw) and mixed (Sm) class-scattering matrices is computed for both cell classes to provide an insight into the extent to which different cell classes in the training data are separated. Our test results suggest that the algorithm is highly accurate for the development of an interactive, computer-assisted diagnosis (CAD) tool.

Toward multidisciplinary use of LANDSAT: Interfacing computerized LANDSAT analysis systems with geographic information systems

NASA Technical Reports Server (NTRS)

Myers, W. L.

1981-01-01

The LANDSAT-geographic information system (GIS) interface must summarize the results of the LANDSAT classification over the same cells that serve as geographic referencing units for the GIS, and output these summaries on a cell-by-cell basis in a form that is readable by the input routines of the GIS. The ZONAL interface for cell-oriented systems consists of two primary programs. The PIXCEL program scans the grid of cells and outputs a channel of pixels. Each pixel contains not the reflectance values but the identifier of the cell in which the center of the pixel is located. This file of pixelized cells along with the results of a pixel-by-pixel classification of the scene produced by the LANDSAT analysis system are input to the CELSUM program which then outputs a cell-by-cell summary formatted according to the requirements of the host GIS. Cross-correlation of the LANDSAT layer with the other layers in the data base is accomplished with the analysis and display facilities of the GIS.

Advances in Spectral-Spatial Classification of Hyperspectral Images

NASA Technical Reports Server (NTRS)

Fauvel, Mathieu; Tarabalka, Yuliya; Benediktsson, Jon Atli; Chanussot, Jocelyn; Tilton, James C.

2012-01-01

Recent advances in spectral-spatial classification of hyperspectral images are presented in this paper. Several techniques are investigated for combining both spatial and spectral information. Spatial information is extracted at the object (set of pixels) level rather than at the conventional pixel level. Mathematical morphology is first used to derive the morphological profile of the image, which includes characteristics about the size, orientation and contrast of the spatial structures present in the image. Then the morphological neighborhood is defined and used to derive additional features for classification. Classification is performed with support vector machines using the available spectral information and the extracted spatial information. Spatial post-processing is next investigated to build more homogeneous and spatially consistent thematic maps. To that end, three presegmentation techniques are applied to define regions that are used to regularize the preliminary pixel-wise thematic map. Finally, a multiple classifier system is defined to produce relevant markers that are exploited to segment the hyperspectral image with the minimum spanning forest algorithm. Experimental results conducted on three real hyperspectral images with different spatial and spectral resolutions and corresponding to various contexts are presented. They highlight the importance of spectral-spatial strategies for the accurate classification of hyperspectral images and validate the proposed methods.

A coarse-to-fine approach for medical hyperspectral image classification with sparse representation

NASA Astrophysics Data System (ADS)

Chang, Lan; Zhang, Mengmeng; Li, Wei

2017-10-01

A coarse-to-fine approach with sparse representation is proposed for medical hyperspectral image classification in this work. Segmentation technique with different scales is employed to exploit edges of the input image, where coarse super-pixel patches provide global classification information while fine ones further provide detail information. Different from common RGB image, hyperspectral image has multi bands to adjust the cluster center with more high precision. After segmentation, each super pixel is classified by recently-developed sparse representation-based classification (SRC), which assigns label for testing samples in one local patch by means of sparse linear combination of all the training samples. Furthermore, segmentation with multiple scales is employed because single scale is not suitable for complicate distribution of medical hyperspectral imagery. Finally, classification results for different sizes of super pixel are fused by some fusion strategy, offering at least two benefits: (1) the final result is obviously superior to that of segmentation with single scale, and (2) the fusion process significantly simplifies the choice of scales. Experimental results using real medical hyperspectral images demonstrate that the proposed method outperforms the state-of-the-art SRC.

Object-based classification of earthquake damage from high-resolution optical imagery using machine learning

NASA Astrophysics Data System (ADS)

Bialas, James; Oommen, Thomas; Rebbapragada, Umaa; Levin, Eugene

2016-07-01

Object-based approaches in the segmentation and classification of remotely sensed images yield more promising results compared to pixel-based approaches. However, the development of an object-based approach presents challenges in terms of algorithm selection and parameter tuning. Subjective methods are often used, but yield less than optimal results. Objective methods are warranted, especially for rapid deployment in time-sensitive applications, such as earthquake damage assessment. Herein, we used a systematic approach in evaluating object-based image segmentation and machine learning algorithms for the classification of earthquake damage in remotely sensed imagery. We tested a variety of algorithms and parameters on post-event aerial imagery for the 2011 earthquake in Christchurch, New Zealand. Results were compared against manually selected test cases representing different classes. In doing so, we can evaluate the effectiveness of the segmentation and classification of different classes and compare different levels of multistep image segmentations. Our classifier is compared against recent pixel-based and object-based classification studies for postevent imagery of earthquake damage. Our results show an improvement against both pixel-based and object-based methods for classifying earthquake damage in high resolution, post-event imagery.

Object-Based Classification as an Alternative Approach to the Traditional Pixel-Based Classification to Identify Potential Habitat of the Grasshopper Sparrow

NASA Astrophysics Data System (ADS)

Jobin, Benoît; Labrecque, Sandra; Grenier, Marcelle; Falardeau, Gilles

2008-01-01

The traditional method of identifying wildlife habitat distribution over large regions consists of pixel-based classification of satellite images into a suite of habitat classes used to select suitable habitat patches. Object-based classification is a new method that can achieve the same objective based on the segmentation of spectral bands of the image creating homogeneous polygons with regard to spatial or spectral characteristics. The segmentation algorithm does not solely rely on the single pixel value, but also on shape, texture, and pixel spatial continuity. The object-based classification is a knowledge base process where an interpretation key is developed using ground control points and objects are assigned to specific classes according to threshold values of determined spectral and/or spatial attributes. We developed a model using the eCognition software to identify suitable habitats for the Grasshopper Sparrow, a rare and declining species found in southwestern Québec. The model was developed in a region with known breeding sites and applied on other images covering adjacent regions where potential breeding habitats may be present. We were successful in locating potential habitats in areas where dairy farming prevailed but failed in an adjacent region covered by a distinct Landsat scene and dominated by annual crops. We discuss the added value of this method, such as the possibility to use the contextual information associated to objects and the ability to eliminate unsuitable areas in the segmentation and land cover classification processes, as well as technical and logistical constraints. A series of recommendations on the use of this method and on conservation issues of Grasshopper Sparrow habitat is also provided.

Parallel exploitation of a spatial-spectral classification approach for hyperspectral images on RVC-CAL

NASA Astrophysics Data System (ADS)

Lazcano, R.; Madroñal, D.; Fabelo, H.; Ortega, S.; Salvador, R.; Callicó, G. M.; Juárez, E.; Sanz, C.

2017-10-01

Hyperspectral Imaging (HI) assembles high resolution spectral information from hundreds of narrow bands across the electromagnetic spectrum, thus generating 3D data cubes in which each pixel gathers the spectral information of the reflectance of every spatial pixel. As a result, each image is composed of large volumes of data, which turns its processing into a challenge, as performance requirements have been continuously tightened. For instance, new HI applications demand real-time responses. Hence, parallel processing becomes a necessity to achieve this requirement, so the intrinsic parallelism of the algorithms must be exploited. In this paper, a spatial-spectral classification approach has been implemented using a dataflow language known as RVCCAL. This language represents a system as a set of functional units, and its main advantage is that it simplifies the parallelization process by mapping the different blocks over different processing units. The spatial-spectral classification approach aims at refining the classification results previously obtained by using a K-Nearest Neighbors (KNN) filtering process, in which both the pixel spectral value and the spatial coordinates are considered. To do so, KNN needs two inputs: a one-band representation of the hyperspectral image and the classification results provided by a pixel-wise classifier. Thus, spatial-spectral classification algorithm is divided into three different stages: a Principal Component Analysis (PCA) algorithm for computing the one-band representation of the image, a Support Vector Machine (SVM) classifier, and the KNN-based filtering algorithm. The parallelization of these algorithms shows promising results in terms of computational time, as the mapping of them over different cores presents a speedup of 2.69x when using 3 cores. Consequently, experimental results demonstrate that real-time processing of hyperspectral images is achievable.

Fuzzy Classification of Ocean Color Satellite Data for Bio-optical Algorithm Constituent Retrievals

NASA Technical Reports Server (NTRS)

Campbell, Janet W.

1998-01-01

The ocean has been traditionally viewed as a 2 class system. Morel and Prieur (1977) classified ocean water according to the dominant absorbent particle suspended in the water column. Case 1 is described as having a high concentration of phytoplankton (and detritus) relative to other particles. Conversely, case 2 is described as having inorganic particles such as suspended sediments in high concentrations. Little work has gone into the problem of mixing bio-optical models for these different water types. An approach is put forth here to blend bio-optical algorithms based on a fuzzy classification scheme. This scheme involves two procedures. First, a clustering procedure identifies classes and builds class statistics from in-situ optical measurements. Next, a classification procedure assigns satellite pixels partial memberships to these classes based on their ocean color reflectance signature. These membership assignments can be used as the basis for a weighting retrievals from class-specific bio-optical algorithms. This technique is demonstrated with in-situ optical measurements and an image from the SeaWiFS ocean color satellite.

Conifer health classification for Colorado, 2008

USGS Publications Warehouse

Cole, Christopher J.; Noble, Suzanne M.; Blauer, Steven L.; Friesen, Beverly A.; Curry, Stacy E.; Bauer, Mark A.

2010-01-01

Colorado has undergone substantial changes in forests due to urbanization, wildfires, insect-caused tree mortality, and other human and environmental factors. The U.S. Geological Survey Rocky Mountain Geographic Science Center evaluated and developed a methodology for applying remotely-sensed imagery for assessing conifer health in Colorado. Two classes were identified for the purposes of this study: healthy and unhealthy (for example, an area the size of a 30- x 30-m pixel with 20 percent or greater visibly dead trees was defined as ?unhealthy?). Medium-resolution Landsat 5 Thematic Mapper imagery were collected. The normalized, reflectance-converted, cloud-filled Landsat scenes were merged to form a statewide image mosaic, and a Normalized Difference Vegetation Index (NDVI) and Renormalized Difference Infrared Index (RDII) were derived. A supervised maximum likelihood classification was done using the Landsat multispectral bands, the NDVI, the RDII, and 30-m U.S. Geological Survey National Elevation Dataset (NED). The classification was constrained to pixels identified in the updated landcover dataset as coniferous or mixed coniferous/deciduous vegetation. The statewide results were merged with a separate health assessment of Grand County, Colo., produced in late 2008. Sampling and validation was done by collecting field data and high-resolution imagery. The 86 percent overall classification accuracy attained in this study suggests that the data and methods used successfully characterized conifer conditions within Colorado. Although forest conditions for Lodgepole Pine (Pinus contorta) are easily characterized, classification uncertainty exists between healthy/unhealthy Ponderosa Pine (Pinus ponderosa), Pi?on (Pinus edulis), and Juniper (Juniperus sp.) vegetation. Some underestimation of conifer mortality in Summit County is likely, where recent (2008) cloud-free imagery was unavailable. These classification uncertainties are primarily due to the spatial and temporal resolution of Landsat, and of the NLCD derived from this sensor. It is believed that high- to moderate-resolution multispectral imagery, coupled with field data, could significantly reduce the uncertainty rates. The USGS produced a four-county follow-up conifer health assessment using high-resolution RapidEye remotely sensed imagery and field data collected in 2009.

Performance analysis of mineral mapping method to delineate mineralization zones under tropical region

NASA Astrophysics Data System (ADS)

Wakila, M. H.; Saepuloh, A.; Heriawan, M. N.; Susanto, A.

2016-09-01

Geothermal explorations and productions are currently being intensively conducted at certain areas in Indonesia such as Wayang Windu Geothermal Field (WWGF) in West Java, Indonesia. The WWGF is located at wide area covering about 40 km2. An accurate method to map the distribution of heterogeneity minerals is necessary for wide areas such as WWGF. Mineral mapping is an important method in geothermal explorations to determine the distribution of minerals which indicate the surface manifestations of geothermal system. This study is aimed to determine the most precise and accurate methods for minerals mapping at geothermal field. Field measurements were performed to assess the accuracy of three proposed methods: 1) Minimum Noise Fraction (MNF), utilizing the linear transformation method to eliminate the correlation among the spectra bands and to reduce the noise in the data, 2) Pixel Purity Index (PPI), a designed method to find the most extreme spectrum pixels and their characteristics due to end-members mixing, 3) Spectral Angle Mapper (SAM), an image classification technique by measuring the spectral similarity between an unknown object with spectral reference in n- dimension. The output of those methods were mineral distribution occurrence. The performance of each mapping method was analyzed based on the ground truth data. Among the three proposed method, the SAM classification method is the most appropriate and accurate for mineral mapping related to spatial distribution of alteration minerals.

Evaluation of seabed mapping methods for fine-scale classification of extremely shallow benthic habitats - Application to the Venice Lagoon, Italy

NASA Astrophysics Data System (ADS)

Montereale Gavazzi, G.; Madricardo, F.; Janowski, L.; Kruss, A.; Blondel, P.; Sigovini, M.; Foglini, F.

2016-03-01

Recent technological developments of multibeam echosounder systems (MBES) allow mapping of benthic habitats with unprecedented detail. MBES can now be employed in extremely shallow waters, challenging data acquisition (as these instruments were often designed for deeper waters) and data interpretation (honed on datasets with resolution sometimes orders of magnitude lower). With extremely high-resolution bathymetry and co-located backscatter data, it is now possible to map the spatial distribution of fine scale benthic habitats, even identifying the acoustic signatures of single sponges. In this context, it is necessary to understand which of the commonly used segmentation methods is best suited to account for such level of detail. At the same time, new sampling protocols for precisely geo-referenced ground truth data need to be developed to validate the benthic environmental classification. This study focuses on a dataset collected in a shallow (2-10 m deep) tidal channel of the Lagoon of Venice, Italy. Using 0.05-m and 0.2-m raster grids, we compared a range of classifications, both pixel-based and object-based approaches, including manual, Maximum Likelihood Classifier, Jenks Optimization clustering, textural analysis and Object Based Image Analysis. Through a comprehensive and accurately geo-referenced ground truth dataset, we were able to identify five different classes of the substrate composition, including sponges, mixed submerged aquatic vegetation, mixed detritic bottom (fine and coarse) and unconsolidated bare sediment. We computed estimates of accuracy (namely Overall, User, Producer Accuracies and the Kappa statistic) by cross tabulating predicted and reference instances. Overall, pixel based segmentations produced the highest accuracies and the accuracy assessment is strongly dependent on the number of classes chosen for the thematic output. Tidal channels in the Venice Lagoon are extremely important in terms of habitats and sediment distribution, particularly within the context of the new tidal barrier being built. However, they had remained largely unexplored until now, because of the surveying challenges. The application of this remote sensing approach, combined with targeted sampling, opens a new perspective in the monitoring of benthic habitats in view of a knowledge-based management of natural resources in shallow coastal areas.

A patch-based convolutional neural network for remote sensing image classification.

PubMed

Sharma, Atharva; Liu, Xiuwen; Yang, Xiaojun; Shi, Di

2017-11-01

Availability of accurate land cover information over large areas is essential to the global environment sustainability; digital classification using medium-resolution remote sensing data would provide an effective method to generate the required land cover information. However, low accuracy of existing per-pixel based classification methods for medium-resolution data is a fundamental limiting factor. While convolutional neural networks (CNNs) with deep layers have achieved unprecedented improvements in object recognition applications that rely on fine image structures, they cannot be applied directly to medium-resolution data due to lack of such fine structures. In this paper, considering the spatial relation of a pixel to its neighborhood, we propose a new deep patch-based CNN system tailored for medium-resolution remote sensing data. The system is designed by incorporating distinctive characteristics of medium-resolution data; in particular, the system computes patch-based samples from multidimensional top of atmosphere reflectance data. With a test site from the Florida Everglades area (with a size of 771 square kilometers), the proposed new system has outperformed pixel-based neural network, pixel-based CNN and patch-based neural network by 24.36%, 24.23% and 11.52%, respectively, in overall classification accuracy. By combining the proposed deep CNN and the huge collection of medium-resolution remote sensing data, we believe that much more accurate land cover datasets can be produced over large areas. Copyright © 2017 Elsevier Ltd. All rights reserved.

Object-oriented and pixel-based classification approach for land cover using airborne long-wave infrared hyperspectral data

NASA Astrophysics Data System (ADS)

Marwaha, Richa; Kumar, Anil; Kumar, Arumugam Senthil

2015-01-01

Our primary objective was to explore a classification algorithm for thermal hyperspectral data. Minimum noise fraction is applied to thermal hyperspectral data and eight pixel-based classifiers, i.e., constrained energy minimization, matched filter, spectral angle mapper (SAM), adaptive coherence estimator, orthogonal subspace projection, mixture-tuned matched filter, target-constrained interference-minimized filter, and mixture-tuned target-constrained interference minimized filter are tested. The long-wave infrared (LWIR) has not yet been exploited for classification purposes. The LWIR data contain emissivity and temperature information about an object. A highest overall accuracy of 90.99% was obtained using the SAM algorithm for the combination of thermal data with a colored digital photograph. Similarly, an object-oriented approach is applied to thermal data. The image is segmented into meaningful objects based on properties such as geometry, length, etc., which are grouped into pixels using a watershed algorithm and an applied supervised classification algorithm, i.e., support vector machine (SVM). The best algorithm in the pixel-based category is the SAM technique. SVM is useful for thermal data, providing a high accuracy of 80.00% at a scale value of 83 and a merge value of 90, whereas for the combination of thermal data with a colored digital photograph, SVM gives the highest accuracy of 85.71% at a scale value of 82 and a merge value of 90.

Estimating Daily Evapotranspiration Based on A Model of Evapotranspiration Fraction (EF) for Mixed Pixels

NASA Astrophysics Data System (ADS)

Xin, X.; Li, F.; Peng, Z.; Qinhuo, L.

2017-12-01

Land surface heterogeneities significantly affect the reliability and accuracy of remotely sensed evapotranspiration (ET), and it gets worse for lower resolution data. At the same time, temporal scale extrapolation of the instantaneous latent heat flux (LE) at satellite overpass time to daily ET are crucial for applications of such remote sensing product. The purpose of this paper is to propose a simple but efficient model for estimating daytime evapotranspiration considering heterogeneity of mixed pixels. In order to do so, an equation to calculate evapotranspiration fraction (EF) of mixed pixels was derived based on two key assumptions. Assumption 1: the available energy (AE) of each sub-pixel equals approximately to that of any other sub-pixels in the same mixed pixel within acceptable margin of bias, and as same as the AE of the mixed pixel. It's only for a simpification of the equation, and its uncertainties and resulted errors in estimated ET are very small. Assumption 2: EF of each sub-pixel equals to the EF of the nearest pure pixel(s) of same land cover type. This equation is supposed to be capable of correcting the spatial scale error of the mixed pixels EF and can be used to calculated daily ET with daily AE data.The model was applied to an artificial oasis in the midstream of Heihe River. HJ-1B satellite data were used to estimate the lumped fluxes at the scale of 300 m after resampling the 30-m resolution datasets to 300 m resolution, which was used to carry on the key step of the model. The results before and after correction were compare to each other and validated using site data of eddy-correlation systems. Results indicated that the new model is capable of improving accuracy of daily ET estimation relative to the lumped method. Validations at 12 sites of eddy-correlation systems for 9 days of HJ-1B overpass showed that the R² increased to 0.82 from 0.62; the RMSE decreased to 1.60 MJ/m² from 2.47MJ/m²; the MBE decreased from 1.92 MJ/m² to 1.18MJ/m², which is a quite significant enhancement.The model is easy to apply. And the moduler of inhomogeneous surfaces is independent and easy to be embedded in the traditional remote sensing algorithms of heat fluxes to get daily ET, which were mainly designed to calculate LE or ET under unsaturated conditions and did not consider heterogeneities of land surface.

Aircraft target detection algorithm based on high resolution spaceborne SAR imagery

NASA Astrophysics Data System (ADS)

Zhang, Hui; Hao, Mengxi; Zhang, Cong; Su, Xiaojing

2018-03-01

In this paper, an image classification algorithm for airport area is proposed, which based on the statistical features of synthetic aperture radar (SAR) images and the spatial information of pixels. The algorithm combines Gamma mixture model and MRF. The algorithm using Gamma mixture model to obtain the initial classification result. Pixel space correlation based on the classification results are optimized by the MRF technique. Additionally, morphology methods are employed to extract airport (ROI) region where the suspected aircraft target samples are clarified to reduce the false alarm and increase the detection performance. Finally, this paper presents the plane target detection, which have been verified by simulation test.

Pneumothorax detection in chest radiographs using convolutional neural networks

NASA Astrophysics Data System (ADS)

Blumenfeld, Aviel; Konen, Eli; Greenspan, Hayit

2018-02-01

This study presents a computer assisted diagnosis system for the detection of pneumothorax (PTX) in chest radiographs based on a convolutional neural network (CNN) for pixel classification. Using a pixel classification approach allows utilization of the texture information in the local environment of each pixel while training a CNN model on millions of training patches extracted from a relatively small dataset. The proposed system uses a pre-processing step of lung field segmentation to overcome the large variability in the input images coming from a variety of imaging sources and protocols. Using a CNN classification, suspected pixel candidates are extracted within each lung segment. A postprocessing step follows to remove non-physiological suspected regions and noisy connected components. The overall percentage of suspected PTX area was used as a robust global decision for the presence of PTX in each lung. The system was trained on a set of 117 chest x-ray images with ground truth segmentations of the PTX regions. The system was tested on a set of 86 images and reached diagnosis accuracy of AUC=0.95. Overall preliminary results are promising and indicate the growing ability of CAD based systems to detect findings in medical imaging on a clinical level accuracy.

Comparison of Pixel-Based and Object-Based Classification Using Parameters and Non-Parameters Approach for the Pattern Consistency of Multi Scale Landcover

NASA Astrophysics Data System (ADS)

Juniati, E.; Arrofiqoh, E. N.

2017-09-01

Information extraction from remote sensing data especially land cover can be obtained by digital classification. In practical some people are more comfortable using visual interpretation to retrieve land cover information. However, it is highly influenced by subjectivity and knowledge of interpreter, also takes time in the process. Digital classification can be done in several ways, depend on the defined mapping approach and assumptions on data distribution. The study compared several classifiers method for some data type at the same location. The data used Landsat 8 satellite imagery, SPOT 6 and Orthophotos. In practical, the data used to produce land cover map in 1:50,000 map scale for Landsat, 1:25,000 map scale for SPOT and 1:5,000 map scale for Orthophotos, but using visual interpretation to retrieve information. Maximum likelihood Classifiers (MLC) which use pixel-based and parameters approach applied to such data, and also Artificial Neural Network classifiers which use pixel-based and non-parameters approach applied too. Moreover, this study applied object-based classifiers to the data. The classification system implemented is land cover classification on Indonesia topographic map. The classification applied to data source, which is expected to recognize the pattern and to assess consistency of the land cover map produced by each data. Furthermore, the study analyse benefits and limitations the use of methods.

Ningaloo Reef: Shallow Marine Habitats Mapped Using a Hyperspectral Sensor

PubMed Central

Kobryn, Halina T.; Wouters, Kristin; Beckley, Lynnath E.; Heege, Thomas

2013-01-01

Research, monitoring and management of large marine protected areas require detailed and up-to-date habitat maps. Ningaloo Marine Park (including the Muiron Islands) in north-western Australia (stretching across three degrees of latitude) was mapped to 20 m depth using HyMap airborne hyperspectral imagery (125 bands) at 3.5 m resolution across the 762 km2 of reef environment between the shoreline and reef slope. The imagery was corrected for atmospheric, air-water interface and water column influences to retrieve bottom reflectance and bathymetry using the physics-based Modular Inversion and Processing System. Using field-validated, image-derived spectra from a representative range of cover types, the classification combined a semi-automated, pixel-based approach with fuzzy logic and derivative techniques. Five thematic classification levels for benthic cover (with probability maps) were generated with varying degrees of detail, ranging from a basic one with three classes (biotic, abiotic and mixed) to the most detailed with 46 classes. The latter consisted of all abiotic and biotic seabed components and hard coral growth forms in dominant or mixed states. The overall accuracy of mapping for the most detailed maps was 70% for the highest classification level. Macro-algal communities formed most of the benthic cover, while hard and soft corals represented only about 7% of the mapped area (58.6 km2). Dense tabulate coral was the largest coral mosaic type (37% of all corals) and the rest of the corals were a mix of tabulate, digitate, massive and soft corals. Our results show that for this shallow, fringing reef environment situated in the arid tropics, hyperspectral remote sensing techniques can offer an efficient and cost-effective approach to mapping and monitoring reef habitats over large, remote and inaccessible areas. PMID:23922921

Aggregation of Sentinel-2 time series classifications as a solution for multitemporal analysis

NASA Astrophysics Data System (ADS)

Lewiński, Stanislaw; Nowakowski, Artur; Malinowski, Radek; Rybicki, Marcin; Kukawska, Ewa; Krupiński, Michał

2017-10-01

The general aim of this work was to elaborate efficient and reliable aggregation method that could be used for creating a land cover map at a global scale from multitemporal satellite imagery. The study described in this paper presents methods for combining results of land cover/land use classifications performed on single-date Sentinel-2 images acquired at different time periods. For that purpose different aggregation methods were proposed and tested on study sites spread on different continents. The initial classifications were performed with Random Forest classifier on individual Sentinel-2 images from a time series. In the following step the resulting land cover maps were aggregated pixel by pixel using three different combinations of information on the number of occurrences of a certain land cover class within a time series and the posterior probability of particular classes resulting from the Random Forest classification. From the proposed methods two are shown superior and in most cases were able to reach or outperform the accuracy of the best individual classifications of single-date images. Moreover, the aggregations results are very stable when used on data with varying cloudiness. They also enable to reduce considerably the number of cloudy pixels in the resulting land cover map what is significant advantage for mapping areas with frequent cloud coverage.

Advances in Spectral-Spatial Classification of Hyperspectral Images

NASA Technical Reports Server (NTRS)

Fauvel, Mathieu; Tarabalka, Yuliya; Benediktsson, Jon Atli; Chanussot, Jocelyn; Tilton, James C.

2012-01-01

Recent advances in spectral-spatial classification of hyperspectral images are presented in this paper. Several techniques are investigated for combining both spatial and spectral information. Spatial information is extracted at the object (set of pixels) level rather than at the conventional pixel level. Mathematical morphology is first used to derive the morphological profile of the image, which includes characteristics about the size, orientation, and contrast of the spatial structures present in the image. Then, the morphological neighborhood is defined and used to derive additional features for classification. Classification is performed with support vector machines (SVMs) using the available spectral information and the extracted spatial information. Spatial postprocessing is next investigated to build more homogeneous and spatially consistent thematic maps. To that end, three presegmentation techniques are applied to define regions that are used to regularize the preliminary pixel-wise thematic map. Finally, a multiple-classifier (MC) system is defined to produce relevant markers that are exploited to segment the hyperspectral image with the minimum spanning forest algorithm. Experimental results conducted on three real hyperspectral images with different spatial and spectral resolutions and corresponding to various contexts are presented. They highlight the importance of spectral–spatial strategies for the accurate classification of hyperspectral images and validate the proposed methods.

SVM Pixel Classification on Colour Image Segmentation

NASA Astrophysics Data System (ADS)

Barui, Subhrajit; Latha, S.; Samiappan, Dhanalakshmi; Muthu, P.

2018-04-01

The aim of image segmentation is to simplify the representation of an image with the help of cluster pixels into something meaningful to analyze. Segmentation is typically used to locate boundaries and curves in an image, precisely to label every pixel in an image to give each pixel an independent identity. SVM pixel classification on colour image segmentation is the topic highlighted in this paper. It holds useful application in the field of concept based image retrieval, machine vision, medical imaging and object detection. The process is accomplished step by step. At first we need to recognize the type of colour and the texture used as an input to the SVM classifier. These inputs are extracted via local spatial similarity measure model and Steerable filter also known as Gabon Filter. It is then trained by using FCM (Fuzzy C-Means). Both the pixel level information of the image and the ability of the SVM Classifier undergoes some sophisticated algorithm to form the final image. The method has a well developed segmented image and efficiency with respect to increased quality and faster processing of the segmented image compared with the other segmentation methods proposed earlier. One of the latest application result is the Light L16 camera.

The effect of imposing 'fractional abundance constraints' onto the multilayer perceptron for sub-pixel land cover classification

NASA Astrophysics Data System (ADS)

Heremans, Stien; Suykens, Johan A. K.; Van Orshoven, Jos

2016-02-01

To be physically interpretable, sub-pixel land cover fractions or abundances should fulfill two constraints, the Abundance Non-negativity Constraint (ANC) and the Abundance Sum-to-one Constraint (ASC). This paper focuses on the effect of imposing these constraints onto the MultiLayer Perceptron (MLP) for a multi-class sub-pixel land cover classification of a time series of low resolution MODIS-images covering the northern part of Belgium. Two constraining modes were compared, (i) an in-training approach that uses 'softmax' as the transfer function in the MLP's output layer and (ii) a post-training approach that linearly rescales the outputs of the unconstrained MLP. Our results demonstrate that the pixel-level prediction accuracy is markedly increased by the explicit enforcement, both in-training and post-training, of the ANC and the ASC. For aggregations of pixels (municipalities), the constrained perceptrons perform at least as well as their unconstrained counterparts. Although the difference in performance between the in-training and post-training approach is small, we recommend the former for integrating the fractional abundance constraints into MLPs meant for sub-pixel land cover estimation, regardless of the targeted level of spatial aggregation.

An experiment in multispectral, multitemporal crop classification using relaxation techniques

NASA Technical Reports Server (NTRS)

Davis, L. S.; Wang, C.-Y.; Xie, H.-C

1983-01-01

The paper describes the result of an experimental study concerning the use of probabilistic relaxation for improving pixel classification rates. Two LACIE sites were used in the study and in both cases, relaxation resulted in a marked improvement in classification rates.

The effect of the atmosphere on the classification of satellite observations to identify surface features

NASA Technical Reports Server (NTRS)

Fraser, R. S.; Bahethi, O. P.; Al-Abbas, A. H.

1977-01-01

The effect of differences in atmospheric turbidity on the classification of Landsat 1 observations of a rural scene is presented. The observations are classified by an unsupervised clustering technique. These clusters serve as a training set for use of a maximum-likelihood algorithm. The measured radiances in each of the four spectral bands are then changed by amounts measured by Landsat 1. These changes can be associated with a decrease in atmospheric turbidity by a factor of 1.3. The classification of 22% of the pixels changes as a result of the modification. The modified observations are then reclassified as an independent set. Only 3% of the pixels have a different classification than the unmodified set. Hence, if classification errors of rural areas are not to exceed 15%, a new training set has to be developed whenever the difference in turbidity between the training and test sets reaches unity.

Subpixel Mapping of Hyperspectral Image Based on Linear Subpixel Feature Detection and Object Optimization

NASA Astrophysics Data System (ADS)

Liu, Zhaoxin; Zhao, Liaoying; Li, Xiaorun; Chen, Shuhan

2018-04-01

Owing to the limitation of spatial resolution of the imaging sensor and the variability of ground surfaces, mixed pixels are widesperead in hyperspectral imagery. The traditional subpixel mapping algorithms treat all mixed pixels as boundary-mixed pixels while ignoring the existence of linear subpixels. To solve this question, this paper proposed a new subpixel mapping method based on linear subpixel feature detection and object optimization. Firstly, the fraction value of each class is obtained by spectral unmixing. Secondly, the linear subpixel features are pre-determined based on the hyperspectral characteristics and the linear subpixel feature; the remaining mixed pixels are detected based on maximum linearization index analysis. The classes of linear subpixels are determined by using template matching method. Finally, the whole subpixel mapping results are iteratively optimized by binary particle swarm optimization algorithm. The performance of the proposed subpixel mapping method is evaluated via experiments based on simulated and real hyperspectral data sets. The experimental results demonstrate that the proposed method can improve the accuracy of subpixel mapping.

Hyperspectral classification of grassland species: towards a UAS application for semi-automatic field surveys

NASA Astrophysics Data System (ADS)

Lopatin, Javier; Fassnacht, Fabian E.; Kattenborn, Teja; Schmidtlein, Sebastian

2017-04-01

Grasslands are one of the ecosystems that have been strongly intervened during the past decades due to anthropogenic impacts, affecting their structural and functional composition. To monitor the spatial and/or temporal changes of these environments, a reliable field survey is first needed. As quality relevés are usually expensive and time consuming, the amount of information available is usually poor or not well spatially distributed at the regional scale. In the present study, we investigate the possibility of a semi-automated method used for repeated surveys of monitoring sites. We analyze the applicability of very high spatial resolution hyperspectral data to classify grassland species at the level of individuals. The AISA+ imaging spectrometer mounted on a scaffold was applied to scan 1 m2 grassland plots and assess the impact of four sources of variation on the predicted species cover: (1) the spatial resolution of the scans, (2) the species number and structural diversity, (3) the species cover, and (4) the species functional types (bryophytes, forbs and graminoids). We found that the spatial resolution and the diversity level (mainly structural diversity) were the most important source of variation for the proposed approach. A spatial resolution below 1 cm produced relatively high model performances, while predictions with pixel sizes over that threshold produced non adequate results. Areas with low interspecies overlap reached classification median values of 0.8 (kappa). On the contrary, results were not satisfactory in plots with frequent interspecies overlap in multiple layers. By means of a bootstrapping procedure, we found that areas with shadows and mixed pixels introduce uncertainties into the classification. We conclude that the application of very high resolution hyperspectral remote sensing as a robust alternative or supplement to field surveys is possible for environments with low structural heterogeneity. This study presents the first try of a full classification of grassland species at the individuum level using spectral data.

Research on a pulmonary nodule segmentation method combining fast self-adaptive FCM and classification.

PubMed

Liu, Hui; Zhang, Cai-Ming; Su, Zhi-Yuan; Wang, Kai; Deng, Kai

2015-01-01

The key problem of computer-aided diagnosis (CAD) of lung cancer is to segment pathologically changed tissues fast and accurately. As pulmonary nodules are potential manifestation of lung cancer, we propose a fast and self-adaptive pulmonary nodules segmentation method based on a combination of FCM clustering and classification learning. The enhanced spatial function considers contributions to fuzzy membership from both the grayscale similarity between central pixels and single neighboring pixels and the spatial similarity between central pixels and neighborhood and improves effectively the convergence rate and self-adaptivity of the algorithm. Experimental results show that the proposed method can achieve more accurate segmentation of vascular adhesion, pleural adhesion, and ground glass opacity (GGO) pulmonary nodules than other typical algorithms.

Interactive classification and content-based retrieval of tissue images

NASA Astrophysics Data System (ADS)

Aksoy, Selim; Marchisio, Giovanni B.; Tusk, Carsten; Koperski, Krzysztof

2002-11-01

We describe a system for interactive classification and retrieval of microscopic tissue images. Our system models tissues in pixel, region and image levels. Pixel level features are generated using unsupervised clustering of color and texture values. Region level features include shape information and statistics of pixel level feature values. Image level features include statistics and spatial relationships of regions. To reduce the gap between low-level features and high-level expert knowledge, we define the concept of prototype regions. The system learns the prototype regions in an image collection using model-based clustering and density estimation. Different tissue types are modeled using spatial relationships of these regions. Spatial relationships are represented by fuzzy membership functions. The system automatically selects significant relationships from training data and builds models which can also be updated using user relevance feedback. A Bayesian framework is used to classify tissues based on these models. Preliminary experiments show that the spatial relationship models we developed provide a flexible and powerful framework for classification and retrieval of tissue images.

Cloud cover analysis with Arctic Advanced Very High Resolution Radiometer data. II - Classification with spectral and textural measures

NASA Technical Reports Server (NTRS)

Key, J.

1990-01-01

The spectral and textural characteristics of polar clouds and surfaces for a 7-day summer series of AVHRR data in two Arctic locations are examined, and the results used in the development of a cloud classification procedure for polar satellite data. Since spatial coherence and texture sensitivity tests indicate that a joint spectral-textural analysis based on the same cell size is inappropriate, cloud detection with AVHRR data and surface identification with passive microwave data are first done on the pixel level as described by Key and Barry (1989). Next, cloud patterns within 250-sq-km regions are described, then the spectral and local textural characteristics of cloud patterns in the image are determined and each cloud pixel is classified by statistical methods. Results indicate that both spectral and textural features can be utilized in the classification of cloudy pixels, although spectral features are most useful for the discrimination between cloud classes.

Prevalence of pure versus mixed snow cover pixels across spatial resolutions in alpine environments: implications for binary and fractional remote sensing approaches

USGS Publications Warehouse

Selkowitz, David J.; Forster, Richard; Caldwell, Megan K.

2014-01-01

Remote sensing of snow-covered area (SCA) can be binary (indicating the presence/absence of snow cover at each pixel) or fractional (indicating the fraction of each pixel covered by snow). Fractional SCA mapping provides more information than binary SCA, but is more difficult to implement and may not be feasible with all types of remote sensing data. The utility of fractional SCA mapping relative to binary SCA mapping varies with the intended application as well as by spatial resolution, temporal resolution and period of interest, and climate. We quantified the frequency of occurrence of partially snow-covered (mixed) pixels at spatial resolutions between 1 m and 500 m over five dates at two study areas in the western U.S., using 0.5 m binary SCA maps derived from high spatial resolution imagery aggregated to fractional SCA at coarser spatial resolutions. In addition, we used in situ monitoring to estimate the frequency of partially snow-covered conditions for the period September 2013–August 2014 at 10 60-m grid cell footprints at two study areas with continental snow climates. Results from the image analysis indicate that at 40 m, slightly above the nominal spatial resolution of Landsat, mixed pixels accounted for 25%–93% of total pixels, while at 500 m, the nominal spatial resolution of MODIS bands used for snow cover mapping, mixed pixels accounted for 67%–100% of total pixels. Mixed pixels occurred more commonly at the continental snow climate site than at the maritime snow climate site. The in situ data indicate that some snow cover was present between 186 and 303 days, and partial snow cover conditions occurred on 10%–98% of days with snow cover. Four sites remained partially snow-free throughout most of the winter and spring, while six sites were entirely snow covered throughout most or all of the winter and spring. Within 60 m grid cells, the late spring/summer transition from snow-covered to snow-free conditions lasted 17–56 days and averaged 37 days. Our results suggest that mixed snow-covered snow-free pixels are common at the spatial resolutions imaged by both the Landsat and MODIS sensors. This highlights the additional information available from fractional SCA products and suggests fractional SCA can provide a major advantage for hydrological and climatological monitoring and modeling, particularly when accurate representation of the spatial distribution of snow cover is critical.

Blob-level active-passive data fusion for Benthic classification

NASA Astrophysics Data System (ADS)

Park, Joong Yong; Kalluri, Hemanth; Mathur, Abhinav; Ramnath, Vinod; Kim, Minsu; Aitken, Jennifer; Tuell, Grady

2012-06-01

We extend the data fusion pixel level to the more semantically meaningful blob level, using the mean-shift algorithm to form labeled blobs having high similarity in the feature domain, and connectivity in the spatial domain. We have also developed Bhattacharyya Distance (BD) and rule-based classifiers, and have implemented these higher-level data fusion algorithms into the CZMIL Data Processing System. Applying these new algorithms to recent SHOALS and CASI data at Plymouth Harbor, Massachusetts, we achieved improved benthic classification accuracies over those produced with either single sensor, or pixel-level fusion strategies. These results appear to validate the hypothesis that classification accuracy may be generally improved by adopting higher spatial and semantic levels of fusion.

PixelLearn

NASA Technical Reports Server (NTRS)

Mazzoni, Dominic; Wagstaff, Kiri; Bornstein, Benjamin; Tang, Nghia; Roden, Joseph

2006-01-01

PixelLearn is an integrated user-interface computer program for classifying pixels in scientific images. Heretofore, training a machine-learning algorithm to classify pixels in images has been tedious and difficult. PixelLearn provides a graphical user interface that makes it faster and more intuitive, leading to more interactive exploration of image data sets. PixelLearn also provides image-enhancement controls to make it easier to see subtle details in images. PixelLearn opens images or sets of images in a variety of common scientific file formats and enables the user to interact with several supervised or unsupervised machine-learning pixel-classifying algorithms while the user continues to browse through the images. The machinelearning algorithms in PixelLearn use advanced clustering and classification methods that enable accuracy much higher than is achievable by most other software previously available for this purpose. PixelLearn is written in portable C++ and runs natively on computers running Linux, Windows, or Mac OS X.

IMPROVING THE ACCURACY OF HISTORIC SATELLITE IMAGE CLASSIFICATION BY COMBINING LOW-RESOLUTION MULTISPECTRAL DATA WITH HIGH-RESOLUTION PANCHROMATIC DATA

DOE Office of Scientific and Technical Information (OSTI.GOV)

Getman, Daniel J

2008-01-01

Many attempts to observe changes in terrestrial systems over time would be significantly enhanced if it were possible to improve the accuracy of classifications of low-resolution historic satellite data. In an effort to examine improving the accuracy of historic satellite image classification by combining satellite and air photo data, two experiments were undertaken in which low-resolution multispectral data and high-resolution panchromatic data were combined and then classified using the ECHO spectral-spatial image classification algorithm and the Maximum Likelihood technique. The multispectral data consisted of 6 multispectral channels (30-meter pixel resolution) from Landsat 7. These data were augmented with panchromatic datamore » (15m pixel resolution) from Landsat 7 in the first experiment, and with a mosaic of digital aerial photography (1m pixel resolution) in the second. The addition of the Landsat 7 panchromatic data provided a significant improvement in the accuracy of classifications made using the ECHO algorithm. Although the inclusion of aerial photography provided an improvement in accuracy, this improvement was only statistically significant at a 40-60% level. These results suggest that once error levels associated with combining aerial photography and multispectral satellite data are reduced, this approach has the potential to significantly enhance the precision and accuracy of classifications made using historic remotely sensed data, as a way to extend the time range of efforts to track temporal changes in terrestrial systems.« less

A Hierarchical Object-oriented Urban Land Cover Classification Using WorldView-2 Imagery and Airborne LiDAR data

NASA Astrophysics Data System (ADS)

Wu, M. F.; Sun, Z. C.; Yang, B.; Yu, S. S.

2016-11-01

In order to reduce the “salt and pepper” in pixel-based urban land cover classification and expand the application of fusion of multi-source data in the field of urban remote sensing, WorldView-2 imagery and airborne Light Detection and Ranging (LiDAR) data were used to improve the classification of urban land cover. An approach of object- oriented hierarchical classification was proposed in our study. The processing of proposed method consisted of two hierarchies. (1) In the first hierarchy, LiDAR Normalized Digital Surface Model (nDSM) image was segmented to objects. The NDVI, Costal Blue and nDSM thresholds were set for extracting building objects. (2) In the second hierarchy, after removing building objects, WorldView-2 fused imagery was obtained by Haze-ratio-based (HR) fusion, and was segmented. A SVM classifier was applied to generate road/parking lot, vegetation and bare soil objects. (3) Trees and grasslands were split based on an nDSM threshold (2.4 meter). The results showed that compared with pixel-based and non-hierarchical object-oriented approach, proposed method provided a better performance of urban land cover classification, the overall accuracy (OA) and overall kappa (OK) improved up to 92.75% and 0.90. Furthermore, proposed method reduced “salt and pepper” in pixel-based classification, improved the extraction accuracy of buildings based on LiDAR nDSM image segmentation, and reduced the confusion between trees and grasslands through setting nDSM threshold.

Mapping forested wetlands in the Great Zhan River Basin through integrating optical, radar, and topographical data classification techniques.

PubMed

Na, X D; Zang, S Y; Wu, C S; Li, W L

2015-11-01

Knowledge of the spatial extent of forested wetlands is essential to many studies including wetland functioning assessment, greenhouse gas flux estimation, and wildlife suitable habitat identification. For discriminating forested wetlands from their adjacent land cover types, researchers have resorted to image analysis techniques applied to numerous remotely sensed data. While with some success, there is still no consensus on the optimal approaches for mapping forested wetlands. To address this problem, we examined two machine learning approaches, random forest (RF) and K-nearest neighbor (KNN) algorithms, and applied these two approaches to the framework of pixel-based and object-based classifications. The RF and KNN algorithms were constructed using predictors derived from Landsat 8 imagery, Radarsat-2 advanced synthetic aperture radar (SAR), and topographical indices. The results show that the objected-based classifications performed better than per-pixel classifications using the same algorithm (RF) in terms of overall accuracy and the difference of their kappa coefficients are statistically significant (p<0.01). There were noticeably omissions for forested and herbaceous wetlands based on the per-pixel classifications using the RF algorithm. As for the object-based image analysis, there were also statistically significant differences (p<0.01) of Kappa coefficient between results performed based on RF and KNN algorithms. The object-based classification using RF provided a more visually adequate distribution of interested land cover types, while the object classifications based on the KNN algorithm showed noticeably commissions for forested wetlands and omissions for agriculture land. This research proves that the object-based classification with RF using optical, radar, and topographical data improved the mapping accuracy of land covers and provided a feasible approach to discriminate the forested wetlands from the other land cover types in forestry area.

Hybrid Optimization of Object-Based Classification in High-Resolution Images Using Continous ANT Colony Algorithm with Emphasis on Building Detection

NASA Astrophysics Data System (ADS)

Tamimi, E.; Ebadi, H.; Kiani, A.

2017-09-01

Automatic building detection from High Spatial Resolution (HSR) images is one of the most important issues in Remote Sensing (RS). Due to the limited number of spectral bands in HSR images, using other features will lead to improve accuracy. By adding these features, the presence probability of dependent features will be increased, which leads to accuracy reduction. In addition, some parameters should be determined in Support Vector Machine (SVM) classification. Therefore, it is necessary to simultaneously determine classification parameters and select independent features according to image type. Optimization algorithm is an efficient method to solve this problem. On the other hand, pixel-based classification faces several challenges such as producing salt-paper results and high computational time in high dimensional data. Hence, in this paper, a novel method is proposed to optimize object-based SVM classification by applying continuous Ant Colony Optimization (ACO) algorithm. The advantages of the proposed method are relatively high automation level, independency of image scene and type, post processing reduction for building edge reconstruction and accuracy improvement. The proposed method was evaluated by pixel-based SVM and Random Forest (RF) classification in terms of accuracy. In comparison with optimized pixel-based SVM classification, the results showed that the proposed method improved quality factor and overall accuracy by 17% and 10%, respectively. Also, in the proposed method, Kappa coefficient was improved by 6% rather than RF classification. Time processing of the proposed method was relatively low because of unit of image analysis (image object). These showed the superiority of the proposed method in terms of time and accuracy.

Evaluating the Visualization of What a Deep Neural Network Has Learned.

PubMed

Samek, Wojciech; Binder, Alexander; Montavon, Gregoire; Lapuschkin, Sebastian; Muller, Klaus-Robert

Deep neural networks (DNNs) have demonstrated impressive performance in complex machine learning tasks such as image classification or speech recognition. However, due to their multilayer nonlinear structure, they are not transparent, i.e., it is hard to grasp what makes them arrive at a particular classification or recognition decision, given a new unseen data sample. Recently, several approaches have been proposed enabling one to understand and interpret the reasoning embodied in a DNN for a single test image. These methods quantify the "importance" of individual pixels with respect to the classification decision and allow a visualization in terms of a heatmap in pixel/input space. While the usefulness of heatmaps can be judged subjectively by a human, an objective quality measure is missing. In this paper, we present a general methodology based on region perturbation for evaluating ordered collections of pixels such as heatmaps. We compare heatmaps computed by three different methods on the SUN397, ILSVRC2012, and MIT Places data sets. Our main result is that the recently proposed layer-wise relevance propagation algorithm qualitatively and quantitatively provides a better explanation of what made a DNN arrive at a particular classification decision than the sensitivity-based approach or the deconvolution method. We provide theoretical arguments to explain this result and discuss its practical implications. Finally, we investigate the use of heatmaps for unsupervised assessment of the neural network performance.Deep neural networks (DNNs) have demonstrated impressive performance in complex machine learning tasks such as image classification or speech recognition. However, due to their multilayer nonlinear structure, they are not transparent, i.e., it is hard to grasp what makes them arrive at a particular classification or recognition decision, given a new unseen data sample. Recently, several approaches have been proposed enabling one to understand and interpret the reasoning embodied in a DNN for a single test image. These methods quantify the "importance" of individual pixels with respect to the classification decision and allow a visualization in terms of a heatmap in pixel/input space. While the usefulness of heatmaps can be judged subjectively by a human, an objective quality measure is missing. In this paper, we present a general methodology based on region perturbation for evaluating ordered collections of pixels such as heatmaps. We compare heatmaps computed by three different methods on the SUN397, ILSVRC2012, and MIT Places data sets. Our main result is that the recently proposed layer-wise relevance propagation algorithm qualitatively and quantitatively provides a better explanation of what made a DNN arrive at a particular classification decision than the sensitivity-based approach or the deconvolution method. We provide theoretical arguments to explain this result and discuss its practical implications. Finally, we investigate the use of heatmaps for unsupervised assessment of the neural network performance.

Image Segmentation Analysis for NASA Earth Science Applications

NASA Technical Reports Server (NTRS)

Tilton, James C.

2010-01-01

NASA collects large volumes of imagery data from satellite-based Earth remote sensing sensors. Nearly all of the computerized image analysis of this data is performed pixel-by-pixel, in which an algorithm is applied directly to individual image pixels. While this analysis approach is satisfactory in many cases, it is usually not fully effective in extracting the full information content from the high spatial resolution image data that s now becoming increasingly available from these sensors. The field of object-based image analysis (OBIA) has arisen in recent years to address the need to move beyond pixel-based analysis. The Recursive Hierarchical Segmentation (RHSEG) software developed by the author is being used to facilitate moving from pixel-based image analysis to OBIA. The key unique aspect of RHSEG is that it tightly intertwines region growing segmentation, which produces spatially connected region objects, with region object classification, which groups sets of region objects together into region classes. No other practical, operational image segmentation approach has this tight integration of region growing object finding with region classification This integration is made possible by the recursive, divide-and-conquer implementation utilized by RHSEG, in which the input image data is recursively subdivided until the image data sections are small enough to successfully mitigat the combinatorial explosion caused by the need to compute the dissimilarity between each pair of image pixels. RHSEG's tight integration of region growing object finding and region classification is what enables the high spatial fidelity of the image segmentations produced by RHSEG. This presentation will provide an overview of the RHSEG algorithm and describe how it is currently being used to support OBIA or Earth Science applications such as snow/ice mapping and finding archaeological sites from remotely sensed data.

As-Built design specification for the CLASFYT program. [production of classification files - crop inventory

NASA Technical Reports Server (NTRS)

Horton, C. L. (Principal Investigator)

1981-01-01

The CLASFYT program is described in detail. The program produces a one-channel universal-formatted classification file. Trajectory coefficients and a composite set of tolerance values are calculated from five acquisitions of radiance values in each of the training fields corresponding to up to ten agricultural products. These coefficients and tolerance values are used to classify each pixel in the test field of the same segment to be the same agricultural product as one of the training fields, none of the products or a screened pixel.

Research on a Pulmonary Nodule Segmentation Method Combining Fast Self-Adaptive FCM and Classification

PubMed Central

Liu, Hui; Zhang, Cai-Ming; Su, Zhi-Yuan; Wang, Kai; Deng, Kai

2015-01-01

The key problem of computer-aided diagnosis (CAD) of lung cancer is to segment pathologically changed tissues fast and accurately. As pulmonary nodules are potential manifestation of lung cancer, we propose a fast and self-adaptive pulmonary nodules segmentation method based on a combination of FCM clustering and classification learning. The enhanced spatial function considers contributions to fuzzy membership from both the grayscale similarity between central pixels and single neighboring pixels and the spatial similarity between central pixels and neighborhood and improves effectively the convergence rate and self-adaptivity of the algorithm. Experimental results show that the proposed method can achieve more accurate segmentation of vascular adhesion, pleural adhesion, and ground glass opacity (GGO) pulmonary nodules than other typical algorithms. PMID:25945120

RIPARIAN CHARACTERIZATION USING SUB-PIXEL ANALYSIS OF LANDSAT TM IMAGERY FOR USE IN ECOLOGICAL RISK ASSESSMENT

EPA Science Inventory

Landuse/land cover and riparian corridor characterization for 7 major watersheds in western Ohio was accomplished using sub-pixel analysis and traditional classification techniques. Areas
representing forest, woodland, shrub, and herbaceous vegetation were delineated using a ...

A systematic comparison of different object-based classification techniques using high spatial resolution imagery in agricultural environments

NASA Astrophysics Data System (ADS)

Li, Manchun; Ma, Lei; Blaschke, Thomas; Cheng, Liang; Tiede, Dirk

2016-07-01

Geographic Object-Based Image Analysis (GEOBIA) is becoming more prevalent in remote sensing classification, especially for high-resolution imagery. Many supervised classification approaches are applied to objects rather than pixels, and several studies have been conducted to evaluate the performance of such supervised classification techniques in GEOBIA. However, these studies did not systematically investigate all relevant factors affecting the classification (segmentation scale, training set size, feature selection and mixed objects). In this study, statistical methods and visual inspection were used to compare these factors systematically in two agricultural case studies in China. The results indicate that Random Forest (RF) and Support Vector Machines (SVM) are highly suitable for GEOBIA classifications in agricultural areas and confirm the expected general tendency, namely that the overall accuracies decline with increasing segmentation scale. All other investigated methods except for RF and SVM are more prone to obtain a lower accuracy due to the broken objects at fine scales. In contrast to some previous studies, the RF classifiers yielded the best results and the k-nearest neighbor classifier were the worst results, in most cases. Likewise, the RF and Decision Tree classifiers are the most robust with or without feature selection. The results of training sample analyses indicated that the RF and adaboost. M1 possess a superior generalization capability, except when dealing with small training sample sizes. Furthermore, the classification accuracies were directly related to the homogeneity/heterogeneity of the segmented objects for all classifiers. Finally, it was suggested that RF should be considered in most cases for agricultural mapping.

Locality-preserving sparse representation-based classification in hyperspectral imagery

NASA Astrophysics Data System (ADS)

Gao, Lianru; Yu, Haoyang; Zhang, Bing; Li, Qingting

2016-10-01

This paper proposes to combine locality-preserving projections (LPP) and sparse representation (SR) for hyperspectral image classification. The LPP is first used to reduce the dimensionality of all the training and testing data by finding the optimal linear approximations to the eigenfunctions of the Laplace Beltrami operator on the manifold, where the high-dimensional data lies. Then, SR codes the projected testing pixels as sparse linear combinations of all the training samples to classify the testing pixels by evaluating which class leads to the minimum approximation error. The integration of LPP and SR represents an innovative contribution to the literature. The proposed approach, called locality-preserving SR-based classification, addresses the imbalance between high dimensionality of hyperspectral data and the limited number of training samples. Experimental results on three real hyperspectral data sets demonstrate that the proposed approach outperforms the original counterpart, i.e., SR-based classification.

Diverse Region-Based CNN for Hyperspectral Image Classification.

PubMed

Zhang, Mengmeng; Li, Wei; Du, Qian

2018-06-01

Convolutional neural network (CNN) is of great interest in machine learning and has demonstrated excellent performance in hyperspectral image classification. In this paper, we propose a classification framework, called diverse region-based CNN, which can encode semantic context-aware representation to obtain promising features. With merging a diverse set of discriminative appearance factors, the resulting CNN-based representation exhibits spatial-spectral context sensitivity that is essential for accurate pixel classification. The proposed method exploiting diverse region-based inputs to learn contextual interactional features is expected to have more discriminative power. The joint representation containing rich spectral and spatial information is then fed to a fully connected network and the label of each pixel vector is predicted by a softmax layer. Experimental results with widely used hyperspectral image data sets demonstrate that the proposed method can surpass any other conventional deep learning-based classifiers and other state-of-the-art classifiers.

Improving urban land use and land cover classification from high-spatial-resolution hyperspectral imagery using contextual information

NASA Astrophysics Data System (ADS)

Yang, He; Ma, Ben; Du, Qian; Yang, Chenghai

2010-08-01

In this paper, we propose approaches to improve the pixel-based support vector machine (SVM) classification for urban land use and land cover (LULC) mapping from airborne hyperspectral imagery with high spatial resolution. Class spatial neighborhood relationship is used to correct the misclassified class pairs, such as roof and trail, road and roof. These classes may be difficult to be separated because they may have similar spectral signatures and their spatial features are not distinct enough to help their discrimination. In addition, misclassification incurred from within-class trivial spectral variation can be corrected by using pixel connectivity information in a local window so that spectrally homogeneous regions can be well preserved. Our experimental results demonstrate the efficiency of the proposed approaches in classification accuracy improvement. The overall performance is competitive to the object-based SVM classification.

Automated cloud classification with a fuzzy logic expert system

NASA Technical Reports Server (NTRS)

Tovinkere, Vasanth; Baum, Bryan A.

1993-01-01

An unresolved problem in current cloud retrieval algorithms concerns the analysis of scenes containing overlapping cloud layers. Cloud parameterizations are very important both in global climate models and in studies of the Earth's radiation budget. Most cloud retrieval schemes, such as the bispectral method used by the International Satellite Cloud Climatology Project (ISCCP), have no way of determining whether overlapping cloud layers exist in any group of satellite pixels. One promising method uses fuzzy logic to determine whether mixed cloud and/or surface types exist within a group of pixels, such as cirrus, land, and water, or cirrus and stratus. When two or more class types are present, fuzzy logic uses membership values to assign the group of pixels partially to the different class types. The strength of fuzzy logic lies in its ability to work with patterns that may include more than one class, facilitating greater information extraction from satellite radiometric data. The development of the fuzzy logic rule-based expert system involves training the fuzzy classifier with spectral and textural features calculated from accurately labeled 32x32 regions of Advanced Very High Resolution Radiometer (AVHRR) 1.1-km data. The spectral data consists of AVHRR channels 1 (0.55-0.68 mu m), 2 (0.725-1.1 mu m), 3 (3.55-3.93 mu m), 4 (10.5-11.5 mu m), and 5 (11.5-12.5 mu m), which include visible, near-infrared, and infrared window regions. The textural features are based on the gray level difference vector (GLDV) method. A sophisticated new interactive visual image Classification System (IVICS) is used to label samples chosen from scenes collected during the FIRE IFO II. The training samples are chosen from predefined classes, chosen to be ocean, land, unbroken stratiform, broken stratiform, and cirrus. The November 28, 1991 NOAA overpasses contain complex multilevel cloud situations ideal for training and validating the fuzzy logic expert system.

Crop identification technology assessment for remote sensing (CITARS). Volume 10: Interpretation of results

NASA Technical Reports Server (NTRS)

Bizzell, R. M.; Feiveson, A. H.; Hall, F. G.; Bauer, M. E.; Davis, B. J.; Malila, W. A.; Rice, D. P.

1975-01-01

The CITARS was an experiment designed to quantitatively evaluate crop identification performance for corn and soybeans in various environments using a well-defined set of automatic data processing (ADP) techniques. Each technique was applied to data acquired to recognize and estimate proportions of corn and soybeans. The CITARS documentation summarizes, interprets, and discusses the crop identification performances obtained using (1) different ADP procedures; (2) a linear versus a quadratic classifier; (3) prior probability information derived from historic data; (4) local versus nonlocal recognition training statistics and the associated use of preprocessing; (5) multitemporal data; (6) classification bias and mixed pixels in proportion estimation; and (7) data with differnt site characteristics, including crop, soil, atmospheric effects, and stages of crop maturity.

Concrete Slump Classification using GLCM Feature Extraction

NASA Astrophysics Data System (ADS)

Andayani, Relly; Madenda, Syarifudin

2016-05-01

Digital image processing technologies have been widely applies in analyzing concrete structure because the accuracy and real time result. The aim of this study is to classify concrete slump by using image processing technique. For this purpose, concrete mix design of 30 MPa compression strength designed with slump of 0-10 mm, 10-30 mm, 30-60 mm, and 60-180 mm were analysed. Image acquired by Nikon Camera D-7000 using high resolution was set up. In the first step RGB converted to greyimage than cropped to 1024 x 1024 pixel. With open-source program, cropped images to be analysed to extract GLCM feature. The result shows for the higher slump contrast getting lower, but higher correlation, energy, and homogeneity.

Data Field Modeling and Spectral-Spatial Feature Fusion for Hyperspectral Data Classification.

PubMed

Liu, Da; Li, Jianxun

2016-12-16

Classification is a significant subject in hyperspectral remote sensing image processing. This study proposes a spectral-spatial feature fusion algorithm for the classification of hyperspectral images (HSI). Unlike existing spectral-spatial classification methods, the influences and interactions of the surroundings on each measured pixel were taken into consideration in this paper. Data field theory was employed as the mathematical realization of the field theory concept in physics, and both the spectral and spatial domains of HSI were considered as data fields. Therefore, the inherent dependency of interacting pixels was modeled. Using data field modeling, spatial and spectral features were transformed into a unified radiation form and further fused into a new feature by using a linear model. In contrast to the current spectral-spatial classification methods, which usually simply stack spectral and spatial features together, the proposed method builds the inner connection between the spectral and spatial features, and explores the hidden information that contributed to classification. Therefore, new information is included for classification. The final classification result was obtained using a random forest (RF) classifier. The proposed method was tested with the University of Pavia and Indian Pines, two well-known standard hyperspectral datasets. The experimental results demonstrate that the proposed method has higher classification accuracies than those obtained by the traditional approaches.

Single-pixel imaging based on compressive sensing with spectral-domain optical mixing

NASA Astrophysics Data System (ADS)

Zhu, Zhijing; Chi, Hao; Jin, Tao; Zheng, Shilie; Jin, Xiaofeng; Zhang, Xianmin

2017-11-01

In this letter a single-pixel imaging structure is proposed based on compressive sensing using a spatial light modulator (SLM)-based spectrum shaper. In the approach, an SLM-based spectrum shaper, the pattern of which is a predetermined pseudorandom bit sequence (PRBS), spectrally codes the optical pulse carrying image information. The energy of the spectrally mixed pulse is detected by a single-pixel photodiode and the measurement results are used to reconstruct the image via a sparse recovery algorithm. As the mixing of the image signal and the PRBS is performed in the spectral domain, optical pulse stretching, modulation, compression and synchronization in the time domain are avoided. Experiments are implemented to verify the feasibility of the approach.

Sub-Pixel Mapping of Tree Canopy, Impervious Surfaces, and Cropland in the Laurentian Great Lakes Basin Using MODIS Time-Series Data

EPA Science Inventory

This research examined sub-pixel land-cover classification performance for tree canopy, impervious surface, and cropland in the Laurentian Great Lakes Basin (GLB) using both timeseries MODIS (MOderate Resolution Imaging Spectroradiometer) NDVI (Normalized Difference Vegetation In...

Evaluating an ensemble classification approach for crop diversity verification in Danish greening subsidy control

NASA Astrophysics Data System (ADS)

Chellasamy, Menaka; Ferré, Ty Paul Andrew; Greve, Mogens Humlekrog

2016-07-01

Beginning in 2015, Danish farmers are obliged to meet specific crop diversification rules based on total land area and number of crops cultivated to be eligible for new greening subsidies. Hence, there is a need for the Danish government to extend their subsidy control system to verify farmers' declarations to warrant greening payments under the new crop diversification rules. Remote Sensing (RS) technology has been used since 1992 to control farmers' subsidies in Denmark. However, a proper RS-based approach is yet to be finalised to validate new crop diversity requirements designed for assessing compliance under the recent subsidy scheme (2014-2020); This study uses an ensemble classification approach (proposed by the authors in previous studies) for validating the crop diversity requirements of the new rules. The approach uses a neural network ensemble classification system with bi-temporal (spring and early summer) WorldView-2 imagery (WV2) and includes the following steps: (1) automatic computation of pixel-based prediction probabilities using multiple neural networks; (2) quantification of the classification uncertainty using Endorsement Theory (ET); (3) discrimination of crop pixels and validation of the crop diversification rules at farm level; and (4) identification of farmers who are violating the requirements for greening subsidies. The prediction probabilities are computed by a neural network ensemble supplied with training samples selected automatically using farmers declared parcels (field vectors containing crop information and the field boundary of each crop). Crop discrimination is performed by considering a set of conclusions derived from individual neural networks based on ET. Verification of the diversification rules is performed by incorporating pixel-based classification uncertainty or confidence intervals with the class labels at the farmer level. The proposed approach was tested with WV2 imagery acquired in 2011 for a study area in Vennebjerg, Denmark, containing 132 farmers, 1258 fields, and 18 crops. The classification results obtained show an overall accuracy of 90.2%. The RS-based results suggest that 36 farmers did not follow the crop diversification rules that would qualify for the greening subsidies. When compared to the farmers' reported crop mixes, irrespective of the rule, the RS results indicate that false crop declarations were made by 8 farmers, covering 15 fields. If the farmers' reports had been submitted for the new greening subsidies, 3 farmers would have made a false claim; while remaining 5 farmers obey the rules of required crop proportion even though they have submitted the false crop code due to their small holding size. The RS results would have supported 96 farmers for greening subsidy claims, with no instances of suggesting a greening subsidy for a holding that the farmer did not report as meeting the required conditions. These results suggest that the proposed RS based method shows great promise for validating the new greening subsidies in Denmark.

Comparison of Sub-pixel Classification Approaches for Crop-specific Mapping

EPA Science Inventory

The Moderate Resolution Imaging Spectroradiometer (MODIS) data has been increasingly used for crop mapping and other agricultural applications. Phenology-based classification approaches using the NDVI (Normalized Difference Vegetation Index) 16-day composite (250 m) data product...

Classification of visible and infrared hyperspectral images based on image segmentation and edge-preserving filtering

NASA Astrophysics Data System (ADS)

Cui, Binge; Ma, Xiudan; Xie, Xiaoyun; Ren, Guangbo; Ma, Yi

2017-03-01

The classification of hyperspectral images with a few labeled samples is a major challenge which is difficult to meet unless some spatial characteristics can be exploited. In this study, we proposed a novel spectral-spatial hyperspectral image classification method that exploited spatial autocorrelation of hyperspectral images. First, image segmentation is performed on the hyperspectral image to assign each pixel to a homogeneous region. Second, the visible and infrared bands of hyperspectral image are partitioned into multiple subsets of adjacent bands, and each subset is merged into one band. Recursive edge-preserving filtering is performed on each merged band which utilizes the spectral information of neighborhood pixels. Third, the resulting spectral and spatial feature band set is classified using the SVM classifier. Finally, bilateral filtering is performed to remove "salt-and-pepper" noise in the classification result. To preserve the spatial structure of hyperspectral image, edge-preserving filtering is applied independently before and after the classification process. Experimental results on different hyperspectral images prove that the proposed spectral-spatial classification approach is robust and offers more classification accuracy than state-of-the-art methods when the number of labeled samples is small.

Error analysis of filtering operations in pixel-duplicated images of diabetic retinopathy

NASA Astrophysics Data System (ADS)

Mehrubeoglu, Mehrube; McLauchlan, Lifford

2010-08-01

In this paper, diabetic retinopathy is chosen for a sample target image to demonstrate the effectiveness of image enlargement through pixel duplication in identifying regions of interest. Pixel duplication is presented as a simpler alternative to data interpolation techniques for detecting small structures in the images. A comparative analysis is performed on different image processing schemes applied to both original and pixel-duplicated images. Structures of interest are detected and and classification parameters optimized for minimum false positive detection in the original and enlarged retinal pictures. The error analysis demonstrates the advantages as well as shortcomings of pixel duplication in image enhancement when spatial averaging operations (smoothing filters) are also applied.

ASSESSMENT OF LANDSCAPE CHARACTERISTICS ON THEMATIC IMAGE CLASSIFICATION ACCURACY

EPA Science Inventory

Landscape characteristics such as small patch size and land cover heterogeneity have been hypothesized to increase the likelihood of misclassifying pixels during thematic image classification. However, there has been a lack of empirical evidence, to support these hypotheses. This...

Supervised classification of brain tissues through local multi-scale texture analysis by coupling DIR and FLAIR MR sequences

NASA Astrophysics Data System (ADS)

Poletti, Enea; Veronese, Elisa; Calabrese, Massimiliano; Bertoldo, Alessandra; Grisan, Enrico

2012-02-01

The automatic segmentation of brain tissues in magnetic resonance (MR) is usually performed on T1-weighted images, due to their high spatial resolution. T1w sequence, however, has some major downsides when brain lesions are present: the altered appearance of diseased tissues causes errors in tissues classification. In order to overcome these drawbacks, we employed two different MR sequences: fluid attenuated inversion recovery (FLAIR) and double inversion recovery (DIR). The former highlights both gray matter (GM) and white matter (WM), the latter highlights GM alone. We propose here a supervised classification scheme that does not require any anatomical a priori information to identify the 3 classes, "GM", "WM", and "background". Features are extracted by means of a local multi-scale texture analysis, computed for each pixel of the DIR and FLAIR sequences. The 9 textures considered are average, standard deviation, kurtosis, entropy, contrast, correlation, energy, homogeneity, and skewness, evaluated on a neighborhood of 3x3, 5x5, and 7x7 pixels. Hence, the total number of features associated to a pixel is 56 (9 textures x3 scales x2 sequences +2 original pixel values). The classifier employed is a Support Vector Machine with Radial Basis Function as kernel. From each of the 4 brain volumes evaluated, a DIR and a FLAIR slice have been selected and manually segmented by 2 expert neurologists, providing 1st and 2nd human reference observations which agree with an average accuracy of 99.03%. SVM performances have been assessed with a 4-fold cross-validation, yielding an average classification accuracy of 98.79%.

Optimizing selection of training and auxiliary data for operational land cover classification for the LCMAP initiative

NASA Astrophysics Data System (ADS)

Zhu, Zhe; Gallant, Alisa L.; Woodcock, Curtis E.; Pengra, Bruce; Olofsson, Pontus; Loveland, Thomas R.; Jin, Suming; Dahal, Devendra; Yang, Limin; Auch, Roger F.

2016-12-01

The U.S. Geological Survey's Land Change Monitoring, Assessment, and Projection (LCMAP) initiative is a new end-to-end capability to continuously track and characterize changes in land cover, use, and condition to better support research and applications relevant to resource management and environmental change. Among the LCMAP product suite are annual land cover maps that will be available to the public. This paper describes an approach to optimize the selection of training and auxiliary data for deriving the thematic land cover maps based on all available clear observations from Landsats 4-8. Training data were selected from map products of the U.S. Geological Survey's Land Cover Trends project. The Random Forest classifier was applied for different classification scenarios based on the Continuous Change Detection and Classification (CCDC) algorithm. We found that extracting training data proportionally to the occurrence of land cover classes was superior to an equal distribution of training data per class, and suggest using a total of 20,000 training pixels to classify an area about the size of a Landsat scene. The problem of unbalanced training data was alleviated by extracting a minimum of 600 training pixels and a maximum of 8000 training pixels per class. We additionally explored removing outliers contained within the training data based on their spectral and spatial criteria, but observed no significant improvement in classification results. We also tested the importance of different types of auxiliary data that were available for the conterminous United States, including: (a) five variables used by the National Land Cover Database, (b) three variables from the cloud screening "Function of mask" (Fmask) statistics, and (c) two variables from the change detection results of CCDC. We found that auxiliary variables such as a Digital Elevation Model and its derivatives (aspect, position index, and slope), potential wetland index, water probability, snow probability, and cloud probability improved the accuracy of land cover classification. Compared to the original strategy of the CCDC algorithm (500 pixels per class), the use of the optimal strategy improved the classification accuracies substantially (15-percentage point increase in overall accuracy and 4-percentage point increase in minimum accuracy).

IMPACTS OF PATCH SIZE AND LANDSCAPE HETEROGENEITY ON THEMATIC IMAGE CLASSIFICATION ACCURACY

EPA Science Inventory

Impacts of Patch Size and Landscape Heterogeneity on Thematic Image Classification Accuracy.
Currently, most thematic accuracy assessments of classified remotely sensed images oily account for errors between the various classes employed, at particular pixels of interest, thu...

3D Spatial and Spectral Fusion of Terrestrial Hyperspectral Imagery and Lidar for Hyperspectral Image Shadow Restoration Applied to a Geologic Outcrop

NASA Astrophysics Data System (ADS)

Hartzell, P. J.; Glennie, C. L.; Hauser, D. L.; Okyay, U.; Khan, S.; Finnegan, D. C.

2016-12-01

Recent advances in remote sensing technology have expanded the acquisition and fusion of active lidar and passive hyperspectral imagery (HSI) from an exclusively airborne technique to terrestrial modalities. This enables high resolution 3D spatial and spectral quantification of vertical geologic structures for applications such as virtual 3D rock outcrop models for hydrocarbon reservoir analog analysis and mineral quantification in open pit mining environments. In contrast to airborne observation geometry, the vertical surfaces observed by horizontal-viewing terrestrial HSI sensors are prone to extensive topography-induced solar shadowing, which leads to reduced pixel classification accuracy or outright removal of shadowed pixels from analysis tasks. Using a precisely calibrated and registered offset cylindrical linear array camera model, we demonstrate the use of 3D lidar data for sub-pixel HSI shadow detection and the restoration of the shadowed pixel spectra via empirical methods that utilize illuminated and shadowed pixels of similar material composition. We further introduce a new HSI shadow restoration technique that leverages collocated backscattered lidar intensity, which is resistant to solar conditions, obtained by projecting the 3D lidar points through the HSI camera model into HSI pixel space. Using ratios derived from the overlapping lidar laser and HSI wavelengths, restored shadow pixel spectra are approximated using a simple scale factor. Simulations of multiple lidar wavelengths, i.e., multi-spectral lidar, indicate the potential for robust HSI spectral restoration that is independent of the complexity and costs associated with rigorous radiometric transfer models, which have yet to be developed for horizontal-viewing terrestrial HSI sensors. The spectral restoration performance is quantified through HSI pixel classification consistency between full sun and partial sun exposures of a single geologic outcrop.

Sensitivity of geographic information system outputs to errors in remotely sensed data

NASA Technical Reports Server (NTRS)

Ramapriyan, H. K.; Boyd, R. K.; Gunther, F. J.; Lu, Y. C.

1981-01-01

The sensitivity of the outputs of a geographic information system (GIS) to errors in inputs derived from remotely sensed data (RSD) is investigated using a suitability model with per-cell decisions and a gridded geographic data base whose cells are larger than the RSD pixels. The process of preparing RSD as input to a GIS is analyzed, and the errors associated with classification and registration are examined. In the case of the model considered, it is found that the errors caused during classification and registration are partially compensated by the aggregation of pixels. The compensation is quantified by means of an analytical model, a Monte Carlo simulation, and experiments with Landsat data. The results show that error reductions of the order of 50% occur because of aggregation when 25 pixels of RSD are used per cell in the geographic data base.

A Visual mining based framework for classification accuracy estimation

NASA Astrophysics Data System (ADS)

Arun, Pattathal Vijayakumar

2013-12-01

Classification techniques have been widely used in different remote sensing applications and correct classification of mixed pixels is a tedious task. Traditional approaches adopt various statistical parameters, however does not facilitate effective visualisation. Data mining tools are proving very helpful in the classification process. We propose a visual mining based frame work for accuracy assessment of classification techniques using open source tools such as WEKA and PREFUSE. These tools in integration can provide an efficient approach for getting information about improvements in the classification accuracy and helps in refining training data set. We have illustrated framework for investigating the effects of various resampling methods on classification accuracy and found that bilinear (BL) is best suited for preserving radiometric characteristics. We have also investigated the optimal number of folds required for effective analysis of LISS-IV images. Techniki klasyfikacji są szeroko wykorzystywane w różnych aplikacjach teledetekcyjnych, w których poprawna klasyfikacja pikseli stanowi poważne wyzwanie. Podejście tradycyjne wykorzystujące różnego rodzaju parametry statystyczne nie zapewnia efektywnej wizualizacji. Wielce obiecujące wydaje się zastosowanie do klasyfikacji narzędzi do eksploracji danych. W artykule zaproponowano podejście bazujące na wizualnej analizie eksploracyjnej, wykorzystujące takie narzędzia typu open source jak WEKA i PREFUSE. Wymienione narzędzia ułatwiają korektę pół treningowych i efektywnie wspomagają poprawę dokładności klasyfikacji. Działanie metody sprawdzono wykorzystując wpływ różnych metod resampling na zachowanie dokładności radiometrycznej i uzyskując najlepsze wyniki dla metody bilinearnej (BL).

Modification of the random forest algorithm to avoid statistical dependence problems when classifying remote sensing imagery

NASA Astrophysics Data System (ADS)

Cánovas-García, Fulgencio; Alonso-Sarría, Francisco; Gomariz-Castillo, Francisco; Oñate-Valdivieso, Fernando

2017-06-01

Random forest is a classification technique widely used in remote sensing. One of its advantages is that it produces an estimation of classification accuracy based on the so called out-of-bag cross-validation method. It is usually assumed that such estimation is not biased and may be used instead of validation based on an external data-set or a cross-validation external to the algorithm. In this paper we show that this is not necessarily the case when classifying remote sensing imagery using training areas with several pixels or objects. According to our results, out-of-bag cross-validation clearly overestimates accuracy, both overall and per class. The reason is that, in a training patch, pixels or objects are not independent (from a statistical point of view) of each other; however, they are split by bootstrapping into in-bag and out-of-bag as if they were really independent. We believe that putting whole patch, rather than pixels/objects, in one or the other set would produce a less biased out-of-bag cross-validation. To deal with the problem, we propose a modification of the random forest algorithm to split training patches instead of the pixels (or objects) that compose them. This modified algorithm does not overestimate accuracy and has no lower predictive capability than the original. When its results are validated with an external data-set, the accuracy is not different from that obtained with the original algorithm. We analysed three remote sensing images with different classification approaches (pixel and object based); in the three cases reported, the modification we propose produces a less biased accuracy estimation.

Soccer player recognition by pixel classification in a hybrid color space

NASA Astrophysics Data System (ADS)

Vandenbroucke, Nicolas; Macaire, Ludovic; Postaire, Jack-Gerard

1997-08-01

Soccer is a very popular sport all over the world, Coaches and sport commentators need accurate information about soccer games, especially about the players behavior. These information can be gathered by inspectors who watch the soccer match and report manually the actions of the players involved in the principal phases of the game. Generally, these inspectors focus their attention on the few players standing near the ball and don't report about the motion of all the other players. So it seems desirable to design a system which automatically tracks all the players in real- time. That's why we propose to automatically track each player through the successive color images of the sequences acquired by a fixed color camera. Each player which is present in the image, is modelized by an active contour model or snake. When, during the soccer match, a player is hidden by another, the snakes which track these two players merge. So, it becomes impossible to track the players, except if the snakes are interactively re-initialized. Fortunately, in most cases, the two players don't belong to the same team. That is why we present an algorithm which recognizes the teams of the players by pixels representing the soccer ground which must be withdrawn before considering the players themselves. To eliminate these pixels, the color characteristics of the ground are determined interactively. In a second step, dealing with windows containing only one player of one team, the color features which yield the best discrimination between the two teams are selected. Thanks to these color features, the pixels associated to the players of the two teams form two separated clusters into a color space. In fact, there are many color representation systems and it's interesting to evaluate the features which provide the best separation between the two classes of pixels according to the players soccer suit. Finally, the classification process for image segmentation is based on the three most discriminating color features which define the coordinates of each pixel in an 'hybrid color space.' Thanks to this hybrid color representation, each pixel can be assigned to one of the two classes by a minimum distance classification.

Non-parametric analysis of LANDSAT maps using neural nets and parallel computers

NASA Technical Reports Server (NTRS)

Salu, Yehuda; Tilton, James

1991-01-01

Nearest neighbor approaches and a new neural network, the Binary Diamond, are used for the classification of images of ground pixels obtained by LANDSAT satellite. The performances are evaluated by comparing classifications of a scene in the vicinity of Washington DC. The problem of optimal selection of categories is addressed as a step in the classification process.

A Comparative Study of Landsat TM and SPOT HRG Images for Vegetation Classification in the Brazilian Amazon.

PubMed

Lu, Dengsheng; Batistella, Mateus; de Miranda, Evaristo E; Moran, Emilio

2008-01-01

Complex forest structure and abundant tree species in the moist tropical regions often cause difficulties in classifying vegetation classes with remotely sensed data. This paper explores improvement in vegetation classification accuracies through a comparative study of different image combinations based on the integration of Landsat Thematic Mapper (TM) and SPOT High Resolution Geometric (HRG) instrument data, as well as the combination of spectral signatures and textures. A maximum likelihood classifier was used to classify the different image combinations into thematic maps. This research indicated that data fusion based on HRG multispectral and panchromatic data slightly improved vegetation classification accuracies: a 3.1 to 4.6 percent increase in the kappa coefficient compared with the classification results based on original HRG or TM multispectral images. A combination of HRG spectral signatures and two textural images improved the kappa coefficient by 6.3 percent compared with pure HRG multispectral images. The textural images based on entropy or second-moment texture measures with a window size of 9 pixels × 9 pixels played an important role in improving vegetation classification accuracy. Overall, optical remote-sensing data are still insufficient for accurate vegetation classifications in the Amazon basin.

A Comparative Study of Landsat TM and SPOT HRG Images for Vegetation Classification in the Brazilian Amazon

PubMed Central

Lu, Dengsheng; Batistella, Mateus; de Miranda, Evaristo E.; Moran, Emilio

2009-01-01

Complex forest structure and abundant tree species in the moist tropical regions often cause difficulties in classifying vegetation classes with remotely sensed data. This paper explores improvement in vegetation classification accuracies through a comparative study of different image combinations based on the integration of Landsat Thematic Mapper (TM) and SPOT High Resolution Geometric (HRG) instrument data, as well as the combination of spectral signatures and textures. A maximum likelihood classifier was used to classify the different image combinations into thematic maps. This research indicated that data fusion based on HRG multispectral and panchromatic data slightly improved vegetation classification accuracies: a 3.1 to 4.6 percent increase in the kappa coefficient compared with the classification results based on original HRG or TM multispectral images. A combination of HRG spectral signatures and two textural images improved the kappa coefficient by 6.3 percent compared with pure HRG multispectral images. The textural images based on entropy or second-moment texture measures with a window size of 9 pixels × 9 pixels played an important role in improving vegetation classification accuracy. Overall, optical remote-sensing data are still insufficient for accurate vegetation classifications in the Amazon basin. PMID:19789716

Extraction and Analysis of Mega Cities’ Impervious Surface on Pixel-based and Object-oriented Support Vector Machine Classification Technology: A case of Bombay

NASA Astrophysics Data System (ADS)

Yu, S. S.; Sun, Z. C.; Sun, L.; Wu, M. F.

2017-02-01

The object of this paper is to study the impervious surface extraction method using remote sensing imagery and monitor the spatiotemporal changing patterns of mega cities. Megacity Bombay was selected as the interesting area. Firstly, the pixel-based and object-oriented support vector machine (SVM) classification methods were used to acquire the land use/land cover (LULC) products of Bombay in 2010. Consequently, the overall accuracy (OA) and overall Kappa (OK) of the pixel-based method were 94.97% and 0.96 with a running time of 78 minutes, the OA and OK of the object-oriented method were 93.72% and 0.94 with a running time of only 17s. Additionally, OA and OK of the object-oriented method after a post-classification were improved up to 95.8% and 0.94. Then, the dynamic impervious surfaces of Bombay in the period 1973-2015 were extracted and the urbanization pattern of Bombay was analysed. Results told that both the two SVM classification methods could accomplish the impervious surface extraction, but the object-oriented method should be a better choice. Urbanization of Bombay experienced a fast extending during the past 42 years, implying a dramatically urban sprawl of mega cities in the developing countries along the One Belt and One Road (OBOR).

TEMPORAL CORRELATION OF CLASSIFICATIONS IN REMOTE SENSING

EPA Science Inventory

A bivariate binary model is developed for estimating the change in land cover from satellite images obtained at two different times. The binary classifications of a pixel at the two times are modeled as potentially correlated random variables, conditional on the true states of th...

IMPACTS OF PATCH SIZE AND LAND COVER HETEROGENEITY ON THEMATIC IMAGE CLASSIFICATION ACCURACY

EPA Science Inventory

Landscape characteristics such as small patch size and land cover heterogeneity have been hypothesized to increase the likelihood of miss-classifying pixels during thematic image classification. However, there has been a lack of empirical evidence to support these hypotheses,...

Microcomputer-based classification of environmental data in municipal areas

NASA Astrophysics Data System (ADS)

Thiergärtner, H.

1995-10-01

Multivariate data-processing methods used in mineral resource identification can be used to classify urban regions. Using elements of expert systems, geographical information systems, as well as known classification and prognosis systems, it is possible to outline a single model that consists of resistant and of temporary parts of a knowledge base including graphical input and output treatment and of resistant and temporary elements of a bank of methods and algorithms. Whereas decision rules created by experts will be stored in expert systems directly, powerful classification rules in form of resistant but latent (implicit) decision algorithms may be implemented in the suggested model. The latent functions will be transformed into temporary explicit decision rules by learning processes depending on the actual task(s), parameter set(s), pixels selection(s), and expert control(s). This takes place both at supervised and nonsupervised classification of multivariately described pixel sets representing municipal subareas. The model is outlined briefly and illustrated by results obtained in a target area covering a part of the city of Berlin (Germany).

A combined reconstruction-classification method for diffuse optical tomography.

PubMed

Hiltunen, P; Prince, S J D; Arridge, S

2009-11-07

We present a combined classification and reconstruction algorithm for diffuse optical tomography (DOT). DOT is a nonlinear ill-posed inverse problem. Therefore, some regularization is needed. We present a mixture of Gaussians prior, which regularizes the DOT reconstruction step. During each iteration, the parameters of a mixture model are estimated. These associate each reconstructed pixel with one of several classes based on the current estimate of the optical parameters. This classification is exploited to form a new prior distribution to regularize the reconstruction step and update the optical parameters. The algorithm can be described as an iteration between an optimization scheme with zeroth-order variable mean and variance Tikhonov regularization and an expectation-maximization scheme for estimation of the model parameters. We describe the algorithm in a general Bayesian framework. Results from simulated test cases and phantom measurements show that the algorithm enhances the contrast of the reconstructed images with good spatial accuracy. The probabilistic classifications of each image contain only a few misclassified pixels.

a Novel 3d Intelligent Fuzzy Algorithm Based on Minkowski-Clustering

NASA Astrophysics Data System (ADS)

Toori, S.; Esmaeily, A.

2017-09-01

Assessing and monitoring the state of the earth surface is a key requirement for global change research. In this paper, we propose a new consensus fuzzy clustering algorithm that is based on the Minkowski distance. This research concentrates on Tehran's vegetation mass and its changes during 29 years using remote sensing technology. The main purpose of this research is to evaluate the changes in vegetation mass using a new process by combination of intelligent NDVI fuzzy clustering and Minkowski distance operation. The dataset includes the images of Landsat8 and Landsat TM, from 1989 to 2016. For each year three images of three continuous days were used to identify vegetation impact and recovery. The result was a 3D NDVI image, with one dimension for each day NDVI. The next step was the classification procedure which is a complicated process of categorizing pixels into a finite number of separate classes, based on their data values. If a pixel satisfies a certain set of standards, the pixel is allocated to the class that corresponds to those criteria. This method is less sensitive to noise and can integrate solutions from multiple samples of data or attributes for processing data in the processing industry. The result was a fuzzy one dimensional image. This image was also computed for the next 28 years. The classification was done in both specified urban and natural park areas of Tehran. Experiments showed that our method worked better in classifying image pixels in comparison with the standard classification methods.

Digital classification of Landsat data for vegetation and land-cover mapping in the Blackfoot River watershed, southeastern Idaho

USGS Publications Warehouse

Pettinger, L.R.

1982-01-01

This paper documents the procedures, results, and final products of a digital analysis of Landsat data used to produce a vegetation and landcover map of the Blackfoot River watershed in southeastern Idaho. Resource classes were identified at two levels of detail: generalized Level I classes (for example, forest land and wetland) and detailed Levels II and III classes (for example, conifer forest, aspen, wet meadow, and riparian hardwoods). Training set statistics were derived using a modified clustering approach. Environmental stratification that separated uplands from lowlands improved discrimination between resource classes having similar spectral signatures. Digital classification was performed using a maximum likelihood algorithm. Classification accuracy was determined on a single-pixel basis from a random sample of 25-pixel blocks. These blocks were transferred to small-scale color-infrared aerial photographs, and the image area corresponding to each pixel was interpreted. Classification accuracy, expressed as percent agreement of digital classification and photo-interpretation results, was 83.0:t 2.1 percent (0.95 probability level) for generalized (Level I) classes and 52.2:t 2.8 percent (0.95 probability level) for detailed (Levels II and III) classes. After the classified images were geometrically corrected, two types of maps were produced of Level I and Levels II and III resource classes: color-coded maps at a 1:250,000 scale, and flatbed-plotter overlays at a 1:24,000 scale. The overlays are more useful because of their larger scale, familiar format to users, and compatibility with other types of topographic and thematic maps of the same scale.

Mediterranean Land Use and Land Cover Classification Assessment Using High Spatial Resolution Data

NASA Astrophysics Data System (ADS)

Elhag, Mohamed; Boteva, Silvena

2016-10-01

Landscape fragmentation is noticeably practiced in Mediterranean regions and imposes substantial complications in several satellite image classification methods. To some extent, high spatial resolution data were able to overcome such complications. For better classification performances in Land Use Land Cover (LULC) mapping, the current research adopts different classification methods comparison for LULC mapping using Sentinel-2 satellite as a source of high spatial resolution. Both of pixel-based and an object-based classification algorithms were assessed; the pixel-based approach employs Maximum Likelihood (ML), Artificial Neural Network (ANN) algorithms, Support Vector Machine (SVM), and, the object-based classification uses the Nearest Neighbour (NN) classifier. Stratified Masking Process (SMP) that integrates a ranking process within the classes based on spectral fluctuation of the sum of the training and testing sites was implemented. An analysis of the overall and individual accuracy of the classification results of all four methods reveals that the SVM classifier was the most efficient overall by distinguishing most of the classes with the highest accuracy. NN succeeded to deal with artificial surface classes in general while agriculture area classes, and forest and semi-natural area classes were segregated successfully with SVM. Furthermore, a comparative analysis indicates that the conventional classification method yielded better accuracy results than the SMP method overall with both classifiers used, ML and SVM.

Some new classification methods for hyperspectral remote sensing

NASA Astrophysics Data System (ADS)

Du, Pei-jun; Chen, Yun-hao; Jones, Simon; Ferwerda, Jelle G.; Chen, Zhi-jun; Zhang, Hua-peng; Tan, Kun; Yin, Zuo-xia

2006-10-01

Hyperspectral Remote Sensing (HRS) is one of the most significant recent achievements of Earth Observation Technology. Classification is the most commonly employed processing methodology. In this paper three new hyperspectral RS image classification methods are analyzed. These methods are: Object-oriented FIRS image classification, HRS image classification based on information fusion and HSRS image classification by Back Propagation Neural Network (BPNN). OMIS FIRS image is used as the example data. Object-oriented techniques have gained popularity for RS image classification in recent years. In such method, image segmentation is used to extract the regions from the pixel information based on homogeneity criteria at first, and spectral parameters like mean vector, texture, NDVI and spatial/shape parameters like aspect ratio, convexity, solidity, roundness and orientation for each region are calculated, finally classification of the image using the region feature vectors and also using suitable classifiers such as artificial neural network (ANN). It proves that object-oriented methods can improve classification accuracy since they utilize information and features both from the point and the neighborhood, and the processing unit is a polygon (in which all pixels are homogeneous and belong to the class). HRS image classification based on information fusion, divides all bands of the image into different groups initially, and extracts features from every group according to the properties of each group. Three levels of information fusion: data level fusion, feature level fusion and decision level fusion are used to HRS image classification. Artificial Neural Network (ANN) can perform well in RS image classification. In order to promote the advances of ANN used for HIRS image classification, Back Propagation Neural Network (BPNN), the most commonly used neural network, is used to HRS image classification.

Estimation of urban surface water at subpixel level from neighborhood pixels using multispectral remote sensing image (Conference Presentation)

NASA Astrophysics Data System (ADS)

Xie, Huan; Luo, Xin; Xu, Xiong; Wang, Chen; Pan, Haiyan; Tong, Xiaohua; Liu, Shijie

2016-10-01

Water body is a fundamental element in urban ecosystems and water mapping is critical for urban and landscape planning and management. As remote sensing has increasingly been used for water mapping in rural areas, this spatially explicit approach applied in urban area is also a challenging work due to the water bodies mainly distributed in a small size and the spectral confusion widely exists between water and complex features in the urban environment. Water index is the most common method for water extraction at pixel level, and spectral mixture analysis (SMA) has been widely employed in analyzing urban environment at subpixel level recently. In this paper, we introduce an automatic subpixel water mapping method in urban areas using multispectral remote sensing data. The objectives of this research consist of: (1) developing an automatic land-water mixed pixels extraction technique by water index; (2) deriving the most representative endmembers of water and land by utilizing neighboring water pixels and adaptive iterative optimal neighboring land pixel for respectively; (3) applying a linear unmixing model for subpixel water fraction estimation. Specifically, to automatically extract land-water pixels, the locally weighted scatter plot smoothing is firstly used to the original histogram curve of WI image . And then the Ostu threshold is derived as the start point to select land-water pixels based on histogram of the WI image with the land threshold and water threshold determination through the slopes of histogram curve . Based on the previous process at pixel level, the image is divided into three parts: water pixels, land pixels, and mixed land-water pixels. Then the spectral mixture analysis (SMA) is applied to land-water mixed pixels for water fraction estimation at subpixel level. With the assumption that the endmember signature of a target pixel should be more similar to adjacent pixels due to spatial dependence, the endmember of water and land are determined by neighboring pure land or pure water pixels within a distance. To obtaining the most representative endmembers in SMA, we designed an adaptive iterative endmember selection method based on the spatial similarity of adjacent pixels. According to the spectral similarity in a spatial adjacent region, the spectrum of land endmember is determined by selecting the most representative land pixel in a local window, and the spectrum of water endmember is determined by calculating an average of the water pixels in the local window. The proposed hierarchical processing method based on WI and SMA (WISMA) is applied to urban areas for reliability evaluation using the Landsat-8 Operational Land Imager (OLI) images. For comparison, four methods at pixel level and subpixel level were chosen respectively. Results indicate that the water maps generated by the proposed method correspond as closely with the truth water maps with subpixel precision. And the results showed that the WISMA achieved the best performance in water mapping with comprehensive analysis of different accuracy evaluation indexes (RMSE and SE).

Extraction of Shrimp Ponds Using Object Oriented Classification vis-a-vis Pixel Based Classification

DTIC Science & Technology

2004-11-01

302 25th ACRS 2004 Chiang Mai , Thailand B-3.6 Data Processing...Proceedings of the 25th Asian Conference on Remote Sensing, Held in Chiang Mai , Thailand on 22-26 November 2004. Copyrighted; Government Purpose Rights... Chiang Mai , Thailand B-3.6 Data Processing

Estimation of a cover-type change matrix from error-prone data

Treesearch

Steen Magnussen

2009-01-01

Coregistration and classification errors seriously compromise per-pixel estimates of land cover change. A more robust estimation of change is proposed in which adjacent pixels are grouped into 3x3 clusters and treated as a unit of observation. A complete change matrix is recovered in a two-step process. The diagonal elements of a change matrix are recovered from...

Retinal vessel segmentation using the 2-D Gabor wavelet and supervised classification.

PubMed

Soares, João V B; Leandro, Jorge J G; Cesar Júnior, Roberto M; Jelinek, Herbert F; Cree, Michael J

2006-09-01

We present a method for automated segmentation of the vasculature in retinal images. The method produces segmentations by classifying each image pixel as vessel or nonvessel, based on the pixel's feature vector. Feature vectors are composed of the pixel's intensity and two-dimensional Gabor wavelet transform responses taken at multiple scales. The Gabor wavelet is capable of tuning to specific frequencies, thus allowing noise filtering and vessel enhancement in a single step. We use a Bayesian classifier with class-conditional probability density functions (likelihoods) described as Gaussian mixtures, yielding a fast classification, while being able to model complex decision surfaces. The probability distributions are estimated based on a training set of labeled pixels obtained from manual segmentations. The method's performance is evaluated on publicly available DRIVE (Staal et al., 2004) and STARE (Hoover et al., 2000) databases of manually labeled images. On the DRIVE database, it achieves an area under the receiver operating characteristic curve of 0.9614, being slightly superior than that presented by state-of-the-art approaches. We are making our implementation available as open source MATLAB scripts for researchers interested in implementation details, evaluation, or development of methods.

A simple and effective method for filling gaps in Landsat ETM+ SLC-off images

USGS Publications Warehouse

Chen, Jin; Zhu, Xiaolin; Vogelmann, James E.; Gao, Feng; Jin, Suming

2011-01-01

The scan-line corrector (SLC) of the Landsat 7 Enhanced Thematic Mapper Plus (ETM+) sensor failed in 2003, resulting in about 22% of the pixels per scene not being scanned. The SLC failure has seriously limited the scientific applications of ETM+ data. While there have been a number of methods developed to fill in the data gaps, each method has shortcomings, especially for heterogeneous landscapes. Based on the assumption that the same-class neighboring pixels around the un-scanned pixels have similar spectral characteristics, and that these neighboring and un-scanned pixels exhibit similar patterns of spectral differences between dates, we developed a simple and effective method to interpolate the values of the pixels within the gaps. We refer to this method as the Neighborhood Similar Pixel Interpolator (NSPI). Simulated and actual SLC-off ETM+ images were used to assess the performance of the NSPI. Results indicate that NSPI can restore the value of un-scanned pixels very accurately, and that it works especially well in heterogeneous regions. In addition, it can work well even if there is a relatively long time interval or significant spectral changes between the input and target image. The filled images appear reasonably spatially continuous without obvious striping patterns. Supervised classification using the maximum likelihood algorithm was done on both gap-filled simulated SLC-off data and the original "gap free" data set, and it was found that classification results, including accuracies, were very comparable. This indicates that gap-filled products generated by NSPI will have relevance to the user community for various land cover applications. In addition, the simple principle and high computational efficiency of NSPI will enable processing large volumes of SLC-off ETM+ data.

Saliency-Guided Change Detection of Remotely Sensed Images Using Random Forest

NASA Astrophysics Data System (ADS)

Feng, W.; Sui, H.; Chen, X.

2018-04-01

Studies based on object-based image analysis (OBIA) representing the paradigm shift in change detection (CD) have achieved remarkable progress in the last decade. Their aim has been developing more intelligent interpretation analysis methods in the future. The prediction effect and performance stability of random forest (RF), as a new kind of machine learning algorithm, are better than many single predictors and integrated forecasting method. In this paper, we present a novel CD approach for high-resolution remote sensing images, which incorporates visual saliency and RF. First, highly homogeneous and compact image super-pixels are generated using super-pixel segmentation, and the optimal segmentation result is obtained through image superimposition and principal component analysis (PCA). Second, saliency detection is used to guide the search of interest regions in the initial difference image obtained via the improved robust change vector analysis (RCVA) algorithm. The salient regions within the difference image that correspond to the binarized saliency map are extracted, and the regions are subject to the fuzzy c-means (FCM) clustering to obtain the pixel-level pre-classification result, which can be used as a prerequisite for superpixel-based analysis. Third, on the basis of the optimal segmentation and pixel-level pre-classification results, different super-pixel change possibilities are calculated. Furthermore, the changed and unchanged super-pixels that serve as the training samples are automatically selected. The spectral features and Gabor features of each super-pixel are extracted. Finally, superpixel-based CD is implemented by applying RF based on these samples. Experimental results on Ziyuan 3 (ZY3) multi-spectral images show that the proposed method outperforms the compared methods in the accuracy of CD, and also confirm the feasibility and effectiveness of the proposed approach.

Orientation selectivity based structure for texture classification

NASA Astrophysics Data System (ADS)

Wu, Jinjian; Lin, Weisi; Shi, Guangming; Zhang, Yazhong; Lu, Liu

2014-10-01

Local structure, e.g., local binary pattern (LBP), is widely used in texture classification. However, LBP is too sensitive to disturbance. In this paper, we introduce a novel structure for texture classification. Researches on cognitive neuroscience indicate that the primary visual cortex presents remarkable orientation selectivity for visual information extraction. Inspired by this, we investigate the orientation similarities among neighbor pixels, and propose an orientation selectivity based pattern for local structure description. Experimental results on texture classification demonstrate that the proposed structure descriptor is quite robust to disturbance.

Use of Landsat-derived temporal profiles for corn-soybean feature extraction and classification

NASA Technical Reports Server (NTRS)

Badhwar, G. D.; Carnes, J. G.; Austin, W. W.

1982-01-01

A physical model is presented, which has been derived from multitemporal-multispectral data acquired by Landsat satellites to describe the behavior and new features that are crop specific. A feasibility study over 40 sites was performed to classify the segment pixels into those of corn, soybeans, and others using the new features and a linear classifier. Results agree well with other existing methods, and it is shown the multitemporal-multispectral scanner data can be transformed into two parameters that are closely related to the target of interest and thus can be used in classification. The approach is less time intensive than other techniques and requires labeling of only pure pixels.

Application of a neural network for reflectance spectrum classification

NASA Astrophysics Data System (ADS)

Yang, Gefei; Gartley, Michael

2017-05-01

Traditional reflectance spectrum classification algorithms are based on comparing spectrum across the electromagnetic spectrum anywhere from the ultra-violet to the thermal infrared regions. These methods analyze reflectance on a pixel by pixel basis. Inspired by high performance that Convolution Neural Networks (CNN) have demonstrated in image classification, we applied a neural network to analyze directional reflectance pattern images. By using the bidirectional reflectance distribution function (BRDF) data, we can reformulate the 4-dimensional into 2 dimensions, namely incident direction × reflected direction × channels. Meanwhile, RIT's micro-DIRSIG model is utilized to simulate additional training samples for improving the robustness of the neural networks training. Unlike traditional classification by using hand-designed feature extraction with a trainable classifier, neural networks create several layers to learn a feature hierarchy from pixels to classifier and all layers are trained jointly. Hence, the our approach of utilizing the angular features are different to traditional methods utilizing spatial features. Although training processing typically has a large computational cost, simple classifiers work well when subsequently using neural network generated features. Currently, most popular neural networks such as VGG, GoogLeNet and AlexNet are trained based on RGB spatial image data. Our approach aims to build a directional reflectance spectrum based neural network to help us to understand from another perspective. At the end of this paper, we compare the difference among several classifiers and analyze the trade-off among neural networks parameters.

Temporal expansion of annual crop classification layers for the CONUS using the C5 decision tree classifier

USGS Publications Warehouse

Friesz, Aaron M.; Wylie, Bruce K.; Howard, Daniel M.

2017-01-01

Crop cover maps have become widely used in a range of research applications. Multiple crop cover maps have been developed to suite particular research interests. The National Agricultural Statistics Service (NASS) Cropland Data Layers (CDL) are a series of commonly used crop cover maps for the conterminous United States (CONUS) that span from 2008 to 2013. In this investigation, we sought to contribute to the availability of consistent CONUS crop cover maps by extending temporal coverage of the NASS CDL archive back eight additional years to 2000 by creating annual NASS CDL-like crop cover maps derived from a classification tree model algorithm. We used over 11 million records to train a classification tree algorithm and develop a crop classification model (CCM). The model was used to create crop cover maps for the CONUS for years 2000–2013 at 250 m spatial resolution. The CCM and the maps for years 2008–2013 were assessed for accuracy relative to resampled NASS CDLs. The CCM performed well against a withheld test data set with a model prediction accuracy of over 90%. The assessment of the crop cover maps indicated that the model performed well spatially, placing crop cover pixels within their known domains; however, the model did show a bias towards the ‘Other’ crop cover class, which caused frequent misclassifications of pixels around the periphery of large crop cover patch clusters and of pixels that form small, sparsely dispersed crop cover patches.

Classification of breast cancer cytological specimen using convolutional neural network

NASA Astrophysics Data System (ADS)

Żejmo, Michał; Kowal, Marek; Korbicz, Józef; Monczak, Roman

2017-01-01

The paper presents a deep learning approach for automatic classification of breast tumors based on fine needle cytology. The main aim of the system is to distinguish benign from malignant cases based on microscopic images. Experiment was carried out on cytological samples derived from 50 patients (25 benign cases + 25 malignant cases) diagnosed in Regional Hospital in Zielona Góra. To classify microscopic images, we used convolutional neural networks (CNN) of two types: GoogLeNet and AlexNet. Due to the very large size of images of cytological specimen (on average 200000 × 100000 pixels), they were divided into smaller patches of size 256 × 256 pixels. Breast cancer classification usually is based on morphometric features of nuclei. Therefore, training and validation patches were selected using Support Vector Machine (SVM) so that suitable amount of cell material was depicted. Neural classifiers were tuned using GPU accelerated implementation of gradient descent algorithm. Training error was defined as a cross-entropy classification loss. Classification accuracy was defined as the percentage ratio of successfully classified validation patches to the total number of validation patches. The best accuracy rate of 83% was obtained by GoogLeNet model. We observed that more misclassified patches belong to malignant cases.

An assessment of commonly employed satellite-based remote sensors for mapping mangrove species in Mexico using an NDVI-based classification scheme.

PubMed

Valderrama-Landeros, L; Flores-de-Santiago, F; Kovacs, J M; Flores-Verdugo, F

2017-12-14

Optimizing the classification accuracy of a mangrove forest is of utmost importance for conservation practitioners. Mangrove forest mapping using satellite-based remote sensing techniques is by far the most common method of classification currently used given the logistical difficulties of field endeavors in these forested wetlands. However, there is now an abundance of options from which to choose in regards to satellite sensors, which has led to substantially different estimations of mangrove forest location and extent with particular concern for degraded systems. The objective of this study was to assess the accuracy of mangrove forest classification using different remotely sensed data sources (i.e., Landsat-8, SPOT-5, Sentinel-2, and WorldView-2) for a system located along the Pacific coast of Mexico. Specifically, we examined a stressed semiarid mangrove forest which offers a variety of conditions such as dead areas, degraded stands, healthy mangroves, and very dense mangrove island formations. The results indicated that Landsat-8 (30 m per pixel) had  the lowest overall accuracy at 64% and that WorldView-2 (1.6 m per pixel) had the highest at 93%. Moreover, the SPOT-5 and the Sentinel-2 classifications (10 m per pixel) were very similar having accuracies of 75 and 78%, respectively. In comparison to WorldView-2, the other sensors overestimated the extent of Laguncularia racemosa and underestimated the extent of Rhizophora mangle. When considering such type of sensors, the higher spatial resolution can be particularly important in mapping small mangrove islands that often occur in degraded mangrove systems.

Efficacy measures associated to a plantar pressure based classification system in diabetic foot medicine.

PubMed

Deschamps, Kevin; Matricali, Giovanni Arnoldo; Desmet, Dirk; Roosen, Philip; Keijsers, Noel; Nobels, Frank; Bruyninckx, Herman; Staes, Filip

2016-09-01

The concept of 'classification' has, similar to many other diseases, been found to be fundamental in the field of diabetic medicine. In the current study, we aimed at determining efficacy measures of a recently published plantar pressure based classification system. Technical efficacy of the classification system was investigated by applying a high resolution, pixel-level analysis on the normalized plantar pressure pedobarographic fields of the original experimental dataset consisting of 97 patients with diabetes and 33 persons without diabetes. Clinical efficacy was assessed by considering the occurence of foot ulcers at the plantar aspect of the forefoot in this dataset. Classification efficacy was assessed by determining the classification recognition rate as well as its sensitivity and specificity using cross-validation subsets of the experimental dataset together with a novel cohort of 12 patients with diabetes. Pixel-level comparison of the four groups associated to the classification system highlighted distinct regional differences. Retrospective analysis showed the occurence of eleven foot ulcers in the experimental dataset since their gait analysis. Eight out of the eleven ulcers developed in a region of the foot which had the highest forces. Overall classification recognition rate exceeded 90% for all cross-validation subsets. Sensitivity and specificity of the four groups associated to the classification system exceeded respectively the 0.7 and 0.8 level in all cross-validation subsets. The results of the current study support the use of the novel plantar pressure based classification system in diabetic foot medicine. It may particularly serve in communication, diagnosis and clinical decision making. Copyright © 2016 Elsevier B.V. All rights reserved.

Supervised pixel classification for segmenting geographic atrophy in fundus autofluorescene images

NASA Astrophysics Data System (ADS)

Hu, Zhihong; Medioni, Gerard G.; Hernandez, Matthias; Sadda, SriniVas R.

2014-03-01

Age-related macular degeneration (AMD) is the leading cause of blindness in people over the age of 65. Geographic atrophy (GA) is a manifestation of the advanced or late-stage of the AMD, which may result in severe vision loss and blindness. Techniques to rapidly and precisely detect and quantify GA lesions would appear to be of important value in advancing the understanding of the pathogenesis of GA and the management of GA progression. The purpose of this study is to develop an automated supervised pixel classification approach for segmenting GA including uni-focal and multi-focal patches in fundus autofluorescene (FAF) images. The image features include region wise intensity (mean and variance) measures, gray level co-occurrence matrix measures (angular second moment, entropy, and inverse difference moment), and Gaussian filter banks. A k-nearest-neighbor (k-NN) pixel classifier is applied to obtain a GA probability map, representing the likelihood that the image pixel belongs to GA. A voting binary iterative hole filling filter is then applied to fill in the small holes. Sixteen randomly chosen FAF images were obtained from sixteen subjects with GA. The algorithm-defined GA regions are compared with manual delineation performed by certified graders. Two-fold cross-validation is applied for the evaluation of the classification performance. The mean Dice similarity coefficients (DSC) between the algorithm- and manually-defined GA regions are 0.84 +/- 0.06 for one test and 0.83 +/- 0.07 for the other test and the area correlations between them are 0.99 (p < 0.05) and 0.94 (p < 0.05) respectively.

Will it Blend? Visualization and Accuracy Evaluation of High-Resolution Fuzzy Vegetation Maps

NASA Astrophysics Data System (ADS)

Zlinszky, A.; Kania, A.

2016-06-01

Instead of assigning every map pixel to a single class, fuzzy classification includes information on the class assigned to each pixel but also the certainty of this class and the alternative possible classes based on fuzzy set theory. The advantages of fuzzy classification for vegetation mapping are well recognized, but the accuracy and uncertainty of fuzzy maps cannot be directly quantified with indices developed for hard-boundary categorizations. The rich information in such a map is impossible to convey with a single map product or accuracy figure. Here we introduce a suite of evaluation indices and visualization products for fuzzy maps generated with ensemble classifiers. We also propose a way of evaluating classwise prediction certainty with "dominance profiles" visualizing the number of pixels in bins according to the probability of the dominant class, also showing the probability of all the other classes. Together, these data products allow a quantitative understanding of the rich information in a fuzzy raster map both for individual classes and in terms of variability in space, and also establish the connection between spatially explicit class certainty and traditional accuracy metrics. These map products are directly comparable to widely used hard boundary evaluation procedures, support active learning-based iterative classification and can be applied for operational use.

Comparing Pixel- and Object-Based Approaches in Effectively Classifying Wetland-Dominated Landscapes

PubMed Central

Berhane, Tedros M.; Lane, Charles R.; Wu, Qiusheng; Anenkhonov, Oleg A.; Chepinoga, Victor V.; Autrey, Bradley C.; Liu, Hongxing

2018-01-01

Wetland ecosystems straddle both terrestrial and aquatic habitats, performing many ecological functions directly and indirectly benefitting humans. However, global wetland losses are substantial. Satellite remote sensing and classification informs wise wetland management and monitoring. Both pixel- and object-based classification approaches using parametric and non-parametric algorithms may be effectively used in describing wetland structure and habitat, but which approach should one select? We conducted both pixel- and object-based image analyses (OBIA) using parametric (Iterative Self-Organizing Data Analysis Technique, ISODATA, and maximum likelihood, ML) and non-parametric (random forest, RF) approaches in the Barguzin Valley, a large wetland (~500 km2) in the Lake Baikal, Russia, drainage basin. Four Quickbird multispectral bands plus various spatial and spectral metrics (e.g., texture, Non-Differentiated Vegetation Index, slope, aspect, etc.) were analyzed using field-based regions of interest sampled to characterize an initial 18 ISODATA-based classes. Parsimoniously using a three-layer stack (Quickbird band 3, water ratio index (WRI), and mean texture) in the analyses resulted in the highest accuracy, 87.9% with pixel-based RF, followed by OBIA RF (segmentation scale 5, 84.6% overall accuracy), followed by pixel-based ML (83.9% overall accuracy). Increasing the predictors from three to five by adding Quickbird bands 2 and 4 decreased the pixel-based overall accuracy while increasing the OBIA RF accuracy to 90.4%. However, McNemar’s chi-square test confirmed no statistically significant difference in overall accuracy among the classifiers (pixel-based ML, RF, or object-based RF) for either the three- or five-layer analyses. Although potentially useful in some circumstances, the OBIA approach requires substantial resources and user input (such as segmentation scale selection—which was found to substantially affect overall accuracy). Hence, we conclude that pixel-based RF approaches are likely satisfactory for classifying wetland-dominated landscapes. PMID:29707381

Comparing Pixel- and Object-Based Approaches in Effectively Classifying Wetland-Dominated Landscapes.

PubMed

Berhane, Tedros M; Lane, Charles R; Wu, Qiusheng; Anenkhonov, Oleg A; Chepinoga, Victor V; Autrey, Bradley C; Liu, Hongxing

2018-01-01

Wetland ecosystems straddle both terrestrial and aquatic habitats, performing many ecological functions directly and indirectly benefitting humans. However, global wetland losses are substantial. Satellite remote sensing and classification informs wise wetland management and monitoring. Both pixel- and object-based classification approaches using parametric and non-parametric algorithms may be effectively used in describing wetland structure and habitat, but which approach should one select? We conducted both pixel- and object-based image analyses (OBIA) using parametric (Iterative Self-Organizing Data Analysis Technique, ISODATA, and maximum likelihood, ML) and non-parametric (random forest, RF) approaches in the Barguzin Valley, a large wetland (~500 km 2 ) in the Lake Baikal, Russia, drainage basin. Four Quickbird multispectral bands plus various spatial and spectral metrics (e.g., texture, Non-Differentiated Vegetation Index, slope, aspect, etc.) were analyzed using field-based regions of interest sampled to characterize an initial 18 ISODATA-based classes. Parsimoniously using a three-layer stack (Quickbird band 3, water ratio index (WRI), and mean texture) in the analyses resulted in the highest accuracy, 87.9% with pixel-based RF, followed by OBIA RF (segmentation scale 5, 84.6% overall accuracy), followed by pixel-based ML (83.9% overall accuracy). Increasing the predictors from three to five by adding Quickbird bands 2 and 4 decreased the pixel-based overall accuracy while increasing the OBIA RF accuracy to 90.4%. However, McNemar's chi-square test confirmed no statistically significant difference in overall accuracy among the classifiers (pixel-based ML, RF, or object-based RF) for either the three- or five-layer analyses. Although potentially useful in some circumstances, the OBIA approach requires substantial resources and user input (such as segmentation scale selection-which was found to substantially affect overall accuracy). Hence, we conclude that pixel-based RF approaches are likely satisfactory for classifying wetland-dominated landscapes.

Improving urban land use and land cover classification from high-spatial-resolution hyperspectral imagery using contextual information

USDA-ARS?s Scientific Manuscript database

In this paper, we propose approaches to improve the pixel-based support vector machine (SVM) classification for urban land use and land cover (LULC) mapping from airborne hyperspectral imagery with high spatial resolution. Class spatial neighborhood relationship is used to correct the misclassified ...

Enhanced Deforestation Mapping in North Korea using Spatial-temporal Image Fusion Method and Phenology-based Index

NASA Astrophysics Data System (ADS)

Jin, Y.; Lee, D.

2017-12-01

North Korea (the Democratic People's Republic of Korea, DPRK) is known to have some of the most degraded forest in the world. The characteristics of forest landscape in North Korea is complex and heterogeneous, the major vegetation cover types in the forest are hillside farm, unstocked forest, natural forest, and plateau vegetation. Better classification of types in high spatial resolution of deforested areas could provide essential information for decisions about forest management priorities and restoration of deforested areas. For mapping heterogeneous vegetation covers, the phenology-based indices are helpful to overcome the reflectance value confusion that occurs when using one season images. Coarse spatial resolution images may be acquired with a high repetition rate and it is useful for analyzing phenology characteristics, but may not capture the spatial detail of the land cover mosaic of the region of interest. Previous spatial-temporal fusion methods were only capture the temporal change, or focused on both temporal change and spatial change but with low accuracy in heterogeneous landscapes and small patches. In this study, a new concept for spatial-temporal image fusion method focus on heterogeneous landscape was proposed to produce fine resolution images at both fine spatial and temporal resolution. We classified the three types of pixels between the base image and target image, the first type is only reflectance changed caused by phenology, this type of pixels supply the reflectance, shape and texture information; the second type is both reflectance and spectrum changed in some bands caused by phenology like rice paddy or farmland, this type of pixels only supply shape and texture information; the third type is reflectance and spectrum changed caused by land cover type change, this type of pixels don't provide any information because we can't know how land cover changed in target image; and each type of pixels were applied different prediction methods. Results show that both STARFM and FSDAF predicted in low accuracy in second type pixels and small patches. Classification results used spatial-temporal image fusion method proposed in this study showed overall classification accuracy of 89.38%, with corresponding kappa coefficients of 0.87.

SU-E-I-59: Investigation of the Usefulness of a Standard Deviation and Mammary Gland Density as Indexes for Mammogram Classification.

PubMed

Takarabe, S; Yabuuchi, H; Morishita, J

2012-06-01

To investigate the usefulness of the standard deviation of pixel values in a whole mammary glands region and the percentage of a high- density mammary glands region to a whole mammary glands region as features for classification of mammograms into four categories based on the ACR BI-RADS breast composition. We used 36 digital mediolateral oblique view mammograms (18 patients) approved by our IRB. These images were classified into the four categories of breast compositions by an experienced breast radiologist and the results of the classification were regarded as a gold standard. First, a whole mammary region in a breast was divided into two regions such as a high-density mammary glands region and a low/iso-density mammary glands region by using a threshold value that was obtained from the pixel values corresponding to a pectoral muscle region. Then the percentage of a high-density mammary glands region to a whole mammary glands region was calculated. In addition, as a new method, the standard deviation of pixel values in a whole mammary glands region was calculated as an index based on the intermingling of mammary glands and fats. Finally, all mammograms were classified by using the combination of the percentage of a high-density mammary glands region and the standard deviation of each image. The agreement rates of the classification between our proposed method and gold standard was 86% (31/36). This result signified that our method has the potential to classify mammograms. The combination of the standard deviation of pixel values in a whole mammary glands region and the percentage of a high-density mammary glands region to a whole mammary glands region was available as features to classify mammograms based on the ACR BI- RADS breast composition. © 2012 American Association of Physicists in Medicine.

Exploring the limits of identifying sub-pixel thermal features using ASTER TIR data

USGS Publications Warehouse

Vaughan, R.G.; Keszthelyi, L.P.; Davies, A.G.; Schneider, D.J.; Jaworowski, C.; Heasler, H.

2010-01-01

Understanding the characteristics of volcanic thermal emissions and how they change with time is important for forecasting and monitoring volcanic activity and potential hazards. Satellite instruments view volcanic thermal features across the globe at various temporal and spatial resolutions. Thermal features that may be a precursor to a major eruption, or indicative of important changes in an on-going eruption can be subtle, making them challenging to reliably identify with satellite instruments. The goal of this study was to explore the limits of the types and magnitudes of thermal anomalies that could be detected using satellite thermal infrared (TIR) data. Specifically, the characterization of sub-pixel thermal features with a wide range of temperatures is considered using ASTER multispectral TIR data. First, theoretical calculations were made to define a "thermal mixing detection threshold" for ASTER, which quantifies the limits of ASTER's ability to resolve sub-pixel thermal mixing over a range of hot target temperatures and % pixel areas. Then, ASTER TIR data were used to model sub-pixel thermal features at the Yellowstone National Park geothermal area (hot spring pools with temperatures from 40 to 90 ??C) and at Mount Erebus Volcano, Antarctica (an active lava lake with temperatures from 200 to 800 ??C). Finally, various sources of uncertainty in sub-pixel thermal calculations were quantified for these empirical measurements, including pixel resampling, atmospheric correction, and background temperature and emissivity assumptions.

Vulnerable land ecosystems classification using spatial context and spectral indices

NASA Astrophysics Data System (ADS)

Ibarrola-Ulzurrun, Edurne; Gonzalo-Martín, Consuelo; Marcello, Javier

2017-10-01

Natural habitats are exposed to growing pressure due to intensification of land use and tourism development. Thus, obtaining information on the vegetation is necessary for conservation and management projects. In this context, remote sensing is an important tool for monitoring and managing habitats, being classification a crucial stage. The majority of image classifications techniques are based upon the pixel-based approach. An alternative is the object-based (OBIA) approach, in which a previous segmentation step merges image pixels to create objects that are then classified. Besides, improved results may be gained by incorporating additional spatial information and specific spectral indices into the classification process. The main goal of this work was to implement and assess object-based classification techniques on very-high resolution imagery incorporating spectral indices and contextual spatial information in the classification models. The study area was Teide National Park in Canary Islands (Spain) using Worldview-2 orthoready imagery. In the classification model, two common indices were selected Normalized Difference Vegetation Index (NDVI) and Optimized Soil Adjusted Vegetation Index (OSAVI), as well as two specific Worldview-2 sensor indices, Worldview Vegetation Index and Worldview Soil Index. To include the contextual information, Grey Level Co-occurrence Matrices (GLCM) were used. The classification was performed training a Support Vector Machine with sufficient and representative number of vegetation samples (Spartocytisus supranubius, Pterocephalus lasiospermus, Descurainia bourgaeana and Pinus canariensis) as well as urban, road and bare soil classes. Confusion Matrices were computed to evaluate the results from each classification model obtaining the highest overall accuracy (90.07%) combining both Worldview indices with the GLCM-dissimilarity.

New DTM Extraction Approach from Airborne Images Derived Dsm

NASA Astrophysics Data System (ADS)

Mousa, Y. A.; Helmholz, P.; Belton, D.

2017-05-01

In this work, a new filtering approach is proposed for a fully automatic Digital Terrain Model (DTM) extraction from very high resolution airborne images derived Digital Surface Models (DSMs). Our approach represents an enhancement of the existing DTM extraction algorithm Multi-directional and Slope Dependent (MSD) by proposing parameters that are more reliable for the selection of ground pixels and the pixelwise classification. To achieve this, four main steps are implemented: Firstly, 8 well-distributed scanlines are used to search for minima as a ground point within a pre-defined filtering window size. These selected ground points are stored with their positions on a 2D surface to create a network of ground points. Then, an initial DTM is created using an interpolation method to fill the gaps in the 2D surface. Afterwards, a pixel to pixel comparison between the initial DTM and the original DSM is performed utilising pixelwise classification of ground and non-ground pixels by applying a vertical height threshold. Finally, the pixels classified as non-ground are removed and the remaining holes are filled. The approach is evaluated using the Vaihingen benchmark dataset provided by the ISPRS working group III/4. The evaluation includes the comparison of our approach, denoted as Network of Ground Points (NGPs) algorithm, with the DTM created based on MSD as well as a reference DTM generated from LiDAR data. The results show that our proposed approach over performs the MSD approach.

Land cover mapping at sub-pixel scales

NASA Astrophysics Data System (ADS)

Makido, Yasuyo Kato

One of the biggest drawbacks of land cover mapping from remotely sensed images relates to spatial resolution, which determines the level of spatial details depicted in an image. Fine spatial resolution images from satellite sensors such as IKONOS and QuickBird are now available. However, these images are not suitable for large-area studies, since a single image is very small and therefore it is costly for large area studies. Much research has focused on attempting to extract land cover types at sub-pixel scale, and little research has been conducted concerning the spatial allocation of land cover types within a pixel. This study is devoted to the development of new algorithms for predicting land cover distribution using remote sensory imagery at sub-pixel level. The "pixel-swapping" optimization algorithm, which was proposed by Atkinson for predicting sub-pixel land cover distribution, is investigated in this study. Two limitations of this method, the arbitrary spatial range value and the arbitrary exponential model of spatial autocorrelation, are assessed. Various weighting functions, as alternatives to the exponential model, are evaluated in order to derive the optimum weighting function. Two different simulation models were employed to develop spatially autocorrelated binary class maps. In all tested models, Gaussian, Exponential, and IDW, the pixel swapping method improved classification accuracy compared with the initial random allocation of sub-pixels. However the results suggested that equal weight could be used to increase accuracy and sub-pixel spatial autocorrelation instead of using these more complex models of spatial structure. New algorithms for modeling the spatial distribution of multiple land cover classes at sub-pixel scales are developed and evaluated. Three methods are examined: sequential categorical swapping, simultaneous categorical swapping, and simulated annealing. These three methods are applied to classified Landsat ETM+ data that has been resampled to 210 meters. The result suggested that the simultaneous method can be considered as the optimum method in terms of accuracy performance and computation time. The case study employs remote sensing imagery at the following sites: tropical forests in Brazil and temperate multiple land mosaic in East China. Sub-areas for both sites are used to examine how the characteristics of the landscape affect the ability of the optimum technique. Three types of measurement: Moran's I, mean patch size (MPS), and patch size standard deviation (STDEV), are used to characterize the landscape. All results suggested that this technique could increase the classification accuracy more than traditional hard classification. The methods developed in this study can benefit researchers who employ coarse remote sensing imagery but are interested in detailed landscape information. In many cases, the satellite sensor that provides large spatial coverage has insufficient spatial detail to identify landscape patterns. Application of the super-resolution technique described in this dissertation could potentially solve this problem by providing detailed land cover predictions from the coarse resolution satellite sensor imagery.

Object-based delineation and classification of alluvial fans by application of mean-shift segmentation and support vector machines

NASA Astrophysics Data System (ADS)

Pipaud, Isabel; Lehmkuhl, Frank

2017-09-01

In the field of geomorphology, automated extraction and classification of landforms is one of the most active research areas. Until the late 2000s, this task has primarily been tackled using pixel-based approaches. As these methods consider pixels and pixel neighborhoods as the sole basic entities for analysis, they cannot account for the irregular boundaries of real-world objects. Object-based analysis frameworks emerging from the field of remote sensing have been proposed as an alternative approach, and were successfully applied in case studies falling in the domains of both general and specific geomorphology. In this context, the a-priori selection of scale parameters or bandwidths is crucial for the segmentation result, because inappropriate parametrization will either result in over-segmentation or insufficient segmentation. In this study, we describe a novel supervised method for delineation and classification of alluvial fans, and assess its applicability using a SRTM 1‧‧ DEM scene depicting a section of the north-eastern Mongolian Altai, located in northwest Mongolia. The approach is premised on the application of mean-shift segmentation and the use of a one-class support vector machine (SVM) for classification. To consider variability in terms of alluvial fan dimension and shape, segmentation is performed repeatedly for different weightings of the incorporated morphometric parameters as well as different segmentation bandwidths. The final classification layer is obtained by selecting, for each real-world object, the most appropriate segmentation result according to fuzzy membership values derived from the SVM classification. Our results show that mean-shift segmentation and SVM-based classification provide an effective framework for delineation and classification of a particular landform. Variable bandwidths and terrain parameter weightings were identified as being crucial for consideration of intra-class variability, and, in turn, for a constantly high segmentation quality. Our analysis further reveals that incorporation of morphometric parameters quantifying specific morphological aspects of a landform is indispensable for developing an accurate classification scheme. Alluvial fans exhibiting accentuated composite morphologies were identified as a major challenge for automatic delineation, as they cannot be fully captured by a single segmentation run. There is, however, a high probability that this shortcoming can be overcome by enhancing the presented approach with a routine merging fan sub-entities based on their spatial relationships.

Using texture analysis to improve per-pixel classification of very high resolution images for mapping plastic greenhouses

NASA Astrophysics Data System (ADS)

Agüera, Francisco; Aguilar, Fernando J.; Aguilar, Manuel A.

The area occupied by plastic-covered greenhouses has undergone rapid growth in recent years, currently exceeding 500,000 ha worldwide. Due to the vast amount of input (water, fertilisers, fuel, etc.) required, and output of different agricultural wastes (vegetable, plastic, chemical, etc.), the environmental impact of this type of production system can be serious if not accompanied by sound and sustainable territorial planning. For this, the new generation of satellites which provide very high resolution imagery, such as QuickBird and IKONOS can be useful. In this study, one QuickBird and one IKONOS satellite image have been used to cover the same area under similar circumstances. The aim of this work was an exhaustive comparison of QuickBird vs. IKONOS images in land-cover detection. In terms of plastic greenhouse mapping, comparative tests were designed and implemented, each with separate objectives. Firstly, the Maximum Likelihood Classification (MLC) was applied using five different approaches combining R, G, B, NIR, and panchromatic bands. The combinations of the bands used, significantly influenced some of the indexes used to classify quality in this work. Furthermore, the quality classification of the QuickBird image was higher in all cases than that of the IKONOS image. Secondly, texture features derived from the panchromatic images at different window sizes and with different grey levels were added as a fifth band to the R, G, B, NIR images to carry out the MLC. The inclusion of texture information in the classification did not improve the classification quality. For classifications with texture information, the best accuracies were found in both images for mean and angular second moment texture parameters. The optimum window size in these texture parameters was 3×3 for IK images, while for QB images it depended on the quality index studied, but the optimum window size was around 15×15. With regard to the grey level, the optimum was 128. Thus, the optimum texture parameter depended on the main objective of the image classification. If the main classification goal is to minimize the number of pixels wrongly classified, the mean texture parameter should be used, whereas if the main classification goal is to minimize the unclassified pixels the angular second moment texture parameter should be used. On the whole, both QuickBird and IKONOS images offered promising results in classifying plastic greenhouses.

Pixel-based flood mapping from SAR imagery: a comparison of approaches

NASA Astrophysics Data System (ADS)

Landuyt, Lisa; Van Wesemael, Alexandra; Van Coillie, Frieke M. B.; Verhoest, Niko E. C.

2017-04-01

Due to their all-weather, day and night capabilities, SAR sensors have been shown to be particularly suitable for flood mapping applications. Thus, they can provide spatially-distributed flood extent data which are valuable for calibrating, validating and updating flood inundation models. These models are an invaluable tool for water managers, to take appropriate measures in times of high water levels. Image analysis approaches to delineate flood extent on SAR imagery are numerous. They can be classified into two categories, i.e. pixel-based and object-based approaches. Pixel-based approaches, e.g. thresholding, are abundant and in general computationally inexpensive. However, large discrepancies between these techniques exist and often subjective user intervention is needed. Object-based approaches require more processing but allow for the integration of additional object characteristics, like contextual information and object geometry, and thus have significant potential to provide an improved classification result. As means of benchmark, a selection of pixel-based techniques is applied on a ERS-2 SAR image of the 2006 flood event of River Dee, United Kingdom. This selection comprises Otsu thresholding, Kittler & Illingworth thresholding, the Fine To Coarse segmentation algorithm and active contour modelling. The different classification results are evaluated and compared by means of several accuracy measures, including binary performance measures.

Automated segmentation of geographic atrophy in fundus autofluorescence images using supervised pixel classification.

PubMed

Hu, Zhihong; Medioni, Gerard G; Hernandez, Matthias; Sadda, Srinivas R

2015-01-01

Geographic atrophy (GA) is a manifestation of the advanced or late stage of age-related macular degeneration (AMD). AMD is the leading cause of blindness in people over the age of 65 in the western world. The purpose of this study is to develop a fully automated supervised pixel classification approach for segmenting GA, including uni- and multifocal patches in fundus autofluorescene (FAF) images. The image features include region-wise intensity measures, gray-level co-occurrence matrix measures, and Gaussian filter banks. A [Formula: see text]-nearest-neighbor pixel classifier is applied to obtain a GA probability map, representing the likelihood that the image pixel belongs to GA. Sixteen randomly chosen FAF images were obtained from 16 subjects with GA. The algorithm-defined GA regions are compared with manual delineation performed by a certified image reading center grader. Eight-fold cross-validation is applied to evaluate the algorithm performance. The mean overlap ratio (OR), area correlation (Pearson's [Formula: see text]), accuracy (ACC), true positive rate (TPR), specificity (SPC), positive predictive value (PPV), and false discovery rate (FDR) between the algorithm- and manually defined GA regions are [Formula: see text], [Formula: see text], [Formula: see text], [Formula: see text], [Formula: see text], [Formula: see text], and [Formula: see text], respectively.

Spotting East African mammals in open savannah from space.

PubMed

Yang, Zheng; Wang, Tiejun; Skidmore, Andrew K; de Leeuw, Jan; Said, Mohammed Y; Freer, Jim

2014-01-01

Knowledge of population dynamics is essential for managing and conserving wildlife. Traditional methods of counting wild animals such as aerial survey or ground counts not only disturb animals, but also can be labour intensive and costly. New, commercially available very high-resolution satellite images offer great potential for accurate estimates of animal abundance over large open areas. However, little research has been conducted in the area of satellite-aided wildlife census, although computer processing speeds and image analysis algorithms have vastly improved. This paper explores the possibility of detecting large animals in the open savannah of Maasai Mara National Reserve, Kenya from very high-resolution GeoEye-1 satellite images. A hybrid image classification method was employed for this specific purpose by incorporating the advantages of both pixel-based and object-based image classification approaches. This was performed in two steps: firstly, a pixel-based image classification method, i.e., artificial neural network was applied to classify potential targets with similar spectral reflectance at pixel level; and then an object-based image classification method was used to further differentiate animal targets from the surrounding landscapes through the applications of expert knowledge. As a result, the large animals in two pilot study areas were successfully detected with an average count error of 8.2%, omission error of 6.6% and commission error of 13.7%. The results of the study show for the first time that it is feasible to perform automated detection and counting of large wild animals in open savannahs from space, and therefore provide a complementary and alternative approach to the conventional wildlife survey techniques.

Benchmark of Machine Learning Methods for Classification of a SENTINEL-2 Image

NASA Astrophysics Data System (ADS)

Pirotti, F.; Sunar, F.; Piragnolo, M.

2016-06-01

Thanks to mainly ESA and USGS, a large bulk of free images of the Earth is readily available nowadays. One of the main goals of remote sensing is to label images according to a set of semantic categories, i.e. image classification. This is a very challenging issue since land cover of a specific class may present a large spatial and spectral variability and objects may appear at different scales and orientations. In this study, we report the results of benchmarking 9 machine learning algorithms tested for accuracy and speed in training and classification of land-cover classes in a Sentinel-2 dataset. The following machine learning methods (MLM) have been tested: linear discriminant analysis, k-nearest neighbour, random forests, support vector machines, multi layered perceptron, multi layered perceptron ensemble, ctree, boosting, logarithmic regression. The validation is carried out using a control dataset which consists of an independent classification in 11 land-cover classes of an area about 60 km2, obtained by manual visual interpretation of high resolution images (20 cm ground sampling distance) by experts. In this study five out of the eleven classes are used since the others have too few samples (pixels) for testing and validating subsets. The classes used are the following: (i) urban (ii) sowable areas (iii) water (iv) tree plantations (v) grasslands. Validation is carried out using three different approaches: (i) using pixels from the training dataset (train), (ii) using pixels from the training dataset and applying cross-validation with the k-fold method (kfold) and (iii) using all pixels from the control dataset. Five accuracy indices are calculated for the comparison between the values predicted with each model and control values over three sets of data: the training dataset (train), the whole control dataset (full) and with k-fold cross-validation (kfold) with ten folds. Results from validation of predictions of the whole dataset (full) show the random forests method with the highest values; kappa index ranging from 0.55 to 0.42 respectively with the most and least number pixels for training. The two neural networks (multi layered perceptron and its ensemble) and the support vector machines - with default radial basis function kernel - methods follow closely with comparable performance.

Thematic accuracy of the National Land Cover Database (NLCD) 2001 land cover for Alaska

USGS Publications Warehouse

Selkowitz, D.J.; Stehman, S.V.

2011-01-01

The National Land Cover Database (NLCD) 2001 Alaska land cover classification is the first 30-m resolution land cover product available covering the entire state of Alaska. The accuracy assessment of the NLCD 2001 Alaska land cover classification employed a geographically stratified three-stage sampling design to select the reference sample of pixels. Reference land cover class labels were determined via fixed wing aircraft, as the high resolution imagery used for determining the reference land cover classification in the conterminous U.S. was not available for most of Alaska. Overall thematic accuracy for the Alaska NLCD was 76.2% (s.e. 2.8%) at Level II (12 classes evaluated) and 83.9% (s.e. 2.1%) at Level I (6 classes evaluated) when agreement was defined as a match between the map class and either the primary or alternate reference class label. When agreement was defined as a match between the map class and primary reference label only, overall accuracy was 59.4% at Level II and 69.3% at Level I. The majority of classification errors occurred at Level I of the classification hierarchy (i.e., misclassifications were generally to a different Level I class, not to a Level II class within the same Level I class). Classification accuracy was higher for more abundant land cover classes and for pixels located in the interior of homogeneous land cover patches. ?? 2011.

A Matlab Program for Textural Classification Using Neural Networks

NASA Astrophysics Data System (ADS)

Leite, E. P.; de Souza, C.

2008-12-01

A new MATLAB code that provides tools to perform classification of textural images for applications in the Geosciences is presented. The program, here coined TEXTNN, comprises the computation of variogram maps in the frequency domain for specific lag distances in the neighborhood of a pixel. The result is then converted back to spatial domain, where directional or ominidirectional semivariograms are extracted. Feature vectors are built with textural information composed of the semivariance values at these lag distances and, moreover, with histogram measures of mean, standard deviation and weighted fill-ratio. This procedure is applied to a selected group of pixels or to all pixels in an image using a moving window. A feed- forward back-propagation Neural Network can then be designed and trained on feature vectors of predefined classes (training set). The training phase minimizes the mean-squared error on the training set. Additionally, at each iteration, the mean-squared error for every validation is assessed and a test set is evaluated. The program also calculates contingency matrices, global accuracy and kappa coefficient for the three data sets, allowing a quantitative appraisal of the predictive power of the Neural Network models. The interpreter is able to select the best model obtained from a k-fold cross-validation or to use a unique split-sample data set for classification of all pixels in a given textural image. The code is opened to the geoscientific community and is very flexible, allowing the experienced user to modify it as necessary. The performance of the algorithms and the end-user program were tested using synthetic images, orbital SAR (RADARSAT) imagery for oil seepage detection, and airborne, multi-polarimetric SAR imagery for geologic mapping. The overall results proved very promising.

VizieR Online Data Catalog: SDSS-DR8 galaxies classified by WND-CHARM (Kuminski+, 2016)

NASA Astrophysics Data System (ADS)

Kuminski, E.; Shamir, L.

2016-06-01

The image analysis method used to classify the images is WND-CHARM (wndchrm; Shamir et al. 2008, BMC Source Code for Biology and Medicine, 3: 13; 2010PLSCB...6E0974S; 2013ascl.soft12002S), which first computes 2885 numerical descriptors from each SDSS image such as textures, edges, shapes), the statistical distribution of the pixel intensities, the polynomial decomposition of the image, and fractal features. These features are extracted from the raw pixels, as well as the image transforms and multi-order image transforms. See section 2 for further explanations. In a similar way than the catalog we also compiled a catalog of all objects with spectra in DR8. For each object, that catalog contains the spec ObjID, the R.A., the decl., the z, z error, the certainty of classification as elliptical, the certainty of classification as spiral, and the certainty of classification as a star. See section 3.1 for further explanations. (2 data files).

Multiresolution texture analysis applied to road surface inspection

NASA Astrophysics Data System (ADS)

Paquis, Stephane; Legeay, Vincent; Konik, Hubert; Charrier, Jean

1999-03-01

Technological advances provide now the opportunity to automate the pavement distress assessment. This paper deals with an approach for achieving an automatic vision system for road surface classification. Road surfaces are composed of aggregates, which have a particular grain size distribution and a mortar matrix. From various physical properties and visual aspects, four road families are generated. We present here a tool using a pyramidal process with the assumption that regions or objects in an image rise up because of their uniform texture. Note that the aim is not to compute another statistical parameter but to include usual criteria in our method. In fact, the road surface classification uses a multiresolution cooccurrence matrix and a hierarchical process through an original intensity pyramid, where a father pixel takes the minimum gray level value of its directly linked children pixels. More precisely, only matrix diagonal is taken into account and analyzed along the pyramidal structure, which allows the classification to be made.

Spectral-Spatial Shared Linear Regression for Hyperspectral Image Classification.

PubMed

Haoliang Yuan; Yuan Yan Tang

2017-04-01

Classification of the pixels in hyperspectral image (HSI) is an important task and has been popularly applied in many practical applications. Its major challenge is the high-dimensional small-sized problem. To deal with this problem, lots of subspace learning (SL) methods are developed to reduce the dimension of the pixels while preserving the important discriminant information. Motivated by ridge linear regression (RLR) framework for SL, we propose a spectral-spatial shared linear regression method (SSSLR) for extracting the feature representation. Comparing with RLR, our proposed SSSLR has the following two advantages. First, we utilize a convex set to explore the spatial structure for computing the linear projection matrix. Second, we utilize a shared structure learning model, which is formed by original data space and a hidden feature space, to learn a more discriminant linear projection matrix for classification. To optimize our proposed method, an efficient iterative algorithm is proposed. Experimental results on two popular HSI data sets, i.e., Indian Pines and Salinas demonstrate that our proposed methods outperform many SL methods.

Alignment, orientation, and Coulomb explosion of difluoroiodobenzene studied with the pixel imaging mass spectrometry (PImMS) camera.

PubMed

Amini, Kasra; Boll, Rebecca; Lauer, Alexandra; Burt, Michael; Lee, Jason W L; Christensen, Lauge; Brauβe, Felix; Mullins, Terence; Savelyev, Evgeny; Ablikim, Utuq; Berrah, Nora; Bomme, Cédric; Düsterer, Stefan; Erk, Benjamin; Höppner, Hauke; Johnsson, Per; Kierspel, Thomas; Krecinic, Faruk; Küpper, Jochen; Müller, Maria; Müller, Erland; Redlin, Harald; Rouzée, Arnaud; Schirmel, Nora; Thøgersen, Jan; Techert, Simone; Toleikis, Sven; Treusch, Rolf; Trippel, Sebastian; Ulmer, Anatoli; Wiese, Joss; Vallance, Claire; Rudenko, Artem; Stapelfeldt, Henrik; Brouard, Mark; Rolles, Daniel

2017-07-07

Laser-induced adiabatic alignment and mixed-field orientation of 2,6-difluoroiodobenzene (C 6 H 3 F 2 I) molecules are probed by Coulomb explosion imaging following either near-infrared strong-field ionization or extreme-ultraviolet multi-photon inner-shell ionization using free-electron laser pulses. The resulting photoelectrons and fragment ions are captured by a double-sided velocity map imaging spectrometer and projected onto two position-sensitive detectors. The ion side of the spectrometer is equipped with a pixel imaging mass spectrometry camera, a time-stamping pixelated detector that can record the hit positions and arrival times of up to four ions per pixel per acquisition cycle. Thus, the time-of-flight trace and ion momentum distributions for all fragments can be recorded simultaneously. We show that we can obtain a high degree of one-and three-dimensional alignment and mixed-field orientation and compare the Coulomb explosion process induced at both wavelengths.

Unmixing AVHRR Imagery to Assess Clearcuts and Forest Regrowth in Oregon

NASA Technical Reports Server (NTRS)

Hlavka, Christine A.; Spanner, Michael A.

1995-01-01

Advanced Very High Resolution Radiometer imagery provides frequent and low-cost coverage of the earth, but its coarse spatial resolution (approx. 1.1 km by 1.1 km) does not lend itself to standard techniques of automated categorization of land cover classes because the pixels are generally mixed; that is, the extent of the pixel includes several land use/cover classes. Unmixing procedures were developed to extract land use/cover class signatures from mixed pixels, using Landsat Thematic Mapper data as a source for the training set, and to estimate fractions of class coverage within pixels. Application of these unmixing procedures to mapping forest clearcuts and regrowth in Oregon indicated that unmixing is a promising approach for mapping major trends in land cover with AVHRR bands 1 and 2. Including thermal bands by unmixing AVHRR bands 1-4 did not lead to significant improvements in accuracy, but experiments with unmixing these four bands did indicate that use of weighted least squares techniques might lead to improvements in other applications of unmixing.

Lagrange constraint neural networks for massive pixel parallel image demixing

NASA Astrophysics Data System (ADS)

Szu, Harold H.; Hsu, Charles C.

2002-03-01

We have shown that the remote sensing optical imaging to achieve detailed sub-pixel decomposition is a unique application of blind source separation (BSS) that is truly linear of far away weak signal, instantaneous speed of light without delay, and along the line of sight without multiple paths. In early papers, we have presented a direct application of statistical mechanical de-mixing method called Lagrange Constraint Neural Network (LCNN). While the BSAO algorithm (using a posteriori MaxEnt ANN and neighborhood pixel average) is not acceptable for remote sensing, a mirror symmetric LCNN approach is all right assuming a priori MaxEnt for unknown sources to be averaged over the source statistics (not neighborhood pixel data) in a pixel-by-pixel independent fashion. LCNN reduces the computation complexity, save a great number of memory devices, and cut the cost of implementation. The Landsat system is designed to measure the radiation to deduce surface conditions and materials. For any given material, the amount of emitted and reflected radiation varies by the wavelength. In practice, a single pixel of a Landsat image has seven channels receiving 0.1 to 12 microns of radiation from the ground within a 20x20 meter footprint containing a variety of radiation materials. A-priori LCNN algorithm provides the spatial-temporal variation of mixture that is hardly de-mixable by other a-posteriori BSS or ICA methods. We have already compared the Landsat remote sensing using both methods in WCCI 2002 Hawaii. Unfortunately the absolute benchmark is not possible because of lacking of the ground truth. We will arbitrarily mix two incoherent sampled images as the ground truth. However, the constant total probability of co-located sources within the pixel footprint is necessary for the remote sensing constraint (since on a clear day the total reflecting energy is constant in neighborhood receiving pixel sensors), we have to normalized two image pixel-by-pixel as well. Then, the result is indeed as expected.

Threshold selection for classification of MR brain images by clustering method

NASA Astrophysics Data System (ADS)

Moldovanu, Simona; Obreja, Cristian; Moraru, Luminita

2015-12-01

Given a grey-intensity image, our method detects the optimal threshold for a suitable binarization of MR brain images. In MR brain image processing, the grey levels of pixels belonging to the object are not substantially different from the grey levels belonging to the background. Threshold optimization is an effective tool to separate objects from the background and further, in classification applications. This paper gives a detailed investigation on the selection of thresholds. Our method does not use the well-known method for binarization. Instead, we perform a simple threshold optimization which, in turn, will allow the best classification of the analyzed images into healthy and multiple sclerosis disease. The dissimilarity (or the distance between classes) has been established using the clustering method based on dendrograms. We tested our method using two classes of images: the first consists of 20 T2-weighted and 20 proton density PD-weighted scans from two healthy subjects and from two patients with multiple sclerosis. For each image and for each threshold, the number of the white pixels (or the area of white objects in binary image) has been determined. These pixel numbers represent the objects in clustering operation. The following optimum threshold values are obtained, T = 80 for PD images and T = 30 for T2w images. Each mentioned threshold separate clearly the clusters that belonging of the studied groups, healthy patient and multiple sclerosis disease.

Application of Bayesian Classification to Content-Based Data Management

NASA Technical Reports Server (NTRS)

Lynnes, Christopher; Berrick, S.; Gopalan, A.; Hua, X.; Shen, S.; Smith, P.; Yang, K-Y.; Wheeler, K.; Curry, C.

2004-01-01

The high volume of Earth Observing System data has proven to be challenging to manage for data centers and users alike. At the Goddard Earth Sciences Distributed Active Archive Center (GES DAAC), about 1 TB of new data are archived each day. Distribution to users is also about 1 TB/day. A substantial portion of this distribution is MODIS calibrated radiance data, which has a wide variety of uses. However, much of the data is not useful for a particular user's needs: for example, ocean color users typically need oceanic pixels that are free of cloud and sun-glint. The GES DAAC is using a simple Bayesian classification scheme to rapidly classify each pixel in the scene in order to support several experimental content-based data services for near-real-time MODIS calibrated radiance products (from Direct Readout stations). Content-based subsetting would allow distribution of, say, only clear pixels to the user if desired. Content-based subscriptions would distribute data to users only when they fit the user's usability criteria in their area of interest within the scene. Content-based cache management would retain more useful data on disk for easy online access. The classification may even be exploited in an automated quality assessment of the geolocation product. Though initially to be demonstrated at the GES DAAC, these techniques have applicability in other resource-limited environments, such as spaceborne data systems.

Satellite mapping of Nile Delta coastal changes

NASA Technical Reports Server (NTRS)

Blodget, H. W.; Taylor, P. T.; Roark, J. H.

1989-01-01

Multitemporal, multispectral scanner (MSS) landsat data have been used to monitor erosion and sedimentation along the Rosetta Promontory of the Nile Delta. These processes have accelerated significantly since the completion of the Aswan High Dam in 1964. Digital differencing of four MSS data sets, using standard algorithms, show that changes observed over a single year period generally occur as strings of single mixed pixels along the coast. Therefore, these can only be used qualitatively to indicate areas where changes occur. Areas of change recorded over a multi-year period are generally larger and thus identified by clusters of pixels; this reduces errors introduced by mixed pixels. Satellites provide a synoptic perspective utilizing data acquired at frequent time intervals. This permits multiple year monitoring of delta evolution on a regional scale.

A study of some nine-element decision rules. [for multispectral recognition of remote sensing

NASA Technical Reports Server (NTRS)

Richardson, W.

1974-01-01

A nine-element rule is one that makes a classification decision for each pixel based on data from that pixel and its eight immediate neighbors. Three such rules, all fast and simple to use, are defined and tested. All performed substantially better on field interiors than the best one-point rule. Qualitative results indicate that fine detail and contradictory testimony tend to be overlooked by the rules.

Small-angle solution scattering using the mixed-mode pixel array detector.

PubMed

Koerner, Lucas J; Gillilan, Richard E; Green, Katherine S; Wang, Suntao; Gruner, Sol M

2011-03-01

Solution small-angle X-ray scattering (SAXS) measurements were obtained using a 128 × 128 pixel X-ray mixed-mode pixel array detector (MMPAD) with an 860 µs readout time. The MMPAD offers advantages for SAXS experiments: a pixel full-well of >2 × 10(7) 10 keV X-rays, a maximum flux rate of 10(8) X-rays pixel(-1) s(-1), and a sub-pixel point-spread function. Data from the MMPAD were quantitatively compared with data from a charge-coupled device (CCD) fiber-optically coupled to a phosphor screen. MMPAD solution SAXS data from lysozyme solutions were of equal or better quality than data captured by the CCD. The read-noise (normalized by pixel area) of the MMPAD was less than that of the CCD by an average factor of 3.0. Short sample-to-detector distances were required owing to the small MMPAD area (19.2 mm × 19.2 mm), and were revealed to be advantageous with respect to detector read-noise. As predicted by the Shannon sampling theory and confirmed by the acquisition of lysozyme solution SAXS curves, the MMPAD at short distances is capable of sufficiently sampling a solution SAXS curve for protein shape analysis. The readout speed of the MMPAD was demonstrated by continuously monitoring lysozyme sample evolution as radiation damage accumulated. These experiments prove that a small suitably configured MMPAD is appropriate for time-resolved solution scattering measurements.

Assessment of land use and land cover change using spatiotemporal analysis of landscape: case study in south of Tehran.

PubMed

Sabr, Abutaleb; Moeinaddini, Mazaher; Azarnivand, Hossein; Guinot, Benjamin

2016-12-01

In the recent years, dust storms originating from local abandoned agricultural lands have increasingly impacted Tehran and Karaj air quality. Designing and implementing mitigation plans are necessary to study land use/land cover change (LUCC). Land use/cover classification is particularly relevant in arid areas. This study aimed to map land use/cover by pixel- and object-based image classification methods, analyse landscape fragmentation and determine the effects of two different classification methods on landscape metrics. The same sets of ground data were used for both classification methods. Because accuracy of classification plays a key role in better understanding LUCC, both methods were employed. Land use/cover maps of the southwest area of Tehran city for the years 1985, 2000 and 2014 were obtained from Landsat digital images and classified into three categories: built-up, agricultural and barren lands. The results of our LUCC analysis showed that the most important changes in built-up agricultural land categories were observed in zone B (Shahriar, Robat Karim and Eslamshahr) between 1985 and 2014. The landscape metrics obtained for all categories pictured high landscape fragmentation in the study area. Despite no significant difference was evidenced between the two classification methods, the object-based classification led to an overall higher accuracy than using the pixel-based classification. In particular, the accuracy of the built-up category showed a marked increase. In addition, both methods showed similar trends in fragmentation metrics. One of the reasons is that the object-based classification is able to identify buildings, impervious surface and roads in dense urban areas, which produced more accurate maps.

Rapid classification of landsat TM imagery for phase 1 stratification using the automated NDVI threshold supervised classification (ANTSC) methodology

Treesearch

William H. Cooke; Dennis M. Jacobs

2002-01-01

FIA annual inventories require rapid updating of pixel-based Phase 1 estimates. Scientists at the Southern Research Station are developing an automated methodology that uses a Normalized Difference Vegetation Index (NDVI) for identifying and eliminating problem FIA plots from the analysis. Problem plots are those that have questionable land useiland cover information....

Can a Forest/Nonforest Change Map Improve the Precision of Forest Area, Volume, Growth, Removals, and Mortality Estimates?

Treesearch

Dale D. Gormanson; Mark H. Hansen; Ronald E. McRoberts

2005-01-01

In an extensive forest inventory, stratifications that use dual-date forest/nonforest classifications of Landsat Thematic Mapper data approximately 10 years apart are tested against similar classifications that use data from only one date. Alternative stratifications that further define edge strata as pixels adjacent to a forest/nonforest boundary are included in the...

Going Deeper With Contextual CNN for Hyperspectral Image Classification.

PubMed

Lee, Hyungtae; Kwon, Heesung

2017-10-01

In this paper, we describe a novel deep convolutional neural network (CNN) that is deeper and wider than other existing deep networks for hyperspectral image classification. Unlike current state-of-the-art approaches in CNN-based hyperspectral image classification, the proposed network, called contextual deep CNN, can optimally explore local contextual interactions by jointly exploiting local spatio-spectral relationships of neighboring individual pixel vectors. The joint exploitation of the spatio-spectral information is achieved by a multi-scale convolutional filter bank used as an initial component of the proposed CNN pipeline. The initial spatial and spectral feature maps obtained from the multi-scale filter bank are then combined together to form a joint spatio-spectral feature map. The joint feature map representing rich spectral and spatial properties of the hyperspectral image is then fed through a fully convolutional network that eventually predicts the corresponding label of each pixel vector. The proposed approach is tested on three benchmark data sets: the Indian Pines data set, the Salinas data set, and the University of Pavia data set. Performance comparison shows enhanced classification performance of the proposed approach over the current state-of-the-art on the three data sets.

Comparison of GOES Cloud Classification Algorithms Employing Explicit and Implicit Physics

NASA Technical Reports Server (NTRS)

Bankert, Richard L.; Mitrescu, Cristian; Miller, Steven D.; Wade, Robert H.

2009-01-01

Cloud-type classification based on multispectral satellite imagery data has been widely researched and demonstrated to be useful for distinguishing a variety of classes using a wide range of methods. The research described here is a comparison of the classifier output from two very different algorithms applied to Geostationary Operational Environmental Satellite (GOES) data over the course of one year. The first algorithm employs spectral channel thresholding and additional physically based tests. The second algorithm was developed through a supervised learning method with characteristic features of expertly labeled image samples used as training data for a 1-nearest-neighbor classification. The latter's ability to identify classes is also based in physics, but those relationships are embedded implicitly within the algorithm. A pixel-to-pixel comparison analysis was done for hourly daytime scenes within a region in the northeastern Pacific Ocean. Considerable agreement was found in this analysis, with many of the mismatches or disagreements providing insight to the strengths and limitations of each classifier. Depending upon user needs, a rule-based or other postprocessing system that combines the output from the two algorithms could provide the most reliable cloud-type classification.

Automated training site selection for large-area remote-sensing image analysis

NASA Astrophysics Data System (ADS)

McCaffrey, Thomas M.; Franklin, Steven E.

1993-11-01

A computer program is presented to select training sites automatically from remotely sensed digital imagery. The basic ideas are to guide the image analyst through the process of selecting typical and representative areas for large-area image classifications by minimizing bias, and to provide an initial list of potential classes for which training sites are required to develop a classification scheme or to verify classification accuracy. Reducing subjectivity in training site selection is achieved by using a purely statistical selection of homogeneous sites which then can be compared to field knowledge, aerial photography, or other remote-sensing imagery and ancillary data to arrive at a final selection of sites to be used to train the classification decision rules. The selection of the homogeneous sites uses simple tests based on the coefficient of variance, the F-statistic, and the Student's i-statistic. Comparisons of site means are conducted with a linear growing list of previously located homogeneous pixels. The program supports a common pixel-interleaved digital image format and has been tested on aerial and satellite optical imagery. The program is coded efficiently in the C programming language and was developed under AIX-Unix on an IBM RISC 6000 24-bit color workstation.

The least-squares mixing models to generate fraction images derived from remote sensing multispectral data

NASA Technical Reports Server (NTRS)

Shimabukuro, Yosio Edemir; Smith, James A.

1991-01-01

Constrained-least-squares and weighted-least-squares mixing models for generating fraction images derived from remote sensing multispectral data are presented. An experiment considering three components within the pixels-eucalyptus, soil (understory), and shade-was performed. The generated fraction images for shade (shade image) derived from these two methods were compared by considering the performance and computer time. The derived shade images are related to the observed variation in forest structure, i.e., the fraction of inferred shade in the pixel is related to different eucalyptus ages.

Wide field imaging - I. Applications of neural networks to object detection and star/galaxy classification

NASA Astrophysics Data System (ADS)

Andreon, S.; Gargiulo, G.; Longo, G.; Tagliaferri, R.; Capuano, N.

2000-12-01

Astronomical wide-field imaging performed with new large-format CCD detectors poses data reduction problems of unprecedented scale, which are difficult to deal with using traditional interactive tools. We present here NExt (Neural Extractor), a new neural network (NN) based package capable of detecting objects and performing both deblending and star/galaxy classification in an automatic way. Traditionally, in astronomical images, objects are first distinguished from the noisy background by searching for sets of connected pixels having brightnesses above a given threshold; they are then classified as stars or as galaxies through diagnostic diagrams having variables chosen according to the astronomer's taste and experience. In the extraction step, assuming that images are well sampled, NExt requires only the simplest a priori definition of `what an object is' (i.e. it keeps all structures composed of more than one pixel) and performs the detection via an unsupervised NN, approaching detection as a clustering problem that has been thoroughly studied in the artificial intelligence literature. The first part of the NExt procedure consists of an optimal compression of the redundant information contained in the pixels via a mapping from pixel intensities to a subspace individualized through principal component analysis. At magnitudes fainter than the completeness limit, stars are usually almost indistinguishable from galaxies, and therefore the parameters characterizing the two classes do not lie in disconnected subspaces, thus preventing the use of unsupervised methods. We therefore adopted a supervised NN (i.e. a NN that first finds the rules to classify objects from examples and then applies them to the whole data set). In practice, each object is classified depending on its membership of the regions mapping the input feature space in the training set. In order to obtain an objective and reliable classification, instead of using an arbitrarily defined set of features we use a NN to select the most significant features among the large number of measured ones, and then we use these selected features to perform the classification task. In order to optimize the performance of the system, we implemented and tested several different models of NN. The comparison of the NExt performance with that of the best detection and classification package known to the authors (SExtractor) shows that NExt is at least as effective as the best traditional packages.

Accounting for sub-pixel variability of clouds and/or unresolved spectral variability, as needed, with generalized radiative transfer theory

DOE PAGES

Davis, Anthony B.; Xu, Feng; Collins, William D.

2015-03-01

Atmospheric hyperspectral VNIR sensing struggles with sub-pixel variability of clouds and limited spectral resolution mixing molecular lines. Our generalized radiative transfer model addresses both issues with new propagation kernels characterized by power-law decay in space.

Operational Tree Species Mapping in a Diverse Tropical Forest with Airborne Imaging Spectroscopy.

PubMed

Baldeck, Claire A; Asner, Gregory P; Martin, Robin E; Anderson, Christopher B; Knapp, David E; Kellner, James R; Wright, S Joseph

2015-01-01

Remote identification and mapping of canopy tree species can contribute valuable information towards our understanding of ecosystem biodiversity and function over large spatial scales. However, the extreme challenges posed by highly diverse, closed-canopy tropical forests have prevented automated remote species mapping of non-flowering tree crowns in these ecosystems. We set out to identify individuals of three focal canopy tree species amongst a diverse background of tree and liana species on Barro Colorado Island, Panama, using airborne imaging spectroscopy data. First, we compared two leading single-class classification methods--binary support vector machine (SVM) and biased SVM--for their performance in identifying pixels of a single focal species. From this comparison we determined that biased SVM was more precise and created a multi-species classification model by combining the three biased SVM models. This model was applied to the imagery to identify pixels belonging to the three focal species and the prediction results were then processed to create a map of focal species crown objects. Crown-level cross-validation of the training data indicated that the multi-species classification model had pixel-level producer's accuracies of 94-97% for the three focal species, and field validation of the predicted crown objects indicated that these had user's accuracies of 94-100%. Our results demonstrate the ability of high spatial and spectral resolution remote sensing to accurately detect non-flowering crowns of focal species within a diverse tropical forest. We attribute the success of our model to recent classification and mapping techniques adapted to species detection in diverse closed-canopy forests, which can pave the way for remote species mapping in a wider variety of ecosystems.

Operational Tree Species Mapping in a Diverse Tropical Forest with Airborne Imaging Spectroscopy

PubMed Central

Baldeck, Claire A.; Asner, Gregory P.; Martin, Robin E.; Anderson, Christopher B.; Knapp, David E.; Kellner, James R.; Wright, S. Joseph

2015-01-01

Remote identification and mapping of canopy tree species can contribute valuable information towards our understanding of ecosystem biodiversity and function over large spatial scales. However, the extreme challenges posed by highly diverse, closed-canopy tropical forests have prevented automated remote species mapping of non-flowering tree crowns in these ecosystems. We set out to identify individuals of three focal canopy tree species amongst a diverse background of tree and liana species on Barro Colorado Island, Panama, using airborne imaging spectroscopy data. First, we compared two leading single-class classification methods—binary support vector machine (SVM) and biased SVM—for their performance in identifying pixels of a single focal species. From this comparison we determined that biased SVM was more precise and created a multi-species classification model by combining the three biased SVM models. This model was applied to the imagery to identify pixels belonging to the three focal species and the prediction results were then processed to create a map of focal species crown objects. Crown-level cross-validation of the training data indicated that the multi-species classification model had pixel-level producer’s accuracies of 94–97% for the three focal species, and field validation of the predicted crown objects indicated that these had user’s accuracies of 94–100%. Our results demonstrate the ability of high spatial and spectral resolution remote sensing to accurately detect non-flowering crowns of focal species within a diverse tropical forest. We attribute the success of our model to recent classification and mapping techniques adapted to species detection in diverse closed-canopy forests, which can pave the way for remote species mapping in a wider variety of ecosystems. PMID:26153693

Design of the low area monotonic trim DAC in 40 nm CMOS technology for pixel readout chips

NASA Astrophysics Data System (ADS)

Drozd, A.; Szczygiel, R.; Maj, P.; Satlawa, T.; Grybos, P.

2014-12-01

The recent research in hybrid pixel detectors working in single photon counting mode focuses on nanometer or 3D technologies which allow making pixels smaller and implementing more complex solutions in each of the pixels. Usually single pixel in readout electronics for X-ray detection comprises of charge amplifier, shaper and discriminator that allow classification of events occurring at the detector as true or false hits by comparing amplitude of the signal obtained with threshold voltage, which minimizes the influence of noise effects. However, making the pixel size smaller often causes problems with pixel to pixel uniformity and additional effects like charge sharing become more visible. To improve channel-to-channel uniformity or implement an algorithm for charge sharing effect minimization, small area trimming DACs working in each pixel independently are necessary. However, meeting the requirement of small area often results in poor linearity and even non-monotonicity. In this paper we present a novel low-area thermometer coded 6-bit DAC implemented in 40 nm CMOS technology. Monte Carlo simulations were performed on the described design proving that under all conditions designed DAC is inherently monotonic. Presented DAC was implemented in the prototype readout chip with 432 pixels working in single photon counting mode, with two trimming DACs in each pixel. Each DAC occupies the area of 8 μm × 18.5 μm. Measurements and chips' tests were performed to obtain reliable statistical results.

Use of multispectral satellite remote sensing to assess mixing of suspended sediment downstream of large river confluences

NASA Astrophysics Data System (ADS)

Umar, M.; Rhoads, Bruce L.; Greenberg, Jonathan A.

2018-01-01

Although past work has noted that contrasts in turbidity often are detectable on remotely sensed images of rivers downstream from confluences, no systematic methodology has been developed for assessing mixing over distance of confluent flows with differing surficial suspended sediment concentrations (SSSC). In contrast to field measurements of mixing below confluences, satellite remote-sensing can provide detailed information on spatial distributions of SSSC over long distances. This paper presents a methodology that uses remote-sensing data to estimate spatial patterns of SSSC downstream of confluences along large rivers and to determine changes in the amount of mixing over distance from confluences. The method develops a calibrated Random Forest (RF) model by relating training SSSC data from river gaging stations to derived spectral indices for the pixels corresponding to gaging-station locations. The calibrated model is then used to predict SSSC values for every river pixel in a remotely sensed image, which provides the basis for mapping of spatial variability in SSSCs along the river. The pixel data are used to estimate average surficial values of SSSC at cross sections spaced uniformly along the river. Based on the cross-section data, a mixing metric is computed for each cross section. The spatial pattern of change in this metric over distance can be used to define rates and length scales of surficial mixing of suspended sediment downstream of a confluence. This type of information is useful for exploring the potential influence of various controlling factors on mixing downstream of confluences, for evaluating how mixing in a river system varies over time and space, and for determining how these variations influence water quality and ecological conditions along the river.

Prostate segmentation by sparse representation based classification

PubMed Central

Gao, Yaozong; Liao, Shu; Shen, Dinggang

2012-01-01

Purpose: The segmentation of prostate in CT images is of essential importance to external beam radiotherapy, which is one of the major treatments for prostate cancer nowadays. During the radiotherapy, the prostate is radiated by high-energy x rays from different directions. In order to maximize the dose to the cancer and minimize the dose to the surrounding healthy tissues (e.g., bladder and rectum), the prostate in the new treatment image needs to be accurately localized. Therefore, the effectiveness and efficiency of external beam radiotherapy highly depend on the accurate localization of the prostate. However, due to the low contrast of the prostate with its surrounding tissues (e.g., bladder), the unpredicted prostate motion, and the large appearance variations across different treatment days, it is challenging to segment the prostate in CT images. In this paper, the authors present a novel classification based segmentation method to address these problems. Methods: To segment the prostate, the proposed method first uses sparse representation based classification (SRC) to enhance the prostate in CT images by pixel-wise classification, in order to overcome the limitation of poor contrast of the prostate images. Then, based on the classification results, previous segmented prostates of the same patient are used as patient-specific atlases to align onto the current treatment image and the majority voting strategy is finally adopted to segment the prostate. In order to address the limitations of the traditional SRC in pixel-wise classification, especially for the purpose of segmentation, the authors extend SRC from the following four aspects: (1) A discriminant subdictionary learning method is proposed to learn a discriminant and compact representation of training samples for each class so that the discriminant power of SRC can be increased and also SRC can be applied to the large-scale pixel-wise classification. (2) The L1 regularized sparse coding is replaced by the elastic net in order to obtain a smooth and clear prostate boundary in the classification result. (3) Residue-based linear regression is incorporated to improve the classification performance and to extend SRC from hard classification to soft classification. (4) Iterative SRC is proposed by using context information to iteratively refine the classification results. Results: The proposed method has been comprehensively evaluated on a dataset consisting of 330 CT images from 24 patients. The effectiveness of the extended SRC has been validated by comparing it with the traditional SRC based on the proposed four extensions. The experimental results show that our extended SRC can obtain not only more accurate classification results but also smoother and clearer prostate boundary than the traditional SRC. Besides, the comparison with other five state-of-the-art prostate segmentation methods indicates that our method can achieve better performance than other methods under comparison. Conclusions: The authors have proposed a novel prostate segmentation method based on the sparse representation based classification, which can achieve considerably accurate segmentation results in CT prostate segmentation. PMID:23039673

Prostate segmentation by sparse representation based classification.

PubMed

Gao, Yaozong; Liao, Shu; Shen, Dinggang

2012-10-01

The segmentation of prostate in CT images is of essential importance to external beam radiotherapy, which is one of the major treatments for prostate cancer nowadays. During the radiotherapy, the prostate is radiated by high-energy x rays from different directions. In order to maximize the dose to the cancer and minimize the dose to the surrounding healthy tissues (e.g., bladder and rectum), the prostate in the new treatment image needs to be accurately localized. Therefore, the effectiveness and efficiency of external beam radiotherapy highly depend on the accurate localization of the prostate. However, due to the low contrast of the prostate with its surrounding tissues (e.g., bladder), the unpredicted prostate motion, and the large appearance variations across different treatment days, it is challenging to segment the prostate in CT images. In this paper, the authors present a novel classification based segmentation method to address these problems. To segment the prostate, the proposed method first uses sparse representation based classification (SRC) to enhance the prostate in CT images by pixel-wise classification, in order to overcome the limitation of poor contrast of the prostate images. Then, based on the classification results, previous segmented prostates of the same patient are used as patient-specific atlases to align onto the current treatment image and the majority voting strategy is finally adopted to segment the prostate. In order to address the limitations of the traditional SRC in pixel-wise classification, especially for the purpose of segmentation, the authors extend SRC from the following four aspects: (1) A discriminant subdictionary learning method is proposed to learn a discriminant and compact representation of training samples for each class so that the discriminant power of SRC can be increased and also SRC can be applied to the large-scale pixel-wise classification. (2) The L1 regularized sparse coding is replaced by the elastic net in order to obtain a smooth and clear prostate boundary in the classification result. (3) Residue-based linear regression is incorporated to improve the classification performance and to extend SRC from hard classification to soft classification. (4) Iterative SRC is proposed by using context information to iteratively refine the classification results. The proposed method has been comprehensively evaluated on a dataset consisting of 330 CT images from 24 patients. The effectiveness of the extended SRC has been validated by comparing it with the traditional SRC based on the proposed four extensions. The experimental results show that our extended SRC can obtain not only more accurate classification results but also smoother and clearer prostate boundary than the traditional SRC. Besides, the comparison with other five state-of-the-art prostate segmentation methods indicates that our method can achieve better performance than other methods under comparison. The authors have proposed a novel prostate segmentation method based on the sparse representation based classification, which can achieve considerably accurate segmentation results in CT prostate segmentation.

Balanced VS Imbalanced Training Data: Classifying Rapideye Data with Support Vector Machines

NASA Astrophysics Data System (ADS)

Ustuner, M.; Sanli, F. B.; Abdikan, S.

2016-06-01

The accuracy of supervised image classification is highly dependent upon several factors such as the design of training set (sample selection, composition, purity and size), resolution of input imagery and landscape heterogeneity. The design of training set is still a challenging issue since the sensitivity of classifier algorithm at learning stage is different for the same dataset. In this paper, the classification of RapidEye imagery with balanced and imbalanced training data for mapping the crop types was addressed. Classification with imbalanced training data may result in low accuracy in some scenarios. Support Vector Machines (SVM), Maximum Likelihood (ML) and Artificial Neural Network (ANN) classifications were implemented here to classify the data. For evaluating the influence of the balanced and imbalanced training data on image classification algorithms, three different training datasets were created. Two different balanced datasets which have 70 and 100 pixels for each class of interest and one imbalanced dataset in which each class has different number of pixels were used in classification stage. Results demonstrate that ML and NN classifications are affected by imbalanced training data in resulting a reduction in accuracy (from 90.94% to 85.94% for ML and from 91.56% to 88.44% for NN) while SVM is not affected significantly (from 94.38% to 94.69%) and slightly improved. Our results highlighted that SVM is proven to be a very robust, consistent and effective classifier as it can perform very well under balanced and imbalanced training data situations. Furthermore, the training stage should be precisely and carefully designed for the need of adopted classifier.

Contemplating case mix: A primer on case mix classification and management.

PubMed

Costa, Andrew P; Poss, Jeffery W; McKillop, Ian

2015-01-01

Case mix classifications are the frameworks that underlie many healthcare funding schemes, including the so-called activity-based funding. Now more than ever, Canadian healthcare administrators are evaluating case mix-based funding and deciphering how they will influence their organization. Case mix is a topic fraught with technical jargon and largely relegated to government agencies or private industries. This article provides an abridged review of case mix classification as well as its implications for management in healthcare. © 2015 The Canadian College of Health Leaders.

Image-classification-based global dimming algorithm for LED backlights in LCDs

NASA Astrophysics Data System (ADS)

Qibin, Feng; Huijie, He; Dong, Han; Lei, Zhang; Guoqiang, Lv

2015-07-01

Backlight dimming can help LCDs reduce power consumption and improve CR. With fixed parameters, dimming algorithm cannot achieve satisfied effects for all kinds of images. The paper introduces an image-classification-based global dimming algorithm. The proposed classification method especially for backlight dimming is based on luminance and CR of input images. The parameters for backlight dimming level and pixel compensation are adaptive with image classifications. The simulation results show that the classification based dimming algorithm presents 86.13% power reduction improvement compared with dimming without classification, with almost same display quality. The prototype is developed. There are no perceived distortions when playing videos. The practical average power reduction of the prototype TV is 18.72%, compared with common TV without dimming.

Rapid Classification of Landsat TM Imagery for Phase 1 Stratification Using the Automated NDVI Threshold Supervised Classification (ANTSC) Methodology

Treesearch

William H. Cooke; Dennis M. Jacobs

2005-01-01

FIA annual inventories require rapid updating of pixel-based Phase 1 estimates. Scientists at the Southern Research Station are developing an automated methodology that uses a Normalized Difference Vegetation Index (NDVI) for identifying and eliminating problem FIA plots from the analysis. Problem plots are those that have questionable land use/land cover information....

Nonlinear hyperspectral unmixing based on sparse non-negative matrix factorization

NASA Astrophysics Data System (ADS)

Li, Jing; Li, Xiaorun; Zhao, Liaoying

2016-01-01

Hyperspectral unmixing aims at extracting pure material spectra, accompanied by their corresponding proportions, from a mixed pixel. Owing to modeling more accurate distribution of real material, nonlinear mixing models (non-LMM) are usually considered to hold better performance than LMMs in complicated scenarios. In the past years, numerous nonlinear models have been successfully applied to hyperspectral unmixing. However, most non-LMMs only think of sum-to-one constraint or positivity constraint while the widespread sparsity among real materials mixing is the very factor that cannot be ignored. That is, for non-LMMs, a pixel is usually composed of a few spectral signatures of different materials from all the pure pixel set. Thus, in this paper, a smooth sparsity constraint is incorporated into the state-of-the-art Fan nonlinear model to exploit the sparsity feature in nonlinear model and use it to enhance the unmixing performance. This sparsity-constrained Fan model is solved with the non-negative matrix factorization. The algorithm was implemented on synthetic and real hyperspectral data and presented its advantage over those competing algorithms in the experiments.

An Iterative Inference Procedure Applying Conditional Random Fields for Simultaneous Classification of Land Cover and Land Use

NASA Astrophysics Data System (ADS)

Albert, L.; Rottensteiner, F.; Heipke, C.

2015-08-01

Land cover and land use exhibit strong contextual dependencies. We propose a novel approach for the simultaneous classification of land cover and land use, where semantic and spatial context is considered. The image sites for land cover and land use classification form a hierarchy consisting of two layers: a land cover layer and a land use layer. We apply Conditional Random Fields (CRF) at both layers. The layers differ with respect to the image entities corresponding to the nodes, the employed features and the classes to be distinguished. In the land cover layer, the nodes represent super-pixels; in the land use layer, the nodes correspond to objects from a geospatial database. Both CRFs model spatial dependencies between neighbouring image sites. The complex semantic relations between land cover and land use are integrated in the classification process by using contextual features. We propose a new iterative inference procedure for the simultaneous classification of land cover and land use, in which the two classification tasks mutually influence each other. This helps to improve the classification accuracy for certain classes. The main idea of this approach is that semantic context helps to refine the class predictions, which, in turn, leads to more expressive context information. Thus, potentially wrong decisions can be reversed at later stages. The approach is designed for input data based on aerial images. Experiments are carried out on a test site to evaluate the performance of the proposed method. We show the effectiveness of the iterative inference procedure and demonstrate that a smaller size of the super-pixels has a positive influence on the classification result.

Classification with spatio-temporal interpixel class dependency contexts

NASA Technical Reports Server (NTRS)

Jeon, Byeungwoo; Landgrebe, David A.

1992-01-01

A contextual classifier which can utilize both spatial and temporal interpixel dependency contexts is investigated. After spatial and temporal neighbors are defined, a general form of maximum a posterior spatiotemporal contextual classifier is derived. This contextual classifier is simplified under several assumptions. Joint prior probabilities of the classes of each pixel and its spatial neighbors are modeled by the Gibbs random field. The classification is performed in a recursive manner to allow a computationally efficient contextual classification. Experimental results with bitemporal TM data show significant improvement of classification accuracy over noncontextual pixelwise classifiers. This spatiotemporal contextual classifier should find use in many applications of remote sensing, especially when the classification accuracy is important.

A hybrid MLP-CNN classifier for very fine resolution remotely sensed image classification

NASA Astrophysics Data System (ADS)

Zhang, Ce; Pan, Xin; Li, Huapeng; Gardiner, Andy; Sargent, Isabel; Hare, Jonathon; Atkinson, Peter M.

2018-06-01

The contextual-based convolutional neural network (CNN) with deep architecture and pixel-based multilayer perceptron (MLP) with shallow structure are well-recognized neural network algorithms, representing the state-of-the-art deep learning method and the classical non-parametric machine learning approach, respectively. The two algorithms, which have very different behaviours, were integrated in a concise and effective way using a rule-based decision fusion approach for the classification of very fine spatial resolution (VFSR) remotely sensed imagery. The decision fusion rules, designed primarily based on the classification confidence of the CNN, reflect the generally complementary patterns of the individual classifiers. In consequence, the proposed ensemble classifier MLP-CNN harvests the complementary results acquired from the CNN based on deep spatial feature representation and from the MLP based on spectral discrimination. Meanwhile, limitations of the CNN due to the adoption of convolutional filters such as the uncertainty in object boundary partition and loss of useful fine spatial resolution detail were compensated. The effectiveness of the ensemble MLP-CNN classifier was tested in both urban and rural areas using aerial photography together with an additional satellite sensor dataset. The MLP-CNN classifier achieved promising performance, consistently outperforming the pixel-based MLP, spectral and textural-based MLP, and the contextual-based CNN in terms of classification accuracy. This research paves the way to effectively address the complicated problem of VFSR image classification.

Automatic parquet block sorting using real-time spectral classification

NASA Astrophysics Data System (ADS)

Astrom, Anders; Astrand, Erik; Johansson, Magnus

1999-03-01

This paper presents a real-time spectral classification system based on the PGP spectrograph and a smart image sensor. The PGP is a spectrograph which extracts the spectral information from a scene and projects the information on an image sensor, which is a method often referred to as Imaging Spectroscopy. The classification is based on linear models and categorizes a number of pixels along a line. Previous systems adopting this method have used standard sensors, which often resulted in poor performance. The new system, however, is based on a patented near-sensor classification method, which exploits analogue features on the smart image sensor. The method reduces the enormous amount of data to be processed at an early stage, thus making true real-time spectral classification possible. The system has been evaluated on hardwood parquet boards showing very good results. The color defects considered in the experiments were blue stain, white sapwood, yellow decay and red decay. In addition to these four defect classes, a reference class was used to indicate correct surface color. The system calculates a statistical measure for each parquet block, giving the pixel defect percentage. The patented method makes it possible to run at very high speeds with a high spectral discrimination ability. Using a powerful illuminator, the system can run with a line frequency exceeding 2000 line/s. This opens up the possibility to maintain high production speed and still measure with good resolution.

Computer program documentation for the patch subsampling processor

NASA Technical Reports Server (NTRS)

Nieves, M. J.; Obrien, S. O.; Oney, J. K. (Principal Investigator)

1981-01-01

The programs presented are intended to provide a way to extract a sample from a full-frame scene and summarize it in a useful way. The sample in each case was chosen to fill a 512-by-512 pixel (sample-by-line) image since this is the largest image that can be displayed on the Integrated Multivariant Data Analysis and Classification System. This sample size provides one megabyte of data for manipulation and storage and contains about 3% of the full-frame data. A patch image processor computes means for 256 32-by-32 pixel squares which constitute the 512-by-512 pixel image. Thus, 256 measurements are available for 8 vegetation indexes over a 100-mile square.

DOE Office of Scientific and Technical Information (OSTI.GOV)

Becker, Julian; Tate, Mark W.; Shanks, Katherine S.

Pixel Array Detectors (PADs) consist of an x-ray sensor layer bonded pixel-by-pixel to an underlying readout chip. This approach allows both the sensor and the custom pixel electronics to be tailored independently to best match the x-ray imaging requirements. Here we describe the hybridization of CdTe sensors to two different charge-integrating readout chips, the Keck PAD and the Mixed-Mode PAD (MM-PAD), both developed previously in our laboratory. The charge-integrating architecture of each of these PADs extends the instantaneous counting rate by many orders of magnitude beyond that obtainable with photon counting architectures. The Keck PAD chip consists of rapid, 8-frame,more » in-pixel storage elements with framing periods <150 ns. The second detector, the MM-PAD, has an extended dynamic range by utilizing an in-pixel overflow counter coupled with charge removal circuitry activated at each overflow. This allows the recording of signals from the single-photon level to tens of millions of x-rays/pixel/frame while framing at 1 kHz. Both detector chips consist of a 128×128 pixel array with (150 µm){sup 2} pixels.« less

Threshold selection for classification of MR brain images by clustering method

DOE Office of Scientific and Technical Information (OSTI.GOV)

Moldovanu, Simona; Dumitru Moţoc High School, 15 Milcov St., 800509, Galaţi; Obreja, Cristian

Given a grey-intensity image, our method detects the optimal threshold for a suitable binarization of MR brain images. In MR brain image processing, the grey levels of pixels belonging to the object are not substantially different from the grey levels belonging to the background. Threshold optimization is an effective tool to separate objects from the background and further, in classification applications. This paper gives a detailed investigation on the selection of thresholds. Our method does not use the well-known method for binarization. Instead, we perform a simple threshold optimization which, in turn, will allow the best classification of the analyzedmore » images into healthy and multiple sclerosis disease. The dissimilarity (or the distance between classes) has been established using the clustering method based on dendrograms. We tested our method using two classes of images: the first consists of 20 T2-weighted and 20 proton density PD-weighted scans from two healthy subjects and from two patients with multiple sclerosis. For each image and for each threshold, the number of the white pixels (or the area of white objects in binary image) has been determined. These pixel numbers represent the objects in clustering operation. The following optimum threshold values are obtained, T = 80 for PD images and T = 30 for T2w images. Each mentioned threshold separate clearly the clusters that belonging of the studied groups, healthy patient and multiple sclerosis disease.« less

A Gaussian Mixture Model Representation of Endmember Variability in Hyperspectral Unmixing

NASA Astrophysics Data System (ADS)

Zhou, Yuan; Rangarajan, Anand; Gader, Paul D.

2018-05-01

Hyperspectral unmixing while considering endmember variability is usually performed by the normal compositional model (NCM), where the endmembers for each pixel are assumed to be sampled from unimodal Gaussian distributions. However, in real applications, the distribution of a material is often not Gaussian. In this paper, we use Gaussian mixture models (GMM) to represent the endmember variability. We show, given the GMM starting premise, that the distribution of the mixed pixel (under the linear mixing model) is also a GMM (and this is shown from two perspectives). The first perspective originates from the random variable transformation and gives a conditional density function of the pixels given the abundances and GMM parameters. With proper smoothness and sparsity prior constraints on the abundances, the conditional density function leads to a standard maximum a posteriori (MAP) problem which can be solved using generalized expectation maximization. The second perspective originates from marginalizing over the endmembers in the GMM, which provides us with a foundation to solve for the endmembers at each pixel. Hence, our model can not only estimate the abundances and distribution parameters, but also the distinct endmember set for each pixel. We tested the proposed GMM on several synthetic and real datasets, and showed its potential by comparing it to current popular methods.

A semi-automatic method for quantification and classification of erythrocytes infected with malaria parasites in microscopic images.

PubMed

Díaz, Gloria; González, Fabio A; Romero, Eduardo

2009-04-01

Visual quantification of parasitemia in thin blood films is a very tedious, subjective and time-consuming task. This study presents an original method for quantification and classification of erythrocytes in stained thin blood films infected with Plasmodium falciparum. The proposed approach is composed of three main phases: a preprocessing step, which corrects luminance differences. A segmentation step that uses the normalized RGB color space for classifying pixels either as erythrocyte or background followed by an Inclusion-Tree representation that structures the pixel information into objects, from which erythrocytes are found. Finally, a two step classification process identifies infected erythrocytes and differentiates the infection stage, using a trained bank of classifiers. Additionally, user intervention is allowed when the approach cannot make a proper decision. Four hundred fifty malaria images were used for training and evaluating the method. Automatic identification of infected erythrocytes showed a specificity of 99.7% and a sensitivity of 94%. The infection stage was determined with an average sensitivity of 78.8% and average specificity of 91.2%.

Remote sensing of submerged aquatic vegetation in lower Chesapeake Bay - A comparison of Landsat MSS to TM imagery

NASA Technical Reports Server (NTRS)

Ackleson, S. G.; Klemas, V.

1987-01-01

Landsat MSS and TM imagery, obtained simultaneously over Guinea Marsh, VA, as analyzed and compares for its ability to detect submerged aquatic vegetation (SAV). An unsupervised clustering algorithm was applied to each image, where the input classification parameters are defined as functions of apparent sensor noise. Class confidence and accuracy were computed for all water areas by comparing the classified images, pixel-by-pixel, to rasterized SAV distributions derived from color aerial photography. To illustrate the effect of water depth on classification error, areas of depth greater than 1.9 m were masked, and class confidence and accuracy recalculated. A single-scattering radiative-transfer model is used to illustrate how percent canopy cover and water depth affect the volume reflectance from a water column containing SAV. For a submerged canopy that is morphologically and optically similar to Zostera marina inhabiting Lower Chesapeake Bay, dense canopies may be isolated by masking optically deep water. For less dense canopies, the effect of increasing water depth is to increase the apparent percent crown cover, which may result in classification error.

Bolivian satellite technology program on ERTS natural resources

NASA Technical Reports Server (NTRS)

Brockmann, H. C. (Principal Investigator); Bartoluccic C., L.; Hoffer, R. M.; Levandowski, D. W.; Ugarte, I.; Valenzuela, R. R.; Urena E., M.; Oros, R.

1977-01-01

The author has identified the following significant results. Application of digital classification for mapping land use permitted the separation of units at more specific levels in less time. A correct classification of data in the computer has a positive effect on the accuracy of the final products. Land use unit comparison with types of soils as represented by the colors of the coded map showed a class relation. Soil types in relation to land cover and land use demonstrated that vegetation was a positive factor in soils classification. Groupings of image resolution elements (pixels) permit studies of land use at different levels, thereby forming parameters for the classification of soils.

LANDSAT landcover information applied to regional planning decisions. [Prince Edward County, Virginia

NASA Technical Reports Server (NTRS)

Dixon, C. M.

1981-01-01

Land cover information derived from LANDSAT is being utilized by Piedmont Planning District Commission located in the State of Virginia. Progress to date is reported on a level one land cover classification map being produced with nine categories. The nine categories of classification are defined. The computer compatible tape selection is presented. Two unsupervised classifications were done, with 50 and 70 classes respectively. Twenty-eight spectral classes were developed using the supervised technique, employing actual ground truth training sites. The accuracy of the unsupervised classifications are estimated through comparison with local county statistics and with an actual pixel count of LANDSAT information compared to ground truth.

Estimation of the Botanical Composition of Clover-Grass Leys from RGB Images Using Data Simulation and Fully Convolutional Neural Networks

PubMed Central

Steen, Kim Arild; Green, Ole; Karstoft, Henrik

2017-01-01

Optimal fertilization of clover-grass fields relies on knowledge of the clover and grass fractions. This study shows how knowledge can be obtained by analyzing images collected in fields automatically. A fully convolutional neural network was trained to create a pixel-wise classification of clover, grass, and weeds in red, green, and blue (RGB) images of clover-grass mixtures. The estimated clover fractions of the dry matter from the images were found to be highly correlated with the real clover fractions of the dry matter, making this a cheap and non-destructive way of monitoring clover-grass fields. The network was trained solely on simulated top-down images of clover-grass fields. This enables the network to distinguish clover, grass, and weed pixels in real images. The use of simulated images for training reduces the manual labor to a few hours, as compared to more than 3000 h when all the real images are annotated for training. The network was tested on images with varied clover/grass ratios and achieved an overall pixel classification accuracy of 83.4%, while estimating the dry matter clover fraction with a standard deviation of 7.8%. PMID:29258215

An intelligent support system for automatic detection of cerebral vascular accidents from brain CT images.

PubMed

Hajimani, Elmira; Ruano, M G; Ruano, A E

2017-07-01

This paper presents a Radial Basis Functions Neural Network (RBFNN) based detection system, for automatic identification of Cerebral Vascular Accidents (CVA) through analysis of Computed Tomographic (CT) images. For the design of a neural network classifier, a Multi Objective Genetic Algorithm (MOGA) framework is used to determine the architecture of the classifier, its corresponding parameters and input features by maximizing the classification precision, while ensuring generalization. This approach considers a large number of input features, comprising first and second order pixel intensity statistics, as well as symmetry/asymmetry information with respect to the ideal mid-sagittal line. Values of specificity of 98% and sensitivity of 98% were obtained, at pixel level, by an ensemble of non-dominated models generated by MOGA, in a set of 150 CT slices (1,867,602pixels), marked by a NeuroRadiologist. This approach also compares favorably at a lesion level with three other published solutions, in terms of specificity (86% compared with 84%), degree of coincidence of marked lesions (89% compared with 77%) and classification accuracy rate (96% compared with 88%). Copyright © 2017. Published by Elsevier B.V.

Small-angle solution scattering using the mixed-mode pixel array detector

PubMed Central

Koerner, Lucas J.; Gillilan, Richard E.; Green, Katherine S.; Wang, Suntao; Gruner, Sol M.

2011-01-01

Solution small-angle X-ray scattering (SAXS) measurements were obtained using a 128 × 128 pixel X-ray mixed-mode pixel array detector (MMPAD) with an 860 µs readout time. The MMPAD offers advantages for SAXS experiments: a pixel full-well of >2 × 107 10 keV X-rays, a maximum flux rate of 108 X-rays pixel−1 s−1, and a sub-pixel point-spread function. Data from the MMPAD were quantitatively compared with data from a charge-coupled device (CCD) fiber-optically coupled to a phosphor screen. MMPAD solution SAXS data from lysozyme solutions were of equal or better quality than data captured by the CCD. The read-noise (normalized by pixel area) of the MMPAD was less than that of the CCD by an average factor of 3.0. Short sample-to-detector distances were required owing to the small MMPAD area (19.2 mm × 19.2 mm), and were revealed to be advantageous with respect to detector read-noise. As predicted by the Shannon sampling theory and confirmed by the acquisition of lysozyme solution SAXS curves, the MMPAD at short distances is capable of sufficiently sampling a solution SAXS curve for protein shape analysis. The readout speed of the MMPAD was demonstrated by continuously monitoring lysozyme sample evolution as radiation damage accumulated. These experiments prove that a small suitably configured MMPAD is appropriate for time-resolved solution scattering measurements. PMID:21335900

Accuracy assessments and areal estimates using two-phase stratified random sampling, cluster plots, and the multivariate composite estimator

Treesearch

Raymond L. Czaplewski

2000-01-01

Consider the following example of an accuracy assessment. Landsat data are used to build a thematic map of land cover for a multicounty region. The map classifier (e.g., a supervised classification algorithm) assigns each pixel into one category of land cover. The classification system includes 12 different types of forest and land cover: black spruce, balsam fir,...

Accuracy assessment of biomass and forested area classification from modis, landstat-tm satellite imagery and forest inventory plot data

Treesearch

Dumitru Salajanu; Dennis M. Jacobs

2007-01-01

The objective of this study was to determine how well forestfnon-forest and biomass classifications obtained from Landsat-TM and MODIS satellite data modeled with FIA plots, compare to each other and with forested area and biomass estimates from the national inventory data, as well as whether there is an increase in overall accuracy when pixel size (spatial resolution...

Developing a case mix classification for child and adolescent mental health services: the influence of presenting problems, complexity factors and service providers on number of appointments.

PubMed

Martin, Peter; Davies, Roger; Macdougall, Amy; Ritchie, Benjamin; Vostanis, Panos; Whale, Andy; Wolpert, Miranda

2017-09-01

Case-mix classification is a focus of international attention in considering how best to manage and fund services, by providing a basis for fairer comparison of resource utilization. Yet there is little evidence of the best ways to establish case mix for child and adolescent mental health services (CAMHS). To develop a case mix classification for CAMHS that is clinically meaningful and predictive of number of appointments attended and to investigate the influence of presenting problems, context and complexity factors and provider variation. We analysed 4573 completed episodes of outpatient care from 11 English CAMHS. Cluster analysis, regression trees and a conceptual classification based on clinical best practice guidelines were compared regarding their ability to predict number of appointments, using mixed effects negative binomial regression. The conceptual classification is clinically meaningful and did as well as data-driven classifications in accounting for number of appointments. There was little evidence for effects of complexity or context factors, with the possible exception of school attendance problems. Substantial variation in resource provision between providers was not explained well by case mix. The conceptually-derived classification merits further testing and development in the context of collaborative decision making.

Classification of Urban Aerial Data Based on Pixel Labelling with Deep Convolutional Neural Networks and Logistic Regression

NASA Astrophysics Data System (ADS)

Yao, W.; Poleswki, P.; Krzystek, P.

2016-06-01

The recent success of deep convolutional neural networks (CNN) on a large number of applications can be attributed to large amounts of available training data and increasing computing power. In this paper, a semantic pixel labelling scheme for urban areas using multi-resolution CNN and hand-crafted spatial-spectral features of airborne remotely sensed data is presented. Both CNN and hand-crafted features are applied to image/DSM patches to produce per-pixel class probabilities with a L1-norm regularized logistical regression classifier. The evidence theory infers a degree of belief for pixel labelling from different sources to smooth regions by handling the conflicts present in the both classifiers while reducing the uncertainty. The aerial data used in this study were provided by ISPRS as benchmark datasets for 2D semantic labelling tasks in urban areas, which consists of two data sources from LiDAR and color infrared camera. The test sites are parts of a city in Germany which is assumed to consist of typical object classes including impervious surfaces, trees, buildings, low vegetation, vehicles and clutter. The evaluation is based on the computation of pixel-based confusion matrices by random sampling. The performance of the strategy with respect to scene characteristics and method combination strategies is analyzed and discussed. The competitive classification accuracy could be not only explained by the nature of input data sources: e.g. the above-ground height of nDSM highlight the vertical dimension of houses, trees even cars and the nearinfrared spectrum indicates vegetation, but also attributed to decision-level fusion of CNN's texture-based approach with multichannel spatial-spectral hand-crafted features based on the evidence combination theory.

Methods of editing cloud and atmospheric layer affected pixels from satellite data

NASA Technical Reports Server (NTRS)

Nixon, P. R. (Principal Investigator); Wiegand, C. L.; Richardson, A. J.; Johnson, M. P.; Goodier, B. G.

1981-01-01

The location and migration of cloud, land and water features were examined in spectral space (reflective VIS vs. emissive IR). Daytime HCMM data showed two distinct types of cloud affected pixels in the south Texas test area. High altitude cirrus and/or cirrostratus and "subvisible cirrus" (SCi) reflected the same or only slightly more than land features. In the emissive band, the digital counts ranged from 1 to over 75 and overlapped land features. Pixels consisting of cumulus clouds, or of mixed cumulus and landscape, clustered in a different area of spectral space than the high altitude cloud pixels. Cumulus affected pixels were more reflective than land and water pixels. In August the high altitude clouds and SCi were more emissive than similar clouds were in July. Four-channel TIROS-N data were examined with the objective of developing a multispectral screening technique for removing SCi contaminated data.

Multi-Scale Fractal Analysis of Image Texture and Pattern

NASA Technical Reports Server (NTRS)

Emerson, Charles W.; Lam, Nina Siu-Ngan; Quattrochi, Dale A.

1999-01-01

Analyses of the fractal dimension of Normalized Difference Vegetation Index (NDVI) images of homogeneous land covers near Huntsville, Alabama revealed that the fractal dimension of an image of an agricultural land cover indicates greater complexity as pixel size increases, a forested land cover gradually grows smoother, and an urban image remains roughly self-similar over the range of pixel sizes analyzed (10 to 80 meters). A similar analysis of Landsat Thematic Mapper images of the East Humboldt Range in Nevada taken four months apart show a more complex relation between pixel size and fractal dimension. The major visible difference between the spring and late summer NDVI images is the absence of high elevation snow cover in the summer image. This change significantly alters the relation between fractal dimension and pixel size. The slope of the fractal dimension-resolution relation provides indications of how image classification or feature identification will be affected by changes in sensor spatial resolution.

Multi-Scale Fractal Analysis of Image Texture and Pattern

NASA Technical Reports Server (NTRS)

Emerson, Charles W.; Lam, Nina Siu-Ngan; Quattrochi, Dale A.

1999-01-01

Analyses of the fractal dimension of Normalized Difference Vegetation Index (NDVI) images of homogeneous land covers near Huntsville, Alabama revealed that the fractal dimension of an image of an agricultural land cover indicates greater complexity as pixel size increases, a forested land cover gradually grows smoother, and an urban image remains roughly self-similar over the range of pixel sizes analyzed (10 to 80 meters). A similar analysis of Landsat Thematic Mapper images of the East Humboldt Range in Nevada taken four months apart show a more complex relation between pixel size and fractal dimension. The major visible difference between the spring and late summer NDVI images of the absence of high elevation snow cover in the summer image. This change significantly alters the relation between fractal dimension and pixel size. The slope of the fractal dimensional-resolution relation provides indications of how image classification or feature identification will be affected by changes in sensor spatial resolution.

Synergistic Use of WorldView-2 Imagery and Airborne LiDAR Data for Urban Land Cover Classification

NASA Astrophysics Data System (ADS)

Wu, M. F.; Sun, Z. C.; Yang, B.; Yu, S. S.

2017-02-01

There are lots of challenges for deriving urban land cover types for high resolution optical imagery because of spectral similarity of different objects, mixed pixels, shadows of buildings and large tree crowns. In order to reduce these uncertainties, recently, it’s a trend of the classification of urban land cover from multi-source sensors in the field of urban remote sensing. In this study, a hierarchical support vector machine (SVM) classification method was applied to the urban land cover mapping, using the WorldView-2 imagery and airborne Light Detection and Ranging (LiDAR) data. The results showed that: (1) The overall accuracy (OA) and overall kappa (OK) were 72.92% and 0.66 for WorldView-2 imagery alone; while the OA and OK were improved up to 89.44% and 0.87 for the synergistic use of the two types of data source. (2) Buildings and road/parking lots extracted from fused data were more precision and well-shaped. The two classes from fused data were optimally classified with higher producer’s accuracy and user’s accuracy than WorldView-2 imagery alone. The trees were also easily separated from the grasslands when the airborne LiDAR data was added. (3) The fused data could reduce the phenomenon of different spectral character of the complex and detailed objects. It was also helpful to address the problem of shadows from the high-rise buildings. The results from this study indicate that the synergistic use of high resolution optical imagery and airborne LiDAR data can be an efficient approach to improving the classification of urban land cover.

Land use mapping from CBERS-2 images with open source tools by applying different classification algorithms

NASA Astrophysics Data System (ADS)

Sanhouse-García, Antonio J.; Rangel-Peraza, Jesús Gabriel; Bustos-Terrones, Yaneth; García-Ferrer, Alfonso; Mesas-Carrascosa, Francisco J.

2016-02-01

Land cover classification is often based on different characteristics between their classes, but with great homogeneity within each one of them. This cover is obtained through field work or by mean of processing satellite images. Field work involves high costs; therefore, digital image processing techniques have become an important alternative to perform this task. However, in some developing countries and particularly in Casacoima municipality in Venezuela, there is a lack of geographic information systems due to the lack of updated information and high costs in software license acquisition. This research proposes a low cost methodology to develop thematic mapping of local land use and types of coverage in areas with scarce resources. Thematic mapping was developed from CBERS-2 images and spatial information available on the network using open source tools. The supervised classification method per pixel and per region was applied using different classification algorithms and comparing them among themselves. Classification method per pixel was based on Maxver algorithms (maximum likelihood) and Euclidean distance (minimum distance), while per region classification was based on the Bhattacharya algorithm. Satisfactory results were obtained from per region classification, where overall reliability of 83.93% and kappa index of 0.81% were observed. Maxver algorithm showed a reliability value of 73.36% and kappa index 0.69%, while Euclidean distance obtained values of 67.17% and 0.61% for reliability and kappa index, respectively. It was demonstrated that the proposed methodology was very useful in cartographic processing and updating, which in turn serve as a support to develop management plans and land management. Hence, open source tools showed to be an economically viable alternative not only for forestry organizations, but for the general public, allowing them to develop projects in economically depressed and/or environmentally threatened areas.

Two-tier tissue decomposition for histopathological image representation and classification.

PubMed

Gultekin, Tunc; Koyuncu, Can Fahrettin; Sokmensuer, Cenk; Gunduz-Demir, Cigdem

2015-01-01

In digital pathology, devising effective image representations is crucial to design robust automated diagnosis systems. To this end, many studies have proposed to develop object-based representations, instead of directly using image pixels, since a histopathological image may contain a considerable amount of noise typically at the pixel-level. These previous studies mostly employ color information to define their objects, which approximately represent histological tissue components in an image, and then use the spatial distribution of these objects for image representation and classification. Thus, object definition has a direct effect on the way of representing the image, which in turn affects classification accuracies. In this paper, our aim is to design a classification system for histopathological images. Towards this end, we present a new model for effective representation of these images that will be used by the classification system. The contributions of this model are twofold. First, it introduces a new two-tier tissue decomposition method for defining a set of multityped objects in an image. Different than the previous studies, these objects are defined combining texture, shape, and size information and they may correspond to individual histological tissue components as well as local tissue subregions of different characteristics. As its second contribution, it defines a new metric, which we call dominant blob scale, to characterize the shape and size of an object with a single scalar value. Our experiments on colon tissue images reveal that this new object definition and characterization provides distinguishing representation of normal and cancerous histopathological images, which is effective to obtain more accurate classification results compared to its counterparts.

A digital pixel cell for address event representation image convolution processing

NASA Astrophysics Data System (ADS)

Camunas-Mesa, Luis; Acosta-Jimenez, Antonio; Serrano-Gotarredona, Teresa; Linares-Barranco, Bernabe

2005-06-01

Address Event Representation (AER) is an emergent neuromorphic interchip communication protocol that allows for real-time virtual massive connectivity between huge number of neurons located on different chips. By exploiting high speed digital communication circuits (with nano-seconds timings), synaptic neural connections can be time multiplexed, while neural activity signals (with mili-seconds timings) are sampled at low frequencies. Also, neurons generate events according to their information levels. Neurons with more information (activity, derivative of activities, contrast, motion, edges,...) generate more events per unit time, and access the interchip communication channel more frequently, while neurons with low activity consume less communication bandwidth. AER technology has been used and reported for the implementation of various type of image sensors or retinae: luminance with local agc, contrast retinae, motion retinae,... Also, there has been a proposal for realizing programmable kernel image convolution chips. Such convolution chips would contain an array of pixels that perform weighted addition of events. Once a pixel has added sufficient event contributions to reach a fixed threshold, the pixel fires an event, which is then routed out of the chip for further processing. Such convolution chips have been proposed to be implemented using pulsed current mode mixed analog and digital circuit techniques. In this paper we present a fully digital pixel implementation to perform the weighted additions and fire the events. This way, for a given technology, there is a fully digital implementation reference against which compare the mixed signal implementations. We have designed, implemented and tested a fully digital AER convolution pixel. This pixel will be used to implement a full AER convolution chip for programmable kernel image convolution processing.

All-passive pixel super-resolution of time-stretch imaging

PubMed Central

Chan, Antony C. S.; Ng, Ho-Cheung; Bogaraju, Sharat C. V.; So, Hayden K. H.; Lam, Edmund Y.; Tsia, Kevin K.

2017-01-01

Based on image encoding in a serial-temporal format, optical time-stretch imaging entails a stringent requirement of state-of-the-art fast data acquisition unit in order to preserve high image resolution at an ultrahigh frame rate — hampering the widespread utilities of such technology. Here, we propose a pixel super-resolution (pixel-SR) technique tailored for time-stretch imaging that preserves pixel resolution at a relaxed sampling rate. It harnesses the subpixel shifts between image frames inherently introduced by asynchronous digital sampling of the continuous time-stretch imaging process. Precise pixel registration is thus accomplished without any active opto-mechanical subpixel-shift control or other additional hardware. Here, we present the experimental pixel-SR image reconstruction pipeline that restores high-resolution time-stretch images of microparticles and biological cells (phytoplankton) at a relaxed sampling rate (≈2–5 GSa/s)—more than four times lower than the originally required readout rate (20 GSa/s) — is thus effective for high-throughput label-free, morphology-based cellular classification down to single-cell precision. Upon integration with the high-throughput image processing technology, this pixel-SR time-stretch imaging technique represents a cost-effective and practical solution for large scale cell-based phenotypic screening in biomedical diagnosis and machine vision for quality control in manufacturing. PMID:28303936

As-built design specification for PARCLS

NASA Technical Reports Server (NTRS)

Tompkins, M. A. (Principal Investigator)

1981-01-01

The PARCLS program, part of the CLASFYG package, reads a parameter file created by the CLASFYG program and a pure pixel ground truth file in order to create to classification file of three separate crop categories in universal format.

A Subpixel Classification of Multispectral Satellite Imagery for Interpetation of Tundra-Taiga Ecotone Vegetation (Case Study on Tuliok River Valley, Khibiny, Russia)

NASA Astrophysics Data System (ADS)

Mikheeva, A. I.; Tutubalina, O. V.; Zimin, M. V.; Golubeva, E. I.

2017-12-01

The tundra-taiga ecotone plays significant role in northern ecosystems. Due to global climatic changes, the vegetation of the ecotone is the key object of many remote-sensing studies. The interpretation of vegetation and nonvegetation objects of the tundra-taiga ecotone on satellite imageries of a moderate resolution is complicated by the difficulty of extracting these objects from the spectral and spatial mixtures within a pixel. This article describes a method for the subpixel classification of Terra ASTER satellite image for vegetation mapping of the tundra-taiga ecotone in the Tuliok River, Khibiny Mountains, Russia. It was demonstrated that this method allows to determine the position of the boundaries of ecotone objects and their abundance on the basis of quantitative criteria, which provides a more accurate characteristic of ecotone vegetation when compared to the per-pixel approach of automatic imagery interpretation.

Three-dimensional object recognition using similar triangles and decision trees

NASA Technical Reports Server (NTRS)

Spirkovska, Lilly

1993-01-01

A system, TRIDEC, that is capable of distinguishing between a set of objects despite changes in the objects' positions in the input field, their size, or their rotational orientation in 3D space is described. TRIDEC combines very simple yet effective features with the classification capabilities of inductive decision tree methods. The feature vector is a list of all similar triangles defined by connecting all combinations of three pixels in a coarse coded 127 x 127 pixel input field. The classification is accomplished by building a decision tree using the information provided from a limited number of translated, scaled, and rotated samples. Simulation results are presented which show that TRIDEC achieves 94 percent recognition accuracy in the 2D invariant object recognition domain and 98 percent recognition accuracy in the 3D invariant object recognition domain after training on only a small sample of transformed views of the objects.

User oriented ERTS-1 images. [vegetation identification in Canada through image enhancement

NASA Technical Reports Server (NTRS)

Shlien, S.; Goodenough, D.

1974-01-01

Photographic reproduction of ERTS-1 images are capable of displaying only a portion of the total information available from the multispectral scanner. Methods are being developed to generate ERTS-1 images oriented towards special users such as agriculturists, foresters, and hydrologists by applying image enhancement techniques and interactive statistical classification schemes. Spatial boundaries and linear features can be emphasized and delineated using simple filters. Linear and nonlinear transformations can be applied to the spectral data to emphasize certain ground information. An automatic classification scheme was developed to identify particular ground cover classes such as fallow, grain, rape seed or various vegetation covers. The scheme applies the maximum likelihood decision rule to the spectral information and classifies the ERTS-1 image on a pixel by pixel basis. Preliminary results indicate that the classifier has limited success in distinguishing crops, but is well adapted for identifying different types of vegetation.

Cloud field classification based on textural features

NASA Technical Reports Server (NTRS)

Sengupta, Sailes Kumar

1989-01-01

An essential component in global climate research is accurate cloud cover and type determination. Of the two approaches to texture-based classification (statistical and textural), only the former is effective in the classification of natural scenes such as land, ocean, and atmosphere. In the statistical approach that was adopted, parameters characterizing the stochastic properties of the spatial distribution of grey levels in an image are estimated and then used as features for cloud classification. Two types of textural measures were used. One is based on the distribution of the grey level difference vector (GLDV), and the other on a set of textural features derived from the MaxMin cooccurrence matrix (MMCM). The GLDV method looks at the difference D of grey levels at pixels separated by a horizontal distance d and computes several statistics based on this distribution. These are then used as features in subsequent classification. The MaxMin tectural features on the other hand are based on the MMCM, a matrix whose (I,J)th entry give the relative frequency of occurrences of the grey level pair (I,J) that are consecutive and thresholded local extremes separated by a given pixel distance d. Textural measures are then computed based on this matrix in much the same manner as is done in texture computation using the grey level cooccurrence matrix. The database consists of 37 cloud field scenes from LANDSAT imagery using a near IR visible channel. The classification algorithm used is the well known Stepwise Discriminant Analysis. The overall accuracy was estimated by the percentage or correct classifications in each case. It turns out that both types of classifiers, at their best combination of features, and at any given spatial resolution give approximately the same classification accuracy. A neural network based classifier with a feed forward architecture and a back propagation training algorithm is used to increase the classification accuracy, using these two classes of features. Preliminary results based on the GLDV textural features alone look promising.

Spectral-spatial hyperspectral image classification using super-pixel-based spatial pyramid representation

NASA Astrophysics Data System (ADS)

Fan, Jiayuan; Tan, Hui Li; Toomik, Maria; Lu, Shijian

2016-10-01

Spatial pyramid matching has demonstrated its power for image recognition task by pooling features from spatially increasingly fine sub-regions. Motivated by the concept of feature pooling at multiple pyramid levels, we propose a novel spectral-spatial hyperspectral image classification approach using superpixel-based spatial pyramid representation. This technique first generates multiple superpixel maps by decreasing the superpixel number gradually along with the increased spatial regions for labelled samples. By using every superpixel map, sparse representation of pixels within every spatial region is then computed through local max pooling. Finally, features learned from training samples are aggregated and trained by a support vector machine (SVM) classifier. The proposed spectral-spatial hyperspectral image classification technique has been evaluated on two public hyperspectral datasets, including the Indian Pines image containing 16 different agricultural scene categories with a 20m resolution acquired by AVIRIS and the University of Pavia image containing 9 land-use categories with a 1.3m spatial resolution acquired by the ROSIS-03 sensor. Experimental results show significantly improved performance compared with the state-of-the-art works. The major contributions of this proposed technique include (1) a new spectral-spatial classification approach to generate feature representation for hyperspectral image, (2) a complementary yet effective feature pooling approach, i.e. the superpixel-based spatial pyramid representation that is used for the spatial correlation study, (3) evaluation on two public hyperspectral image datasets with superior image classification performance.

Automatic classification of endoscopic images for premalignant conditions of the esophagus

NASA Astrophysics Data System (ADS)

Boschetto, Davide; Gambaretto, Gloria; Grisan, Enrico

2016-03-01

Barrett's esophagus (BE) is a precancerous complication of gastroesophageal reflux disease in which normal stratified squamous epithelium lining the esophagus is replaced by intestinal metaplastic columnar epithelium. Repeated endoscopies and multiple biopsies are often necessary to establish the presence of intestinal metaplasia. Narrow Band Imaging (NBI) is an imaging technique commonly used with endoscopies that enhances the contrast of vascular pattern on the mucosa. We present a computer-based method for the automatic normal/metaplastic classification of endoscopic NBI images. Superpixel segmentation is used to identify and cluster pixels belonging to uniform regions. From each uniform clustered region of pixels, eight features maximizing differences among normal and metaplastic epithelium are extracted for the classification step. For each superpixel, the three mean intensities of each color channel are firstly selected as features. Three added features are the mean intensities for each superpixel after separately applying to the red-channel image three different morphological filters (top-hat filtering, entropy filtering and range filtering). The last two features require the computation of the Grey-Level Co-Occurrence Matrix (GLCM), and are reflective of the contrast and the homogeneity of each superpixel. The classification step is performed using an ensemble of 50 classification trees, with a 10-fold cross-validation scheme by training the classifier at each step on a random 70% of the images and testing on the remaining 30% of the dataset. Sensitivity and Specificity are respectively of 79.2% and 87.3%, with an overall accuracy of 83.9%.

Mapping Sub-Antarctic Cushion Plants Using Random Forests to Combine Very High Resolution Satellite Imagery and Terrain Modelling

PubMed Central

Bricher, Phillippa K.; Lucieer, Arko; Shaw, Justine; Terauds, Aleks; Bergstrom, Dana M.

2013-01-01

Monitoring changes in the distribution and density of plant species often requires accurate and high-resolution baseline maps of those species. Detecting such change at the landscape scale is often problematic, particularly in remote areas. We examine a new technique to improve accuracy and objectivity in mapping vegetation, combining species distribution modelling and satellite image classification on a remote sub-Antarctic island. In this study, we combine spectral data from very high resolution WorldView-2 satellite imagery and terrain variables from a high resolution digital elevation model to improve mapping accuracy, in both pixel- and object-based classifications. Random forest classification was used to explore the effectiveness of these approaches on mapping the distribution of the critically endangered cushion plant Azorella macquariensis Orchard (Apiaceae) on sub-Antarctic Macquarie Island. Both pixel- and object-based classifications of the distribution of Azorella achieved very high overall validation accuracies (91.6–96.3%, κ = 0.849–0.924). Both two-class and three-class classifications were able to accurately and consistently identify the areas where Azorella was absent, indicating that these maps provide a suitable baseline for monitoring expected change in the distribution of the cushion plants. Detecting such change is critical given the threats this species is currently facing under altering environmental conditions. The method presented here has applications to monitoring a range of species, particularly in remote and isolated environments. PMID:23940805

Simulating urban land cover changes at sub-pixel level in a coastal city

NASA Astrophysics Data System (ADS)

Zhao, Xiaofeng; Deng, Lei; Feng, Huihui; Zhao, Yanchuang

2014-10-01

The simulation of urban expansion or land cover changes is a major theme in both geographic information science and landscape ecology. Yet till now, almost all of previous studies were based on grid computations at pixel level. With the prevalence of spectral mixture analysis in urban land cover research, the simulation of urban land cover at sub-pixel level is being put into agenda. This study provided a new approach of land cover simulation at sub-pixel level. Landsat TM/ETM+ images of Xiamen city, China on both the January of 2002 and 2007 were used to acquire land cover data through supervised classification. Then the two classified land cover data were utilized to extract the transformation rule between 2002 and 2007 using logistic regression. The transformation possibility of each land cover type in a certain pixel was taken as its percent in the same pixel after normalization. And cellular automata (CA) based grid computation was carried out to acquire simulated land cover on 2007. The simulated 2007 sub-pixel land cover was testified with a validated sub-pixel land cover achieved by spectral mixture analysis in our previous studies on the same date. And finally the sub-pixel land cover of 2017 was simulated for urban planning and management. The results showed that our method is useful in land cover simulation at sub-pixel level. Although the simulation accuracy is not quite satisfactory for all the land cover types, it provides an important idea and a good start in the CA-based urban land cover simulation.

Predicting relative species composition within mixed conifer forest pixels using zero‐inflated models and Landsat imagery

Treesearch

Shannon L. Savage; Rick L. Lawrence; John R. Squires

2015-01-01

Ecological and land management applications would often benefit from maps of relative canopy cover of each species present within a pixel, instead of traditional remote-sensing based maps of either dominant species or percent canopy cover without regard to species composition. Widely used statistical models for remote sensing, such as randomForest (RF),...

Hyperspectral image classification based on local binary patterns and PCANet

NASA Astrophysics Data System (ADS)

Yang, Huizhen; Gao, Feng; Dong, Junyu; Yang, Yang

2018-04-01

Hyperspectral image classification has been well acknowledged as one of the challenging tasks of hyperspectral data processing. In this paper, we propose a novel hyperspectral image classification framework based on local binary pattern (LBP) features and PCANet. In the proposed method, linear prediction error (LPE) is first employed to select a subset of informative bands, and LBP is utilized to extract texture features. Then, spectral and texture features are stacked into a high dimensional vectors. Next, the extracted features of a specified position are transformed to a 2-D image. The obtained images of all pixels are fed into PCANet for classification. Experimental results on real hyperspectral dataset demonstrate the effectiveness of the proposed method.

Adaptive local thresholding for robust nucleus segmentation utilizing shape priors

NASA Astrophysics Data System (ADS)

Wang, Xiuzhong; Srinivas, Chukka

2016-03-01

This paper describes a novel local thresholding method for foreground detection. First, a Canny edge detection method is used for initial edge detection. Then, tensor voting is applied on the initial edge pixels, using a nonsymmetric tensor field tailored to encode prior information about nucleus size, shape, and intensity spatial distribution. Tensor analysis is then performed to generate the saliency image and, based on that, the refined edge. Next, the image domain is divided into blocks. In each block, at least one foreground and one background pixel are sampled for each refined edge pixel. The saliency weighted foreground histogram and background histogram are then created. These two histograms are used to calculate a threshold by minimizing the background and foreground pixel classification error. The block-wise thresholds are then used to generate the threshold for each pixel via interpolation. Finally, the foreground is obtained by comparing the original image with the threshold image. The effective use of prior information, combined with robust techniques, results in far more reliable foreground detection, which leads to robust nucleus segmentation.

Comparing Pixel- and Object-Based Approaches in Effectively Classifying Wetland-Dominated Landscapes

EPA Science Inventory

Wetland ecosystems straddle both terrestrial and aquatic habitats, performing many ecological functions directly and indirectly benefitting humans. However, global wetland losses are substantial. Satellite remote sensing and classification informs wise wetland management and moni...

Spatial-spectral blood cell classification with microscopic hyperspectral imagery

NASA Astrophysics Data System (ADS)

Ran, Qiong; Chang, Lan; Li, Wei; Xu, Xiaofeng

2017-10-01

Microscopic hyperspectral images provide a new way for blood cell examination. The hyperspectral imagery can greatly facilitate the classification of different blood cells. In this paper, the microscopic hyperspectral images are acquired by connecting the microscope and the hyperspectral imager, and then tested for blood cell classification. For combined use of the spectral and spatial information provided by hyperspectral images, a spatial-spectral classification method is improved from the classical extreme learning machine (ELM) by integrating spatial context into the image classification task with Markov random field (MRF) model. Comparisons are done among ELM, ELM-MRF, support vector machines(SVM) and SVMMRF methods. Results show the spatial-spectral classification methods(ELM-MRF, SVM-MRF) perform better than pixel-based methods(ELM, SVM), and the proposed ELM-MRF has higher precision and show more accurate location of cells.

Application of Polynomial Neural Networks to Classification of Acoustic Warfare Signals

DTIC Science & Technology

1993-04-01

on Neural Networks, Vol. II, Jun’e, 1987. [66] Shynk, J.J., "Adaptive IIR filtering," IEEE ASSP Magazine, Vol. 6, No. 2, Apr. 1989. 175 I [67] Specht ...rows This is the size of the yellow capture window which will be displayed on the screen. The best setting for pixel-rows is two greater than exemplar...exemplar size of 4 to be captured by the PNN. The pixel-rows setting is 6, which allows all four rows of I the retina data to fit inside yellow capture

A Proposed Extension to the Soil Moisture and Ocean Salinity Level 2 Algorithm for Mixed Forest and Moderate Vegetation Pixels

NASA Technical Reports Server (NTRS)

Panciera, Rocco; Walker, Jeffrey P.; Kalma, Jetse; Kim, Edward

2011-01-01

The Soil Moisture and Ocean Salinity (SMOS)mission, launched in November 2009, provides global maps of soil moisture and ocean salinity by measuring the L-band (1.4 GHz) emission of the Earth's surface with a spatial resolution of 40-50 km.Uncertainty in the retrieval of soilmoisture over large heterogeneous areas such as SMOS pixels is expected, due to the non-linearity of the relationship between soil moisture and the microwave emission. The current baseline soilmoisture retrieval algorithm adopted by SMOS and implemented in the SMOS Level 2 (SMOS L2) processor partially accounts for the sub-pixel heterogeneity of the land surface, by modelling the individual contributions of different pixel fractions to the overall pixel emission. This retrieval approach is tested in this study using airborne L-band data over an area the size of a SMOS pixel characterised by a mix Eucalypt forest and moderate vegetation types (grassland and crops),with the objective of assessing its ability to correct for the soil moisture retrieval error induced by the land surface heterogeneity. A preliminary analysis using a traditional uniform pixel retrieval approach shows that the sub-pixel heterogeneity of land cover type causes significant errors in soil moisture retrieval (7.7%v/v RMSE, 2%v/v bias) in pixels characterised by a significant amount of forest (40-60%). Although the retrieval approach adopted by SMOS partially reduces this error, it is affected by errors beyond the SMOS target accuracy, presenting in particular a strong dry bias when a fraction of the pixel is occupied by forest (4.1%v/v RMSE,-3.1%v/v bias). An extension to the SMOS approach is proposed that accounts for the heterogeneity of vegetation optical depth within the SMOS pixel. The proposed approach is shown to significantly reduce the error in retrieved soil moisture (2.8%v/v RMSE, -0.3%v/v bias) in pixels characterised by a critical amount of forest (40-60%), at the limited cost of only a crude estimate of the optical depth of the forested area (better than 35% uncertainty). This study makes use of an unprecedented data set of airborne L-band observations and ground supporting data from the National Airborne Field Experiment 2005 (NAFE'05), which allowed accurate characterisation of the land surface heterogeneity over an area equivalent in size to a SMOS pixel.

Effect of mixing scanner types and reconstruction kernels on the characterization of lung parenchymal pathologies: emphysema, interstitial pulmonary fibrosis and normal non-smokers

NASA Astrophysics Data System (ADS)

Xu, Ye; van Beek, Edwin J.; McLennan, Geoffrey; Guo, Junfeng; Sonka, Milan; Hoffman, Eric

2006-03-01

In this study we utilize our texture characterization software (3-D AMFM) to characterize interstitial lung diseases (including emphysema) based on MDCT generated volumetric data using 3-dimensional texture features. We have sought to test whether the scanner and reconstruction filter (kernel) type affect the classification of lung diseases using the 3-D AMFM. We collected MDCT images in three subject groups: emphysema (n=9), interstitial pulmonary fibrosis (IPF) (n=10), and normal non-smokers (n=9). In each group, images were scanned either on a Siemens Sensation 16 or 64-slice scanner, (B50f or B30 recon. kernel) or a Philips 4-slice scanner (B recon. kernel). A total of 1516 volumes of interest (VOIs; 21x21 pixels in plane) were marked by two chest imaging experts using the Iowa Pulmonary Analysis Software Suite (PASS). We calculated 24 volumetric features. Bayesian methods were used for classification. Images from different scanners/kernels were combined in all possible combinations to test how robust the tissue classification was relative to the differences in image characteristics. We used 10-fold cross validation for testing the result. Sensitivity, specificity and accuracy were calculated. One-way Analysis of Variances (ANOVA) was used to compare the classification result between the various combinations of scanner and reconstruction kernel types. This study yielded a sensitivity of 94%, 91%, 97%, and 93% for emphysema, ground-glass, honeycombing, and normal non-smoker patterns respectively using a mixture of all three subject groups. The specificity for these characterizations was 97%, 99%, 99%, and 98%, respectively. The F test result of ANOVA shows there is no significant difference (p <0.05) between different combinations of data with respect to scanner and convolution kernel type. Since different MDCT and reconstruction kernel types did not show significant differences in regards to the classification result, this study suggests that the 3-D AMFM can be generally introduced.

Multiple Scale Landscape Pattern Index Interpretation for the Persistent Monitoring of Land-Cover and Land-Use

NASA Astrophysics Data System (ADS)

Spivey, Alvin J.

Mapping land-cover land-use change (LCLUC) over regional and continental scales, and long time scales (years and decades), can be accomplished using thematically identified classification maps of a landscape---a LCLU class map. Observations of a landscape's LCLU class map pattern can indicate the most relevant process, like hydrologic or ecologic function, causing landscape scale environmental change. Quantified as Landscape Pattern Metrics (LPM), emergent landscape patterns act as Landscape Indicators (LI) when physically interpreted. The common mathematical approach to quantifying observed landscape scale pattern is to have LPM measure how connected a class exists within the landscape, through nonlinear local kernel operations of edges and gradients in class maps. Commonly applied kernel-based LPM that consistently reveal causal processes are Dominance, Contagion, and Fractal Dimension. These kernel-based LPM can be difficult to interpret. The emphasis on an image pixel's edge by gradient operations and dependence on an image pixel's existence according to classification accuracy limit the interpretation of LPM. For example, the Dominance and Contagion kernel-based LPM very similarly measure how connected a landscape is. Because of this, their reported edge measurements of connected pattern correlate strongly, making their results ambiguous. Additionally, each of these kernel-based LPM are unscalable when comparing class maps from separate imaging system sensor scenarios that change the image pixel's edge position (i.e. changes in landscape extent, changes in pixel size, changes in orientation, etc), and can only interpret landscape pattern as accurately as the LCLU map classification will allow. This dissertation discusses the reliability of common LPM in light of imaging system effects such as: algorithm classification likelihoods, LCLU classification accuracy due to random image sensor noise, and image scale. A description of an approach to generating well behaved LPM through a Fourier system analysis of the entire class map, or any subset of the class map (e.g. the watershed) is the focus of this work. The Fourier approach provides four improvements for LPM. First, the approach reduces any correlation between metrics by developing them within an independent (i.e. orthogonal) Fourier vector space; a Fourier vector space that includes relevant physically representative parameters ( i.e. between class Euclidean distance). Second, accounting for LCLU classification accuracy the LPM measurement precision and measurement accuracy are reported. Third, the mathematics of this approach makes it possible to compare image data captured at separate pixel resolutions or even from separate landscape scenes. Fourth, Fourier interpreted landscape pattern measurement can be a measure of the entire landscape shape, of individual landscape cover change, or as exchanges between class map subsets by operating on the entire class map, subset of class map, or separate subsets of class map[s] respectively. These LCLUC LPM are examined along the 1991-1992 and 2000-2001 records of National Land Cover Database Landsat data products. Those LPM results are used in a predictive fecal coliform model at the South Carolina watershed level in the context of past (validation study) change. Finally, the proposed LPM ability to be used as ecologically relevant environmental indicators is tested by correlating metrics with other, well known LI that consistently reveal causal processes in the literature.

Fuzzy C-means classification for corrosion evolution of steel images

NASA Astrophysics Data System (ADS)

Trujillo, Maite; Sadki, Mustapha

2004-05-01

An unavoidable problem of metal structures is their exposure to rust degradation during their operational life. Thus, the surfaces need to be assessed in order to avoid potential catastrophes. There is considerable interest in the use of patch repair strategies which minimize the project costs. However, to operate such strategies with confidence in the long useful life of the repair, it is essential that the condition of the existing coatings and the steel substrate can be accurately quantified and classified. This paper describes the application of fuzzy set theory for steel surfaces classification according to the steel rust time. We propose a semi-automatic technique to obtain image clustering using the Fuzzy C-means (FCM) algorithm and we analyze two kinds of data to study the classification performance. Firstly, we investigate the use of raw images" pixels without any pre-processing methods and neighborhood pixels. Secondly, we apply Gaussian noise to the images with different standard deviation to study the FCM method tolerance to Gaussian noise. The noisy images simulate the possible perturbations of the images due to the weather or rust deposits in the steel surfaces during typical on-site acquisition procedures

High-resolution land cover classification using low resolution global data

NASA Astrophysics Data System (ADS)

Carlotto, Mark J.

2013-05-01

A fusion approach is described that combines texture features from high-resolution panchromatic imagery with land cover statistics derived from co-registered low-resolution global databases to obtain high-resolution land cover maps. The method does not require training data or any human intervention. We use an MxN Gabor filter bank consisting of M=16 oriented bandpass filters (0-180°) at N resolutions (3-24 meters/pixel). The size range of these spatial filters is consistent with the typical scale of manmade objects and patterns of cultural activity in imagery. Clustering reduces the complexity of the data by combining pixels that have similar texture into clusters (regions). Texture classification assigns a vector of class likelihoods to each cluster based on its textural properties. Classification is unsupervised and accomplished using a bank of texture anomaly detectors. Class likelihoods are modulated by land cover statistics derived from lower resolution global data over the scene. Preliminary results from a number of Quickbird scenes show our approach is able to classify general land cover features such as roads, built up area, forests, open areas, and bodies of water over a wide range of scenes.

The computer treatment of remotely sensed data: An introduction to techniques which have geologic applications. [image enhancement and thematic classification in Brazil

NASA Technical Reports Server (NTRS)

Parada, N. D. J. (Principal Investigator); Paradella, W. R.; Vitorello, I.

1982-01-01

Several aspects of computer-assisted analysis techniques for image enhancement and thematic classification by which LANDSAT MSS imagery may be treated quantitatively are explained. On geological applications, computer processing of digital data allows, possibly, the fullest use of LANDSAT data, by displaying enhanced and corrected data for visual analysis and by evaluating and assigning each spectral pixel information to a given class.

Parallel processing implementations of a contextual classifier for multispectral remote sensing data

NASA Technical Reports Server (NTRS)

Siegel, H. J.; Swain, P. H.; Smith, B. W.

1980-01-01

Contextual classifiers are being developed as a method to exploit the spatial/spectral context of a pixel to achieve accurate classification. Classification algorithms such as the contextual classifier typically require large amounts of computation time. One way to reduce the execution time of these tasks is through the use of parallelism. The applicability of the CDC flexible processor system and of a proposed multimicroprocessor system (PASM) for implementing contextual classifiers is examined.

Evaluation of linear discriminant analysis for automated Raman histological mapping of esophageal high-grade dysplasia

NASA Astrophysics Data System (ADS)

Hutchings, Joanne; Kendall, Catherine; Shepherd, Neil; Barr, Hugh; Stone, Nicholas

2010-11-01

Rapid Raman mapping has the potential to be used for automated histopathology diagnosis, providing an adjunct technique to histology diagnosis. The aim of this work is to evaluate the feasibility of automated and objective pathology classification of Raman maps using linear discriminant analysis. Raman maps of esophageal tissue sections are acquired. Principal component (PC)-fed linear discriminant analysis (LDA) is carried out using subsets of the Raman map data (6483 spectra). An overall (validated) training classification model performance of 97.7% (sensitivity 95.0 to 100% and specificity 98.6 to 100%) is obtained. The remainder of the map spectra (131,672 spectra) are projected onto the classification model resulting in Raman images, demonstrating good correlation with contiguous hematoxylin and eosin (HE) sections. Initial results suggest that LDA has the potential to automate pathology diagnosis of esophageal Raman images, but since the classification of test spectra is forced into existing training groups, further work is required to optimize the training model. A small pixel size is advantageous for developing the training datasets using mapping data, despite lengthy mapping times, due to additional morphological information gained, and could facilitate differentiation of further tissue groups, such as the basal cells/lamina propria, in the future, but larger pixels sizes (and faster mapping) may be more feasible for clinical application.

Towards the Optimal Pixel Size of dem for Automatic Mapping of Landslide Areas

NASA Astrophysics Data System (ADS)

Pawłuszek, K.; Borkowski, A.; Tarolli, P.

2017-05-01

Determining appropriate spatial resolution of digital elevation model (DEM) is a key step for effective landslide analysis based on remote sensing data. Several studies demonstrated that choosing the finest DEM resolution is not always the best solution. Various DEM resolutions can be applicable for diverse landslide applications. Thus, this study aims to assess the influence of special resolution on automatic landslide mapping. Pixel-based approach using parametric and non-parametric classification methods, namely feed forward neural network (FFNN) and maximum likelihood classification (ML), were applied in this study. Additionally, this allowed to determine the impact of used classification method for selection of DEM resolution. Landslide affected areas were mapped based on four DEMs generated at 1 m, 2 m, 5 m and 10 m spatial resolution from airborne laser scanning (ALS) data. The performance of the landslide mapping was then evaluated by applying landslide inventory map and computation of confusion matrix. The results of this study suggests that the finest scale of DEM is not always the best fit, however working at 1 m DEM resolution on micro-topography scale, can show different results. The best performance was found at 5 m DEM-resolution for FFNN and 1 m DEM resolution for results. The best performance was found to be using 5 m DEM-resolution for FFNN and 1 m DEM resolution for ML classification.

Application of classification methods for mapping Mercury's surface composition: analysis on Rudaki's Area

NASA Astrophysics Data System (ADS)

Zambon, F.; De Sanctis, M. C.; Capaccioni, F.; Filacchione, G.; Carli, C.; Ammanito, E.; Friggeri, A.

2011-10-01

During the first two MESSENGER flybys (14th January 2008 and 6th October 2008) the Mercury Dual Imaging System (MDIS) has extended the coverage of the Mercury surface, obtained by Mariner 10 and now we have images of about 90% of the Mercury surface [1]. MDIS is equipped with a Narrow Angle Camera (NAC) and a Wide Angle Camera (WAC). The NAC uses an off-axis reflective design with a 1.5° field of view (FOV) centered at 747 nm. The WAC has a re- fractive design with a 10.5° FOV and 12-position filters that cover a 395-1040 nm spectral range [2]. The color images can be used to infer information on the surface composition and classification meth- ods are an interesting technique for multispectral image analysis which can be applied to the study of the planetary surfaces. Classification methods are based on clustering algorithms and they can be divided in two categories: unsupervised and supervised. The unsupervised classifiers do not require the analyst feedback, and the algorithm automatically organizes pixels values into classes. In the supervised method, instead, the analyst must choose the "training area" that define the pixels value of a given class [3]. Here we will describe the classification in different compositional units of the region near the Rudaki Crater on Mercury.

Inert Reassessment Document for Poly(oxyethylene) adducts of mixed phytosterols

EPA Pesticide Factsheets

Poly(oxyethy1ene) adducts of mixed phytosterols is uncategorized as to list classification status. Based upon the reasonable certainty of no harm safety finding, the List 4B classification for poly(oxyethy1ene) adducts of mixed phytosterols is affirmed.

Object-oriented feature extraction approach for mapping supraglacial debris in Schirmacher Oasis using very high-resolution satellite data

NASA Astrophysics Data System (ADS)

Jawak, Shridhar D.; Jadhav, Ajay; Luis, Alvarinho J.

2016-05-01

Supraglacial debris was mapped in the Schirmacher Oasis, east Antarctica, by using WorldView-2 (WV-2) high resolution optical remote sensing data consisting of 8-band calibrated Gram Schmidt (GS)-sharpened and atmospherically corrected WV-2 imagery. This study is a preliminary attempt to develop an object-oriented rule set to extract supraglacial debris for Antarctic region using 8-spectral band imagery. Supraglacial debris was manually digitized from the satellite imagery to generate the ground reference data. Several trials were performed using few existing traditional pixel-based classification techniques and color-texture based object-oriented classification methods to extract supraglacial debris over a small domain of the study area. Multi-level segmentation and attributes such as scale, shape, size, compactness along with spectral information from the data were used for developing the rule set. The quantitative analysis of error was carried out against the manually digitized reference data to test the practicability of our approach over the traditional pixel-based methods. Our results indicate that OBIA-based approach (overall accuracy: 93%) for extracting supraglacial debris performed better than all the traditional pixel-based methods (overall accuracy: 80-85%). The present attempt provides a comprehensive improved method for semiautomatic feature extraction in supraglacial environment and a new direction in the cryospheric research.

Urban Density Indices Using Mean Shift-Based Upsampled Elevetion Data

NASA Astrophysics Data System (ADS)

Charou, E.; Gyftakis, S.; Bratsolis, E.; Tsenoglou, T.; Papadopoulou, Th. D.; Vassilas, N.

2015-04-01

Urban density is an important factor for several fields, e.g. urban design, planning and land management. Modern remote sensors deliver ample information for the estimation of specific urban land classification classes (2D indicators), and the height of urban land classification objects (3D indicators) within an Area of Interest (AOI). In this research, two of these indicators, Building Coverage Ratio (BCR) and Floor Area Ratio (FAR) are numerically and automatically derived from high-resolution airborne RGB orthophotos and LiDAR data. In the pre-processing step the low resolution elevation data are fused with the high resolution optical data through a mean-shift based discontinuity preserving smoothing algorithm. The outcome is an improved normalized digital surface model (nDSM) is an upsampled elevation data with considerable improvement regarding region filling and "straightness" of elevation discontinuities. In a following step, a Multilayer Feedforward Neural Network (MFNN) is used to classify all pixels of the AOI to building or non-building categories. For the total surface of the block and the buildings we consider the number of their pixels and the surface of the unit pixel. Comparisons of the automatically derived BCR and FAR indicators with manually derived ones shows the applicability and effectiveness of the methodology proposed.

Semantic segmentation of mFISH images using convolutional networks.

PubMed

Pardo, Esteban; Morgado, José Mário T; Malpica, Norberto

2018-04-30

Multicolor in situ hybridization (mFISH) is a karyotyping technique used to detect major chromosomal alterations using fluorescent probes and imaging techniques. Manual interpretation of mFISH images is a time consuming step that can be automated using machine learning; in previous works, pixel or patch wise classification was employed, overlooking spatial information which can help identify chromosomes. In this work, we propose a fully convolutional semantic segmentation network for the interpretation of mFISH images, which uses both spatial and spectral information to classify each pixel in an end-to-end fashion. The semantic segmentation network developed was tested on samples extracted from a public dataset using cross validation. Despite having no labeling information of the image it was tested on, our algorithm yielded an average correct classification ratio (CCR) of 87.41%. Previously, this level of accuracy was only achieved with state of the art algorithms when classifying pixels from the same image in which the classifier has been trained. These results provide evidence that fully convolutional semantic segmentation networks may be employed in the computer aided diagnosis of genetic diseases with improved performance over the current image analysis methods. © 2018 International Society for Advancement of Cytometry. © 2018 International Society for Advancement of Cytometry.

Automated Detection of Synapses in Serial Section Transmission Electron Microscopy Image Stacks

PubMed Central

Kreshuk, Anna; Koethe, Ullrich; Pax, Elizabeth; Bock, Davi D.; Hamprecht, Fred A.

2014-01-01

We describe a method for fully automated detection of chemical synapses in serial electron microscopy images with highly anisotropic axial and lateral resolution, such as images taken on transmission electron microscopes. Our pipeline starts from classification of the pixels based on 3D pixel features, which is followed by segmentation with an Ising model MRF and another classification step, based on object-level features. Classifiers are learned on sparse user labels; a fully annotated data subvolume is not required for training. The algorithm was validated on a set of 238 synapses in 20 serial 7197×7351 pixel images (4.5×4.5×45 nm resolution) of mouse visual cortex, manually labeled by three independent human annotators and additionally re-verified by an expert neuroscientist. The error rate of the algorithm (12% false negative, 7% false positive detections) is better than state-of-the-art, even though, unlike the state-of-the-art method, our algorithm does not require a prior segmentation of the image volume into cells. The software is based on the ilastik learning and segmentation toolkit and the vigra image processing library and is freely available on our website, along with the test data and gold standard annotations (http://www.ilastik.org/synapse-detection/sstem). PMID:24516550

Pixel decomposition for tracking in low resolution videos

NASA Astrophysics Data System (ADS)

Govinda, Vivekanand; Ralph, Jason F.; Spencer, Joseph W.; Goulermas, John Y.; Yang, Lihua; Abbas, Alaa M.

2008-04-01

This paper describes a novel set of algorithms that allows indoor activity to be monitored using data from very low resolution imagers and other non-intrusive sensors. The objects are not resolved but activity may still be determined. This allows the use of such technology in sensitive environments where privacy must be maintained. Spectral un-mixing algorithms from remote sensing were adapted for this environment. These algorithms allow the fractional contributions from different colours within each pixel to be estimated and this is used to assist in the detection and monitoring of small objects or sub-pixel motion.

Mixed Linear/Square-Root Encoded Single Slope Ramp Provides a Fast, Low Noise Analog to Digital Converter with Very High Linearity for Focal Plane Arrays

NASA Technical Reports Server (NTRS)

Wrigley, Christopher James (Inventor); Hancock, Bruce R. (Inventor); Cunningham, Thomas J. (Inventor); Newton, Kenneth W. (Inventor)

2014-01-01

An analog-to-digital converter (ADC) converts pixel voltages from a CMOS image into a digital output. A voltage ramp generator generates a voltage ramp that has a linear first portion and a non-linear second portion. A digital output generator generates a digital output based on the voltage ramp, the pixel voltages, and comparator output from an array of comparators that compare the voltage ramp to the pixel voltages. A return lookup table linearizes the digital output values.

Using Trained Pixel Classifiers to Select Images of Interest

NASA Technical Reports Server (NTRS)

Mazzoni, D.; Wagstaff, K.; Castano, R.

2004-01-01

We present a machine-learning-based approach to ranking images based on learned priorities. Unlike previous methods for image evaluation, which typically assess the value of each image based on the presence of predetermined specific features, this method involves using two levels of machine-learning classifiers: one level is used to classify each pixel as belonging to one of a group of rather generic classes, and another level is used to rank the images based on these pixel classifications, given some example rankings from a scientist as a guide. Initial results indicate that the technique works well, producing new rankings that match the scientist's rankings significantly better than would be expected by chance. The method is demonstrated for a set of images collected by a Mars field-test rover.

Experimental study of digital image processing techniques for LANDSAT data

NASA Technical Reports Server (NTRS)

Rifman, S. S. (Principal Investigator); Allendoerfer, W. B.; Caron, R. H.; Pemberton, L. J.; Mckinnon, D. M.; Polanski, G.; Simon, K. W.

1976-01-01

The author has identified the following significant results. Results are reported for: (1) subscene registration, (2) full scene rectification and registration, (3) resampling techniques, (4) and ground control point (GCP) extraction. Subscenes (354 pixels x 234 lines) were registered to approximately 1/4 pixel accuracy and evaluated by change detection imagery for three cases: (1) bulk data registration, (2) precision correction of a reference subscene using GCP data, and (3) independently precision processed subscenes. Full scene rectification and registration results were evaluated by using a correlation technique to measure registration errors of 0.3 pixel rms thoughout the full scene. Resampling evaluations of nearest neighbor and TRW cubic convolution processed data included change detection imagery and feature classification. Resampled data were also evaluated for an MSS scene containing specular solar reflections.

Multiple Sparse Representations Classification

PubMed Central

Plenge, Esben; Klein, Stefan S.; Niessen, Wiro J.; Meijering, Erik

2015-01-01

Sparse representations classification (SRC) is a powerful technique for pixelwise classification of images and it is increasingly being used for a wide variety of image analysis tasks. The method uses sparse representation and learned redundant dictionaries to classify image pixels. In this empirical study we propose to further leverage the redundancy of the learned dictionaries to achieve a more accurate classifier. In conventional SRC, each image pixel is associated with a small patch surrounding it. Using these patches, a dictionary is trained for each class in a supervised fashion. Commonly, redundant/overcomplete dictionaries are trained and image patches are sparsely represented by a linear combination of only a few of the dictionary elements. Given a set of trained dictionaries, a new patch is sparse coded using each of them, and subsequently assigned to the class whose dictionary yields the minimum residual energy. We propose a generalization of this scheme. The method, which we call multiple sparse representations classification (mSRC), is based on the observation that an overcomplete, class specific dictionary is capable of generating multiple accurate and independent estimates of a patch belonging to the class. So instead of finding a single sparse representation of a patch for each dictionary, we find multiple, and the corresponding residual energies provides an enhanced statistic which is used to improve classification. We demonstrate the efficacy of mSRC for three example applications: pixelwise classification of texture images, lumen segmentation in carotid artery magnetic resonance imaging (MRI), and bifurcation point detection in carotid artery MRI. We compare our method with conventional SRC, K-nearest neighbor, and support vector machine classifiers. The results show that mSRC outperforms SRC and the other reference methods. In addition, we present an extensive evaluation of the effect of the main mSRC parameters: patch size, dictionary size, and sparsity level. PMID:26177106

Classification of Urban Feature from Unmanned Aerial Vehicle Images Using Gasvm Integration and Multi-Scale Segmentation

NASA Astrophysics Data System (ADS)

Modiri, M.; Salehabadi, A.; Mohebbi, M.; Hashemi, A. M.; Masumi, M.

2015-12-01

The use of UAV in the application of photogrammetry to obtain cover images and achieve the main objectives of the photogrammetric mapping has been a boom in the region. The images taken from REGGIOLO region in the province of, Italy Reggio -Emilia by UAV with non-metric camera Canon Ixus and with an average height of 139.42 meters were used to classify urban feature. Using the software provided SURE and cover images of the study area, to produce dense point cloud, DSM and Artvqvtv spatial resolution of 10 cm was prepared. DTM area using Adaptive TIN filtering algorithm was developed. NDSM area was prepared with using the difference between DSM and DTM and a separate features in the image stack. In order to extract features, using simultaneous occurrence matrix features mean, variance, homogeneity, contrast, dissimilarity, entropy, second moment, and correlation for each of the RGB band image was used Orthophoto area. Classes used to classify urban problems, including buildings, trees and tall vegetation, grass and vegetation short, paved road and is impervious surfaces. Class consists of impervious surfaces such as pavement conditions, the cement, the car, the roof is stored. In order to pixel-based classification and selection of optimal features of classification was GASVM pixel basis. In order to achieve the classification results with higher accuracy and spectral composition informations, texture, and shape conceptual image featureOrthophoto area was fencing. The segmentation of multi-scale segmentation method was used.it belonged class. Search results using the proposed classification of urban feature, suggests the suitability of this method of classification complications UAV is a city using images. The overall accuracy and kappa coefficient method proposed in this study, respectively, 47/93% and 84/91% was.

Health Instruction Packages: Drug Dosage, Classification, and Mixing.

ERIC Educational Resources Information Center

Bracchi, Dorothy P.; And Others

Text, illustrations, and exercises are utilized in a set of seven learning modules to instruct nursing students in the fundamentals of drug classification, dosage, and mixing. The first module, by Dorothy Bracchi, teaches the student to identify six classifications of medication often administered to orthopedic patients: anti-neurospasmolytic…

A new computer approach to mixed feature classification for forestry application

NASA Technical Reports Server (NTRS)

Kan, E. P.

1976-01-01

A computer approach for mapping mixed forest features (i.e., types, classes) from computer classification maps is discussed. Mixed features such as mixed softwood/hardwood stands are treated as admixtures of softwood and hardwood areas. Large-area mixed features are identified and small-area features neglected when the nominal size of a mixed feature can be specified. The computer program merges small isolated areas into surrounding areas by the iterative manipulation of the postprocessing algorithm that eliminates small connected sets. For a forestry application, computer-classified LANDSAT multispectral scanner data of the Sam Houston National Forest were used to demonstrate the proposed approach. The technique was successful in cleaning the salt-and-pepper appearance of multiclass classification maps and in mapping admixtures of softwood areas and hardwood areas. However, the computer-mapped mixed areas matched very poorly with the ground truth because of inadequate resolution and inappropriate definition of mixed features.

Automatic Building Detection based on Supervised Classification using High Resolution Google Earth Images

NASA Astrophysics Data System (ADS)

Ghaffarian, S.; Ghaffarian, S.

2014-08-01

This paper presents a novel approach to detect the buildings by automization of the training area collecting stage for supervised classification. The method based on the fact that a 3d building structure should cast a shadow under suitable imaging conditions. Therefore, the methodology begins with the detection and masking out the shadow areas using luminance component of the LAB color space, which indicates the lightness of the image, and a novel double thresholding technique. Further, the training areas for supervised classification are selected by automatically determining a buffer zone on each building whose shadow is detected by using the shadow shape and the sun illumination direction. Thereafter, by calculating the statistic values of each buffer zone which is collected from the building areas the Improved Parallelepiped Supervised Classification is executed to detect the buildings. Standard deviation thresholding applied to the Parallelepiped classification method to improve its accuracy. Finally, simple morphological operations conducted for releasing the noises and increasing the accuracy of the results. The experiments were performed on set of high resolution Google Earth images. The performance of the proposed approach was assessed by comparing the results of the proposed approach with the reference data by using well-known quality measurements (Precision, Recall and F1-score) to evaluate the pixel-based and object-based performances of the proposed approach. Evaluation of the results illustrates that buildings detected from dense and suburban districts with divers characteristics and color combinations using our proposed method have 88.4 % and 853 % overall pixel-based and object-based precision performances, respectively.

Comparison of Random Forest and Support Vector Machine classifiers using UAV remote sensing imagery

NASA Astrophysics Data System (ADS)

Piragnolo, Marco; Masiero, Andrea; Pirotti, Francesco

2017-04-01

Since recent years surveying with unmanned aerial vehicles (UAV) is getting a great amount of attention due to decreasing costs, higher precision and flexibility of usage. UAVs have been applied for geomorphological investigations, forestry, precision agriculture, cultural heritage assessment and for archaeological purposes. It can be used for land use and land cover classification (LULC). In literature, there are two main types of approaches for classification of remote sensing imagery: pixel-based and object-based. On one hand, pixel-based approach mostly uses training areas to define classes and respective spectral signatures. On the other hand, object-based classification considers pixels, scale, spatial information and texture information for creating homogeneous objects. Machine learning methods have been applied successfully for classification, and their use is increasing due to the availability of faster computing capabilities. The methods learn and train the model from previous computation. Two machine learning methods which have given good results in previous investigations are Random Forest (RF) and Support Vector Machine (SVM). The goal of this work is to compare RF and SVM methods for classifying LULC using images collected with a fixed wing UAV. The processing chain regarding classification uses packages in R, an open source scripting language for data analysis, which provides all necessary algorithms. The imagery was acquired and processed in November 2015 with cameras providing information over the red, blue, green and near infrared wavelength reflectivity over a testing area in the campus of Agripolis, in Italy. Images were elaborated and ortho-rectified through Agisoft Photoscan. The ortho-rectified image is the full data set, and the test set is derived from partial sub-setting of the full data set. Different tests have been carried out, using a percentage from 2 % to 20 % of the total. Ten training sets and ten validation sets are obtained from each test set. The control dataset consist of an independent visual classification done by an expert over the whole area. The classes are (i) broadleaf, (ii) building, (iii) grass, (iv) headland access path, (v) road, (vi) sowed land, (vii) vegetable. The RF and SVM are applied to the test set. The performances of the methods are evaluated using the three following accuracy metrics: Kappa index, Classification accuracy and Classification Error. All three are calculated in three different ways: with K-fold cross validation, using the validation test set and using the full test set. The analysis indicates that SVM gets better results in terms of good scores using K-fold cross or validation test set. Using the full test set, RF achieves a better result in comparison to SVM. It also seems that SVM performs better with smaller training sets, whereas RF performs better as training sets get larger.

SVGA and XGA LCOS microdisplays for HMD applications

NASA Astrophysics Data System (ADS)

Bolotski, Michael; Alvelda, Phillip

1999-07-01

MicroDisplay liquid crystal on silicon (LCOS) display devices are based on a combination of technologies combined with the extreme integration capability of conventionally fabricated CMOS substrates. Two recent SVGA (800 X 600) pixel resolution designs were demonstrated based on 10 micron and 12.5-micron pixel pitch architectures. The resulting microdisplays measure approximately 10 mm and 12 mm in diagonal respectively. Further, an XGA (1024 X 768) resolution display fabricated with a 12.5-micron pixel pitch with a 16-mm diagonal was also demonstrated. Both the larger SVGA and the XGA design were based on the same 12.5-micron pixel-pitch design, demonstrating a quickly scalable design architecture for rapid prototyping life-cycles. All three microdisplay designs described above function in grayscale and high-performance Field-Sequential-Color (FSC) operating modes. The fast liquid crystal operating modes and new scalable high- performance pixel addressing architectures presented in this paper enable substantially improved color, contrast, and brightness while still satisfying the optical, packaging, and power requirements of portable commercial and defense applications including ultra-portable helmet, eyeglass, and heat-mounted systems. The entire suite of The MicroDisplay Corporation's technologies was devised to create a line of mixed-signal application-specific integrated circuits (ASIC) in single-chip display systems. Mixed-signal circuits can integrate computing, memory, and communication circuitry on the same substrate as the display drivers and pixel array for a multifunctional complete system-on-a-chip. For helmet and head-mounted displays this can include capabilities such as the incorporation of customized symbology and information storage directly on the display substrate. System-on-a-chip benefits also include reduced head supported weight requirements through the elimination of off-chip drive electronics.

Segmentation of white blood cells and comparison of cell morphology by linear and naïve Bayes classifiers.

PubMed

Prinyakupt, Jaroonrut; Pluempitiwiriyawej, Charnchai

2015-06-30

Blood smear microscopic images are routinely investigated by haematologists to diagnose most blood diseases. However, the task is quite tedious and time consuming. An automatic detection and classification of white blood cells within such images can accelerate the process tremendously. In this paper we propose a system to locate white blood cells within microscopic blood smear images, segment them into nucleus and cytoplasm regions, extract suitable features and finally, classify them into five types: basophil, eosinophil, neutrophil, lymphocyte and monocyte. Two sets of blood smear images were used in this study's experiments. Dataset 1, collected from Rangsit University, were normal peripheral blood slides under light microscope with 100× magnification; 555 images with 601 white blood cells were captured by a Nikon DS-Fi2 high-definition color camera and saved in JPG format of size 960 × 1,280 pixels at 15 pixels per 1 μm resolution. In dataset 2, 477 cropped white blood cell images were downloaded from CellaVision.com. They are in JPG format of size 360 × 363 pixels. The resolution is estimated to be 10 pixels per 1 μm. The proposed system comprises a pre-processing step, nucleus segmentation, cell segmentation, feature extraction, feature selection and classification. The main concept of the segmentation algorithm employed uses white blood cell's morphological properties and the calibrated size of a real cell relative to image resolution. The segmentation process combined thresholding, morphological operation and ellipse curve fitting. Consequently, several features were extracted from the segmented nucleus and cytoplasm regions. Prominent features were then chosen by a greedy search algorithm called sequential forward selection. Finally, with a set of selected prominent features, both linear and naïve Bayes classifiers were applied for performance comparison. This system was tested on normal peripheral blood smear slide images from two datasets. Two sets of comparison were performed: segmentation and classification. The automatically segmented results were compared to the ones obtained manually by a haematologist. It was found that the proposed method is consistent and coherent in both datasets, with dice similarity of 98.9 and 91.6% for average segmented nucleus and cell regions, respectively. Furthermore, the overall correction rate in the classification phase is about 98 and 94% for linear and naïve Bayes models, respectively. The proposed system, based on normal white blood cell morphology and its characteristics, was applied to two different datasets. The results of the calibrated segmentation process on both datasets are fast, robust, efficient and coherent. Meanwhile, the classification of normal white blood cells into five types shows high sensitivity in both linear and naïve Bayes models, with slightly better results in the linear classifier.

Spatial assessment of intertidal seagrass meadows using optical imaging systems and a lightweight drone

NASA Astrophysics Data System (ADS)

Duffy, James P.; Pratt, Laura; Anderson, Karen; Land, Peter E.; Shutler, Jamie D.

2018-01-01

Seagrass ecosystems are highly sensitive to environmental change. They are also in global decline and under threat from a variety of anthropogenic factors. There is now an urgency to establish robust monitoring methodologies so that changes in seagrass abundance and distribution in these sensitive coastal environments can be understood. Typical monitoring approaches have included remote sensing from satellites and airborne platforms, ground based ecological surveys and snorkel/scuba surveys. These techniques can suffer from temporal and spatial inconsistency, or are very localised making it hard to assess seagrass meadows in a structured manner. Here we present a novel technique using a lightweight (sub 7 kg) drone and consumer grade cameras to produce very high spatial resolution (∼4 mm pixel-1) mosaics of two intertidal sites in Wales, UK. We present a full data collection methodology followed by a selection of classification techniques to produce coverage estimates at each site. We trialled three classification approaches of varying complexity to investigate and illustrate the differing performance and capabilities of each. Our results show that unsupervised classifications perform better than object-based methods in classifying seagrass cover. We also found that the more sparsely vegetated of the two meadows studied was more accurately classified - it had lower root mean squared deviation (RMSD) between observed and classified coverage (9-9.5%) compared to a more densely vegetated meadow (RMSD 16-22%). Furthermore, we examine the potential to detect other biotic features, finding that lugworm mounds can be detected visually at coarser resolutions such as 43 mm pixel-1, whereas smaller features such as cockle shells within seagrass require finer grained data (<17 mm pixel-1).

An embedded face-classification system for infrared images on an FPGA

NASA Astrophysics Data System (ADS)

Soto, Javier E.; Figueroa, Miguel

2014-10-01

We present a face-classification architecture for long-wave infrared (IR) images implemented on a Field Programmable Gate Array (FPGA). The circuit is fast, compact and low power, can recognize faces in real time and be embedded in a larger image-processing and computer vision system operating locally on an IR camera. The algorithm uses Local Binary Patterns (LBP) to perform feature extraction on each IR image. First, each pixel in the image is represented as an LBP pattern that encodes the similarity between the pixel and its neighbors. Uniform LBP codes are then used to reduce the number of patterns to 59 while preserving more than 90% of the information contained in the original LBP representation. Then, the image is divided into 64 non-overlapping regions, and each region is represented as a 59-bin histogram of patterns. Finally, the algorithm concatenates all 64 regions to create a 3,776-bin spatially enhanced histogram. We reduce the dimensionality of this histogram using Linear Discriminant Analysis (LDA), which improves clustering and enables us to store an entire database of 53 subjects on-chip. During classification, the circuit applies LBP and LDA to each incoming IR image in real time, and compares the resulting feature vector to each pattern stored in the local database using the Manhattan distance. We implemented the circuit on a Xilinx Artix-7 XC7A100T FPGA and tested it with the UCHThermalFace database, which consists of 28 81 x 150-pixel images of 53 subjects in indoor and outdoor conditions. The circuit achieves a 98.6% hit ratio, trained with 16 images and tested with 12 images of each subject in the database. Using a 100 MHz clock, the circuit classifies 8,230 images per second, and consumes only 309mW.

SUVI Thematic Maps: A new tool for space weather forecasting

NASA Astrophysics Data System (ADS)

Hughes, J. M.; Seaton, D. B.; Darnel, J.

2017-12-01

The new Solar Ultraviolet Imager (SUVI) instruments aboard NOAA's GOES-R series satellites collect continuous, high-quality imagery of the Sun in six wavelengths. SUVI imagers produce at least one image every 10 seconds, or 8,640 images per day, considerably more data than observers can digest in real time. Over the projected 20-year lifetime of the four GOES-R series spacecraft, SUVI will provide critical imagery for space weather forecasters and produce an extensive but unwieldy archive. In order to condense the database into a dynamic and searchable form we have developed solar thematic maps, maps of the Sun with key features, such as coronal holes, flares, bright regions, quiet corona, and filaments, identified. Thematic maps will be used in NOAA's Space Weather Prediction Center to improve forecaster response time to solar events and generate several derivative products. Likewise, scientists use thematic maps to find observations of interest more easily. Using an expert-trained, naive Bayesian classifier to label each pixel, we create thematic maps in real-time. We created software to collect expert classifications of solar features based on SUVI images. Using this software, we compiled a database of expert classifications, from which we could characterize the distribution of pixels associated with each theme. Given new images, the classifier assigns each pixel the most appropriate label according to the trained distribution. Here we describe the software to collect expert training and the successes and limitations of the classifier. The algorithm excellently identifies coronal holes but fails to consistently detect filaments and prominences. We compare the Bayesian classifier to an artificial neural network, one of our attempts to overcome the aforementioned limitations. These results are very promising and encourage future research into an ensemble classification approach.

Classification of Hyperspectral or Trichromatic Measurements of Ocean Color Data into Spectral Classes.

PubMed

Prasad, Dilip K; Agarwal, Krishna

2016-03-22

We propose a method for classifying radiometric oceanic color data measured by hyperspectral satellite sensors into known spectral classes, irrespective of the downwelling irradiance of the particular day, i.e., the illumination conditions. The focus is not on retrieving the inherent optical properties but to classify the pixels according to the known spectral classes of the reflectances from the ocean. The method compensates for the unknown downwelling irradiance by white balancing the radiometric data at the ocean pixels using the radiometric data of bright pixels (typically from clouds). The white-balanced data is compared with the entries in a pre-calibrated lookup table in which each entry represents the spectral properties of one class. The proposed approach is tested on two datasets of in situ measurements and 26 different daylight illumination spectra for medium resolution imaging spectrometer (MERIS), moderate-resolution imaging spectroradiometer (MODIS), sea-viewing wide field-of-view sensor (SeaWiFS), coastal zone color scanner (CZCS), ocean and land colour instrument (OLCI), and visible infrared imaging radiometer suite (VIIRS) sensors. Results are also shown for CIMEL's SeaPRISM sun photometer sensor used on-board field trips. Accuracy of more than 92% is observed on the validation dataset and more than 86% is observed on the other dataset for all satellite sensors. The potential of applying the algorithms to non-satellite and non-multi-spectral sensors mountable on airborne systems is demonstrated by showing classification results for two consumer cameras. Classification on actual MERIS data is also shown. Additional results comparing the spectra of remote sensing reflectance with level 2 MERIS data and chlorophyll concentration estimates of the data are included.

Comparison of Hyperspectral and Multispectral Satellites for Forest Alliance Classification in the San Francisco Bay Area

NASA Astrophysics Data System (ADS)

Clark, M. L.

2016-12-01

The goal of this study was to assess multi-temporal, Hyperspectral Infrared Imager (HyspIRI) satellite imagery for improved forest class mapping relative to multispectral satellites. The study area was the western San Francisco Bay Area, California and forest alliances (e.g., forest communities defined by dominant or co-dominant trees) were defined using the U.S. National Vegetation Classification System. Simulated 30-m HyspIRI, Landsat 8 and Sentinel-2 imagery were processed from image data acquired by NASA's AVIRIS airborne sensor in year 2015, with summer and multi-temporal (spring, summer, fall) data analyzed separately. HyspIRI reflectance was used to generate a suite of hyperspectral metrics that targeted key spectral features related to chemical and structural properties. The Random Forests classifier was applied to the simulated images and overall accuracies (OA) were compared to those from real Landsat 8 images. For each image group, broad land cover (e.g., Needle-leaf Trees, Broad-leaf Trees, Annual agriculture, Herbaceous, Built-up) was classified first, followed by a finer-detail forest alliance classification for pixels mapped as closed-canopy forest. There were 5 needle-leaf tree alliances and 16 broad-leaf tree alliances, including 7 Quercus (oak) alliance types. No forest alliance classification exceeded 50% OA, indicating that there was broad spectral similarity among alliances, most of which were not spectrally pure but rather a mix of tree species. In general, needle-leaf (Pine, Redwood, Douglas Fir) alliances had better class accuracies than broad-leaf alliances (Oaks, Madrone, Bay Laurel, Buckeye, etc). Multi-temporal data classifications all had 5-6% greater OA than with comparable summer data. For simulated data, HyspIRI metrics had 4-5% greater OA than Landsat 8 and Sentinel-2 multispectral imagery and 3-4% greater OA than HyspIRI reflectance. Finally, HyspIRI metrics had 8% greater OA than real Landsat 8 imagery. In conclusion, forest alliance classification was found to be a difficult remote sensing application with moderate resolution (30 m) satellite imagery; however, of the data tested, HyspIRI spectral metrics had the best performance relative to multispectral satellites.

Marker-Based Hierarchical Segmentation and Classification Approach for Hyperspectral Imagery

NASA Technical Reports Server (NTRS)

Tarabalka, Yuliya; Tilton, James C.; Benediktsson, Jon Atli; Chanussot, Jocelyn

2011-01-01

The Hierarchical SEGmentation (HSEG) algorithm, which is a combination of hierarchical step-wise optimization and spectral clustering, has given good performances for hyperspectral image analysis. This technique produces at its output a hierarchical set of image segmentations. The automated selection of a single segmentation level is often necessary. We propose and investigate the use of automatically selected markers for this purpose. In this paper, a novel Marker-based HSEG (M-HSEG) method for spectral-spatial classification of hyperspectral images is proposed. First, pixelwise classification is performed and the most reliably classified pixels are selected as markers, with the corresponding class labels. Then, a novel constrained marker-based HSEG algorithm is applied, resulting in a spectral-spatial classification map. The experimental results show that the proposed approach yields accurate segmentation and classification maps, and thus is attractive for hyperspectral image analysis.

Evaluation of different shadow detection and restoration methods and their impact on vegetation indices using UAV high-resolution imageries over vineyards

NASA Astrophysics Data System (ADS)

Aboutalebi, M.; Torres-Rua, A. F.; McKee, M.; Kustas, W. P.; Nieto, H.

2017-12-01

Shadows are an unavoidable component of high-resolution imagery. Although shadows can be a useful source of information about terrestrial features, they are a hindrance for image processing and lead to misclassification errors and increased uncertainty in defining surface reflectance properties. In precision agriculture activities, shadows may affect the performance of vegetation indices at pixel and plant scales. Thus, it becomes necessary to evaluate existing shadow detection and restoration methods, especially for applications that makes direct use of pixel information to estimate vegetation biomass, leaf area index (LAI), plant water use and stress, chlorophyll content, just to name a few. In this study, four high-resolution imageries captured by the Utah State University - AggieAir Unmanned Aerial Vehicle (UAV) system flown in 2014, 2015, and 2016 over a commercial vineyard located in the California for the USDA-Agricultural Research Service Grape Remote sensing Atmospheric Profile and Evapotranspiration Experiment (GRAPEX) Program are used for shadow detection and restoration. Four different methods for shadow detection are compared: (1) unsupervised classification, (2) supervised classification, (3) index-based method, and (4) physically-based method. Also, two different shadow restoration methods are evaluated: (1) linear correlation correction, and (2) gamma correction. The models' performance is evaluated over two vegetation indices: normalized difference vegetation index (NDVI) and LAI for both sunlit and shadowed pixels. Histogram and analysis of variance (ANOVA) are used as performance indicators. Results indicated that the performance of the supervised classification and the index-based method are better than other methods. In addition, there is a statistical difference between the average of NDVI and LAI on the sunlit and shadowed pixels. Among the shadow restoration methods, gamma correction visually works better than the linear correlation correction. Moreover, the statistical difference between sunlit and shadowed NDVI and LAI decreases after the application of the gamma restoration method. Potential effects of shadows on modeling surface energy balance and evapotranspiration using very high resolution UAV imagery over the GRAPEX vineyard will be discussed.

A PIXEL COMPOSITION-BASED REFERENCE DATA SET FOR THEMATIC ACCURACY ASSESSMENT

EPA Science Inventory

Developing reference data sets for accuracy assessment of land-cover classifications derived from coarse spatial resolution sensors such as MODIS can be difficult due to the large resolution differences between the image data and available reference data sources. Ideally, the spa...

MKID digital readout tuning with deep learning

NASA Astrophysics Data System (ADS)

Dodkins, R.; Mahashabde, S.; O'Brien, K.; Thatte, N.; Fruitwala, N.; Walter, A. B.; Meeker, S. R.; Szypryt, P.; Mazin, B. A.

2018-04-01

Microwave Kinetic Inductance Detector (MKID) devices offer inherent spectral resolution, simultaneous read out of thousands of pixels, and photon-limited sensitivity at optical wavelengths. Before taking observations the readout power and frequency of each pixel must be individually tuned, and if the equilibrium state of the pixels change, then the readout must be retuned. This process has previously been performed through manual inspection, and typically takes one hour per 500 resonators (20 h for a ten-kilo-pixel array). We present an algorithm based on a deep convolution neural network (CNN) architecture to determine the optimal bias power for each resonator. The bias point classifications from this CNN model, and those from alternative automated methods, are compared to those from human decisions, and the accuracy of each method is assessed. On a test feed-line dataset, the CNN achieves an accuracy of 90% within 1 dB of the designated optimal value, which is equivalent accuracy to a randomly selected human operator, and superior to the highest scoring alternative automated method by 10%. On a full ten-kilopixel array, the CNN performs the characterization in a matter of minutes - paving the way for future mega-pixel MKID arrays.

Classification of high dimensional multispectral image data

NASA Technical Reports Server (NTRS)

Hoffbeck, Joseph P.; Landgrebe, David A.

1993-01-01

A method for classifying high dimensional remote sensing data is described. The technique uses a radiometric adjustment to allow a human operator to identify and label training pixels by visually comparing the remotely sensed spectra to laboratory reflectance spectra. Training pixels for material without obvious spectral features are identified by traditional means. Features which are effective for discriminating between the classes are then derived from the original radiance data and used to classify the scene. This technique is applied to Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) data taken over Cuprite, Nevada in 1992, and the results are compared to an existing geologic map. This technique performed well even with noisy data and the fact that some of the materials in the scene lack absorption features. No adjustment for the atmosphere or other scene variables was made to the data classified. While the experimental results compare favorably with an existing geologic map, the primary purpose of this research was to demonstrate the classification method, as compared to the geology of the Cuprite scene.

Classifying multispectral data by neural networks

NASA Technical Reports Server (NTRS)

Telfer, Brian A.; Szu, Harold H.; Kiang, Richard K.

1993-01-01

Several energy functions for synthesizing neural networks are tested on 2-D synthetic data and on Landsat-4 Thematic Mapper data. These new energy functions, designed specifically for minimizing misclassification error, in some cases yield significant improvements in classification accuracy over the standard least mean squares energy function. In addition to operating on networks with one output unit per class, a new energy function is tested for binary encoded outputs, which result in smaller network sizes. The Thematic Mapper data (four bands were used) is classified on a single pixel basis, to provide a starting benchmark against which further improvements will be measured. Improvements are underway to make use of both subpixel and superpixel (i.e. contextual or neighborhood) information in tile processing. For single pixel classification, the best neural network result is 78.7 percent, compared with 71.7 percent for a classical nearest neighbor classifier. The 78.7 percent result also improves on several earlier neural network results on this data.

Deep learning based classification for head and neck cancer detection with hyperspectral imaging in an animal model

NASA Astrophysics Data System (ADS)

Ma, Ling; Lu, Guolan; Wang, Dongsheng; Wang, Xu; Chen, Zhuo Georgia; Muller, Susan; Chen, Amy; Fei, Baowei

2017-03-01

Hyperspectral imaging (HSI) is an emerging imaging modality that can provide a noninvasive tool for cancer detection and image-guided surgery. HSI acquires high-resolution images at hundreds of spectral bands, providing big data to differentiating different types of tissue. We proposed a deep learning based method for the detection of head and neck cancer with hyperspectral images. Since the deep learning algorithm can learn the feature hierarchically, the learned features are more discriminative and concise than the handcrafted features. In this study, we adopt convolutional neural networks (CNN) to learn the deep feature of pixels for classifying each pixel into tumor or normal tissue. We evaluated our proposed classification method on the dataset containing hyperspectral images from 12 tumor-bearing mice. Experimental results show that our method achieved an average accuracy of 91.36%. The preliminary study demonstrated that our deep learning method can be applied to hyperspectral images for detecting head and neck tumors in animal models.

Classification of simulated and actual NOAA-6 AVHRR data for hydrologic land-surface feature definition. [Advanced Very High Resolution Radiometer

NASA Technical Reports Server (NTRS)

Ormsby, J. P.

1982-01-01

An examination of the possibilities of using Landsat data to simulate NOAA-6 Advanced Very High Resolution Radiometer (AVHRR) data on two channels, as well as using actual NOAA-6 imagery, for large-scale hydrological studies is presented. A running average was obtained of 18 consecutive pixels of 1 km resolution taken by the Landsat scanners were scaled up to 8-bit data and investigated for different gray levels. AVHRR data comprising five channels of 10-bit, band-interleaved information covering 10 deg latitude were analyzed and a suitable pixel grid was chosen for comparison with the Landsat data in a supervised classification format, an unsupervised mode, and with ground truth. Landcover delineation was explored by removing snow, water, and cloud features from the cluster analysis, and resulted in less than 10% difference. Low resolution large-scale data was determined useful for characterizing some landcover features if weekly and/or monthly updates are maintained.

Classification by diagnosing all absorption features (CDAF) for the most abundant minerals in airborne hyperspectral images

NASA Astrophysics Data System (ADS)

Mobasheri, Mohammad Reza; Ghamary-Asl, Mohsen

2011-12-01

Imaging through hyperspectral technology is a powerful tool that can be used to spectrally identify and spatially map materials based on their specific absorption characteristics in electromagnetic spectrum. A robust method called Tetracorder has shown its effectiveness at material identification and mapping, using a set of algorithms within an expert system decision-making framework. In this study, using some stages of Tetracorder, a technique called classification by diagnosing all absorption features (CDAF) is introduced. This technique enables one to assign a class to the most abundant mineral in each pixel with high accuracy. The technique is based on the derivation of information from reflectance spectra of the image. This can be done through extraction of spectral absorption features of any minerals from their respected laboratory-measured reflectance spectra, and comparing it with those extracted from the pixels in the image. The CDAF technique has been executed on the AVIRIS image where the results show an overall accuracy of better than 96%.

Early Validation of Sentinel-2 L2A Processor and Products

NASA Astrophysics Data System (ADS)

Pflug, Bringfried; Main-Knorn, Magdalena; Bieniarz, Jakub; Debaecker, Vincent; Louis, Jerome

2016-08-01

Sentinel-2 is a constellation of two polar orbiting satellite units each one equipped with an optical imaging sensor MSI (Multi-Spectral Instrument). Sentinel-2A was launched on June 23, 2015 and Sentinel-2B will follow in 2017.The Level-2A (L2A) processor Sen2Cor implemented for Sentinel-2 data provides a scene classification image, aerosol optical thickness (AOT) and water vapour (WV) maps and the Bottom-Of-Atmosphere (BOA) corrected reflectance product. First validation results of Sen2Cor scene classification showed an overall accuracy of 81%. AOT at 550 nm is estimated by Sen2Cor with uncertainty of 0.035 for cloudless images and locations with dense dark vegetation (DDV) pixels present in the image. Aerosol estimation fails if the image contains no DDV-pixels. Mean difference between Sen2Cor WV and ground-truth is 0.29 cm. Uncertainty of up to 0.04 was found for the BOA- reflectance product.

Thematic accuracy of the 1992 National Land-Cover Data for the eastern United States: Statistical methodology and regional results

USGS Publications Warehouse

Stehman, S.V.; Wickham, J.D.; Smith, J.H.; Yang, L.

2003-01-01

The accuracy of the 1992 National Land-Cover Data (NLCD) map is assessed via a probability sampling design incorporating three levels of stratification and two stages of selection. Agreement between the map and reference land-cover labels is defined as a match between the primary or alternate reference label determined for a sample pixel and a mode class of the mapped 3×3 block of pixels centered on the sample pixel. Results are reported for each of the four regions comprising the eastern United States for both Anderson Level I and II classifications. Overall accuracies for Levels I and II are 80% and 46% for New England, 82% and 62% for New York/New Jersey (NY/NJ), 70% and 43% for the Mid-Atlantic, and 83% and 66% for the Southeast.

Application of Skylab EREP data for land use management

NASA Technical Reports Server (NTRS)

Simonett, D. S. (Principal Investigator)

1976-01-01

The author has identified the following significant results. The 1.09-1.19 micron band proved to be very valuable for discriminating a variety of land use categories, including agriculture, forest, and urban classes. The 1.55-1.75 micron band proved very useful in combination with the 1.09-1.19 micron band. Misregistration between spectral bands, even by as little as 1/2 pixel, may degrade classification accuracy. Identification accuracy of boundary or border pixels was as much as 13% lower than the accuracy for identifying internal field pixels. The principal conclusion with respect to the S190B camera system is that the higher resolution of the S190B system in comparison to previous space photography (Gemini, Apollo), to the S190A system (Skylab), and to LANDSAT imagery significantly increases the range of additional discrimination achievable.

Lattice algebra approach to multispectral analysis of ancient documents.

PubMed

Valdiviezo-N, Juan C; Urcid, Gonzalo

2013-02-01

This paper introduces a lattice algebra procedure that can be used for the multispectral analysis of historical documents and artworks. Assuming the presence of linearly mixed spectral pixels captured in a multispectral scene, the proposed method computes the scaled min- and max-lattice associative memories to determine the purest pixels that best represent the spectra of single pigments. The estimation of fractional proportions of pure spectra at each image pixel is used to build pigment abundance maps that can be used for subsequent restoration of damaged parts. Application examples include multispectral images acquired from the Archimedes Palimpsest and a Mexican pre-Hispanic codex.

Challenges of rehabilitation case mix measurement in Ontario hospitals.

PubMed

Sutherland, Jason Murray; Walker, Jan

2008-03-01

Case mix classification systems have been adopted in many countries as a method to manage and finance healthcare in acute care settings; the most popular systems are based on diagnosis related groups. The most successful of those case mix systems differentiate patient types by reflecting both the intensity of resources consumed and patient acuity. Case mix systems for use with non-acute hospital activity have not been as wide-spread; other than in the United States, little attention has been directed towards case mix classification for rehabilitation activity. In a province with over 13 million inhabitants with 2496 rehabilitation beds, inpatient rehabilitation is an important component of hospital care in Ontario, Canada, and consists of the spectrum of intensive rehabilitation activities intended to restore function. Although case mix adjusted activity has been the currency in Ontario's Integrated Population Based Allocation hospital funding formula, rehabilitation activity has not been subjected to case mix measurement. A project to examine case mix classification for adult inpatient rehabilitation activity was initiated by the Ontario Ministry of Health and Long-Term Care whose outcome was a case mix system and associated cost weights that would result in rehabilitation activity being incorporated into the hospital funding formula. The process described in this study provides Ontario's provincial government with a case mix classification system for adult inpatient rehabilitation activity although there remain areas for improvement.

The Effects of Radiation on Imagery Sensors in Space

NASA Technical Reports Server (NTRS)

Mathis, Dylan

2007-01-01

Recent experience using high definition video on the International Space Station reveals camera pixel degradation due to particle radiation to be a much more significant problem with high definition cameras than with standard definition video. Although it may at first appear that increased pixel density on the imager is the logical explanation for this, the ISS implementations of high definition suggest a more complex causal and mediating factor mix. The degree of damage seems to vary from one type of camera to another, and this variation prompts a reconsideration of the possible factors in pixel loss, such as imager size, number of pixels, pixel aperture ratio, imager type (CCD or CMOS), method of error correction/concealment, and the method of compression used for recording or transmission. The problem of imager pixel loss due to particle radiation is not limited to out-of-atmosphere applications. Since particle radiation increases with altitude, it is not surprising to find anecdotal evidence that video cameras subject to many hours of airline travel show an increased incidence of pixel loss. This is even evident in some standard definition video applications, and pixel loss due to particle radiation only stands to become a more salient issue considering the continued diffusion of high definition video cameras in the marketplace.

A custom hardware classifier for bruised apple detection in hyperspectral images

NASA Astrophysics Data System (ADS)

Cárdenas, Javier; Figueroa, Miguel; Pezoa, Jorge E.

2015-09-01

We present a custom digital architecture for bruised apple classification using hyperspectral images in the near infrared (NIR) spectrum. The algorithm classifies each pixel in an image into one of three classes: bruised, non-bruised, and background. We extract two 5-element feature vectors for each pixel using only 10 out of the 236 spectral bands provided by the hyperspectral camera, thereby greatly reducing both the requirements of the imager and the computational complexity of the algorithm. We then use two linear-kernel support vector machine (SVM) to classify each pixel. Each SVM was trained with 504 windows of size 17×17-pixel taken from 14 hyperspectral images of 320×320 pixels each, for each class. The architecture then computes the percentage of bruised pixels in each apple in order to adequately classify the fruit. We implemented the architecture on a Xilinx Zynq Z-7010 field-programmable gate array (FPGA) and tested it on images from a NIR N17E push-broom camera with a frame rate of 25 fps, a band-pixel rate of 1.888 MHz, and 236 spectral bands between 900 and 1700 nanometers in laboratory conditions. Using 28-bit fixed-point arithmetic, the circuit accurately discriminates 95.2% of the pixels corresponding to an apple, 81% of the pixels corresponding to a bruised apple, and 96.4% of the background. With the default threshold settings, the highest false positive (FP) for a bruised apple is 18.7%. The circuit operates at the native frame rate of the camera, consumes 67 mW of dynamic power, and uses less than 10% of the logic resources on the FPGA.

A 75-ps Gated CMOS Image Sensor with Low Parasitic Light Sensitivity

PubMed Central

Zhang, Fan; Niu, Hanben

2016-01-01

In this study, a 40 × 48 pixel global shutter complementary metal-oxide-semiconductor (CMOS) image sensor with an adjustable shutter time as low as 75 ps was implemented using a 0.5-μm mixed-signal CMOS process. The implementation consisted of a continuous contact ring around each p+/n-well photodiode in the pixel array in order to apply sufficient light shielding. The parasitic light sensitivity of the in-pixel storage node was measured to be 1/8.5 × 107 when illuminated by a 405-nm diode laser and 1/1.4 × 104 when illuminated by a 650-nm diode laser. The pixel pitch was 24 μm, the size of the square p+/n-well photodiode in each pixel was 7 μm per side, the measured random readout noise was 217 e− rms, and the measured dynamic range of the pixel of the designed chip was 5500:1. The type of gated CMOS image sensor (CIS) that is proposed here can be used in ultra-fast framing cameras to observe non-repeatable fast-evolving phenomena. PMID:27367699

A 75-ps Gated CMOS Image Sensor with Low Parasitic Light Sensitivity.

PubMed

Zhang, Fan; Niu, Hanben

2016-06-29

In this study, a 40 × 48 pixel global shutter complementary metal-oxide-semiconductor (CMOS) image sensor with an adjustable shutter time as low as 75 ps was implemented using a 0.5-μm mixed-signal CMOS process. The implementation consisted of a continuous contact ring around each p+/n-well photodiode in the pixel array in order to apply sufficient light shielding. The parasitic light sensitivity of the in-pixel storage node was measured to be 1/8.5 × 10⁷ when illuminated by a 405-nm diode laser and 1/1.4 × 10⁴ when illuminated by a 650-nm diode laser. The pixel pitch was 24 μm, the size of the square p+/n-well photodiode in each pixel was 7 μm per side, the measured random readout noise was 217 e(-) rms, and the measured dynamic range of the pixel of the designed chip was 5500:1. The type of gated CMOS image sensor (CIS) that is proposed here can be used in ultra-fast framing cameras to observe non-repeatable fast-evolving phenomena.

Comparison of Random Forest, k-Nearest Neighbor, and Support Vector Machine Classifiers for Land Cover Classification Using Sentinel-2 Imagery

PubMed Central

Thanh Noi, Phan; Kappas, Martin

2017-01-01

In previous classification studies, three non-parametric classifiers, Random Forest (RF), k-Nearest Neighbor (kNN), and Support Vector Machine (SVM), were reported as the foremost classifiers at producing high accuracies. However, only a few studies have compared the performances of these classifiers with different training sample sizes for the same remote sensing images, particularly the Sentinel-2 Multispectral Imager (MSI). In this study, we examined and compared the performances of the RF, kNN, and SVM classifiers for land use/cover classification using Sentinel-2 image data. An area of 30 × 30 km2 within the Red River Delta of Vietnam with six land use/cover types was classified using 14 different training sample sizes, including balanced and imbalanced, from 50 to over 1250 pixels/class. All classification results showed a high overall accuracy (OA) ranging from 90% to 95%. Among the three classifiers and 14 sub-datasets, SVM produced the highest OA with the least sensitivity to the training sample sizes, followed consecutively by RF and kNN. In relation to the sample size, all three classifiers showed a similar and high OA (over 93.85%) when the training sample size was large enough, i.e., greater than 750 pixels/class or representing an area of approximately 0.25% of the total study area. The high accuracy was achieved with both imbalanced and balanced datasets. PMID:29271909

Comparison of Random Forest, k-Nearest Neighbor, and Support Vector Machine Classifiers for Land Cover Classification Using Sentinel-2 Imagery.

PubMed

Thanh Noi, Phan; Kappas, Martin

2017-12-22

In previous classification studies, three non-parametric classifiers, Random Forest (RF), k-Nearest Neighbor (kNN), and Support Vector Machine (SVM), were reported as the foremost classifiers at producing high accuracies. However, only a few studies have compared the performances of these classifiers with different training sample sizes for the same remote sensing images, particularly the Sentinel-2 Multispectral Imager (MSI). In this study, we examined and compared the performances of the RF, kNN, and SVM classifiers for land use/cover classification using Sentinel-2 image data. An area of 30 × 30 km² within the Red River Delta of Vietnam with six land use/cover types was classified using 14 different training sample sizes, including balanced and imbalanced, from 50 to over 1250 pixels/class. All classification results showed a high overall accuracy (OA) ranging from 90% to 95%. Among the three classifiers and 14 sub-datasets, SVM produced the highest OA with the least sensitivity to the training sample sizes, followed consecutively by RF and kNN. In relation to the sample size, all three classifiers showed a similar and high OA (over 93.85%) when the training sample size was large enough, i.e., greater than 750 pixels/class or representing an area of approximately 0.25% of the total study area. The high accuracy was achieved with both imbalanced and balanced datasets.

a Region-Based Multi-Scale Approach for Object-Based Image Analysis

NASA Astrophysics Data System (ADS)

Kavzoglu, T.; Yildiz Erdemir, M.; Tonbul, H.

2016-06-01

Within the last two decades, object-based image analysis (OBIA) considering objects (i.e. groups of pixels) instead of pixels has gained popularity and attracted increasing interest. The most important stage of the OBIA is image segmentation that groups spectrally similar adjacent pixels considering not only the spectral features but also spatial and textural features. Although there are several parameters (scale, shape, compactness and band weights) to be set by the analyst, scale parameter stands out the most important parameter in segmentation process. Estimating optimal scale parameter is crucially important to increase the classification accuracy that depends on image resolution, image object size and characteristics of the study area. In this study, two scale-selection strategies were implemented in the image segmentation process using pan-sharped Qickbird-2 image. The first strategy estimates optimal scale parameters for the eight sub-regions. For this purpose, the local variance/rate of change (LV-RoC) graphs produced by the ESP-2 tool were analysed to determine fine, moderate and coarse scales for each region. In the second strategy, the image was segmented using the three candidate scale values (fine, moderate, coarse) determined from the LV-RoC graph calculated for whole image. The nearest neighbour classifier was applied in all segmentation experiments and equal number of pixels was randomly selected to calculate accuracy metrics (overall accuracy and kappa coefficient). Comparison of region-based and image-based segmentation was carried out on the classified images and found that region-based multi-scale OBIA produced significantly more accurate results than image-based single-scale OBIA. The difference in classification accuracy reached to 10% in terms of overall accuracy.

Classification of Active Microwave and Passive Optical Data Based on Bayesian Theory and Mrf

NASA Astrophysics Data System (ADS)

Yu, F.; Li, H. T.; Han, Y. S.; Gu, H. Y.

2012-08-01

A classifier based on Bayesian theory and Markov random field (MRF) is presented to classify the active microwave and passive optical remote sensing data, which have demonstrated their respective advantages in inversion of surface soil moisture content. In the method, the VV, VH polarization of ASAR and all the 7 TM bands are taken as the input of the classifier to get the class labels of each pixel of the images. And the model is validated for the necessities of integration of TM and ASAR, it shows that, the total precision of classification in this paper is 89.4%. Comparing with the classification with single TM, the accuracy increase 11.5%, illustrating that synthesis of active and passive optical remote sensing data is efficient and potential in classification.

Multimodal segmentation of optic disc and cup from stereo fundus and SD-OCT images

NASA Astrophysics Data System (ADS)

Miri, Mohammad Saleh; Lee, Kyungmoo; Niemeijer, Meindert; Abràmoff, Michael D.; Kwon, Young H.; Garvin, Mona K.

2013-03-01

Glaucoma is one of the major causes of blindness worldwide. One important structural parameter for the diagnosis and management of glaucoma is the cup-to-disc ratio (CDR), which tends to become larger as glaucoma progresses. While approaches exist for segmenting the optic disc and cup within fundus photographs, and more recently, within spectral-domain optical coherence tomography (SD-OCT) volumes, no approaches have been reported for the simultaneous segmentation of these structures within both modalities combined. In this work, a multimodal pixel-classification approach for the segmentation of the optic disc and cup within fundus photographs and SD-OCT volumes is presented. In particular, after segmentation of other important structures (such as the retinal layers and retinal blood vessels) and fundus-to-SD-OCT image registration, features are extracted from both modalities and a k-nearest-neighbor classification approach is used to classify each pixel as cup, rim, or background. The approach is evaluated on 70 multimodal image pairs from 35 subjects in a leave-10%-out fashion (by subject). A significant improvement in classification accuracy is obtained using the multimodal approach over that obtained from the corresponding unimodal approach (97.8% versus 95.2%; p < 0:05; paired t-test).

Multi-resolution analysis using integrated microscopic configuration with local patterns for benign-malignant mass classification

NASA Astrophysics Data System (ADS)

Rabidas, Rinku; Midya, Abhishek; Chakraborty, Jayasree; Sadhu, Anup; Arif, Wasim

2018-02-01

In this paper, Curvelet based local attributes, Curvelet-Local configuration pattern (C-LCP), is introduced for the characterization of mammographic masses as benign or malignant. Amid different anomalies such as micro- calcification, bilateral asymmetry, architectural distortion, and masses, the reason for targeting the mass lesions is due to their variation in shape, size, and margin which makes the diagnosis a challenging task. Being efficient in classification, multi-resolution property of the Curvelet transform is exploited and local information is extracted from the coefficients of each subband using Local configuration pattern (LCP). The microscopic measures in concatenation with the local textural information provide more discriminating capability than individual. The measures embody the magnitude information along with the pixel-wise relationships among the neighboring pixels. The performance analysis is conducted with 200 mammograms of the DDSM database containing 100 mass cases of each benign and malignant. The optimal set of features is acquired via stepwise logistic regression method and the classification is carried out with Fisher linear discriminant analysis. The best area under the receiver operating characteristic curve and accuracy of 0.95 and 87.55% are achieved with the proposed method, which is further compared with some of the state-of-the-art competing methods.

Unsupervised classification of scattering behavior using radar polarimetry data

NASA Technical Reports Server (NTRS)

Van Zyl, Jakob J.

1989-01-01

The use of an imaging radar polarimeter data for unsupervised classification of scattering behavior is described by comparing the polarization properties of each pixel in a image to that of simple classes of scattering such as even number of reflections, odd number of reflections, and diffuse scattering. For example, when this algorithm is applied to data acquired over the San Francisco Bay area in California, it classifies scattering by the ocean as being similar to that predicted by the class of odd number of reflections, scattering by the urban area as being similar to that predicted by the class of even number of reflections, and scattering by the Golden Gate Park as being similar to that predicted by the diffuse scattering class. It also classifies the scattering by a lighthouse in the ocean and boats on the ocean surface as being similar to that predicted by the even number of reflections class, making it easy to identify these objects against the background of the surrounding ocean. The algorithm is also applied to forested areas and shows that scattering from clear-cut areas and agricultural fields is mostly similar to that predicted by the odd number of reflections class, while the scattering from tree-covered areas generally is classified as being a mixture of pixels exhibiting the characteristics of all three classes, although each pixel is identified with only a single class.

Mangrove classification through the use of object oriented classification and support vector machine of lidar datasets: a case study in Naawan and Manticao, Misamis Oriental, Philippines

NASA Astrophysics Data System (ADS)

Jalbuena, Rey L.; Peralta, Rudolph V.; Tamondong, Ayin M.

2016-10-01

Mangroves are trees or shrubs that grows at the surface between the land and the sea in tropical and sub-tropical latitudes. Mangroves are essential in supporting various marine life, thus, it is important to preserve and manage these areas. There are many approaches in creating Mangroves maps, one of which is through the use of Light Detection and Ranging (LiDAR). It is a remote sensing technique which uses light pulses to measure distances and to generate three-dimensional point clouds of the Earth's surface. In this study, the topographic LiDAR Data will be used to analyze the geophysical features of the terrain and create a Mangrove map. The dataset that we have were first pre-processed using the LAStools software. It is a software that is used to process LiDAR data sets and create different layers such as DSM, DTM, nDSM, Slope, LiDAR Intensity, LiDAR number of first returns, and CHM. All the aforementioned layers together was used to derive the Mangrove class. Then, an Object-based Image Analysis (OBIA) was performed using eCognition. OBIA analyzes a group of pixels with similar properties called objects, as compared to the traditional pixel-based which only examines a single pixel. Multi-threshold and multiresolution segmentation were used to delineate the different classes and split the image into objects. There are four levels of classification, first is the separation of the Land from the Water. Then the Land class was further dived into Ground and Non-ground objects. Furthermore classification of Nonvegetation, Mangroves, and Other Vegetation was done from the Non-ground objects. Lastly Separation of the mangrove class was done through the Use of field verified training points which was then run into a Support Vector Machine (SVM) classification. Different classes were separated using the different layer feature properties, such as mean, mode, standard deviation, geometrical properties, neighbor-related properties, and textural properties. Accuracy assessment was done using a different set of field validation points. This workflow was applied in the classification of Mangroves to a LiDAR dataset of Naawan and Manticao, Misamis Oriental, Philippines. The process presented in this study shows that LiDAR data and its derivatives can be used in extracting and creating Mangrove maps, which can be helpful in managing coastal environment.

Implications of sensor design for coral reef detection: Upscaling ground hyperspectral imagery in spatial and spectral scales

NASA Astrophysics Data System (ADS)

Caras, Tamir; Hedley, John; Karnieli, Arnon

2017-12-01

Remote sensing offers a potential tool for large scale environmental surveying and monitoring. However, remote observations of coral reefs are difficult especially due to the spatial and spectral complexity of the target compared to sensor specifications as well as the environmental implications of the water medium above. The development of sensors is driven by technological advances and the desired products. Currently, spaceborne systems are technologically limited to a choice between high spectral resolution and high spatial resolution, but not both. The current study explores the dilemma of whether future sensor design for marine monitoring should prioritise on improving their spatial or spectral resolution. To address this question, a spatially and spectrally resampled ground-level hyperspectral image was used to test two classification elements: (1) how the tradeoff between spatial and spectral resolutions affects classification; and (2) how a noise reduction by majority filter might improve classification accuracy. The studied reef, in the Gulf of Aqaba (Eilat), Israel, is heterogeneous and complex so the local substrate patches are generally finer than currently available imagery. Therefore, the tested spatial resolution was broadly divided into four scale categories from five millimeters to one meter. Spectral resolution resampling aimed to mimic currently available and forthcoming spaceborne sensors such as (1) Environmental Mapping and Analysis Program (EnMAP) that is characterized by 25 bands of 6.5 nm width; (2) VENμS with 12 narrow bands; and (3) the WorldView series with broadband multispectral resolution. Results suggest that spatial resolution should generally be prioritized for coral reef classification because the finer spatial scale tested (pixel size < 0.1 m) may compensate for some low spectral resolution drawbacks. In this regard, it is shown that the post-classification majority filtering substantially improves the accuracy of all pixel sizes up to the point where the kernel size reaches the average unit size (pixel < 0.25 m). However, careful investigation as to the effect of band distribution and choice could improve the sensor suitability for the marine environment task. This in mind, while the focus in this study was on the technologically limited spaceborne design, aerial sensors may presently provide an opportunity to implement the suggested setup.

Effects of Digitization and JPEG Compression on Land Cover Classification Using Astronaut-Acquired Orbital Photographs

NASA Technical Reports Server (NTRS)

Robinson, Julie A.; Webb, Edward L.; Evangelista, Arlene

2000-01-01

Studies that utilize astronaut-acquired orbital photographs for visual or digital classification require high-quality data to ensure accuracy. The majority of images available must be digitized from film and electronically transferred to scientific users. This study examined the effect of scanning spatial resolution (1200, 2400 pixels per inch [21.2 and 10.6 microns/pixel]), scanning density range option (Auto, Full) and compression ratio (non-lossy [TIFF], and lossy JPEG 10:1, 46:1, 83:1) on digital classification results of an orbital photograph from the NASA - Johnson Space Center archive. Qualitative results suggested that 1200 ppi was acceptable for visual interpretive uses for major land cover types. Moreover, Auto scanning density range was superior to Full density range. Quantitative assessment of the processing steps indicated that, while 2400 ppi scanning spatial resolution resulted in more classified polygons as well as a substantially greater proportion of polygons < 0.2 ha, overall agreement between 1200 ppi and 2400 ppi was quite high. JPEG compression up to approximately 46:1 also did not appear to have a major impact on quantitative classification characteristics. We conclude that both 1200 and 2400 ppi scanning resolutions are acceptable options for this level of land cover classification, as well as a compression ratio at or below approximately 46:1. Auto range density should always be used during scanning because it acquires more of the information from the film. The particular combination of scanning spatial resolution and compression level will require a case-by-case decision and will depend upon memory capabilities, analytical objectives and the spatial properties of the objects in the image.

Detection of Aspens Using High Resolution Aerial Laser Scanning Data and Digital Aerial Images

PubMed Central

Säynäjoki, Raita; Packalén, Petteri; Maltamo, Matti; Vehmas, Mikko; Eerikäinen, Kalle

2008-01-01

The aim was to use high resolution Aerial Laser Scanning (ALS) data and aerial images to detect European aspen (Populus tremula L.) from among other deciduous trees. The field data consisted of 14 sample plots of 30 m × 30 m size located in the Koli National Park in the North Karelia, Eastern Finland. A Canopy Height Model (CHM) was interpolated from the ALS data with a pulse density of 3.86/m2, low-pass filtered using Height-Based Filtering (HBF) and binarized to create the mask needed to separate the ground pixels from the canopy pixels within individual areas. Watershed segmentation was applied to the low-pass filtered CHM in order to create preliminary canopy segments, from which the non-canopy elements were extracted to obtain the final canopy segmentation, i.e. the ground mask was analysed against the canopy mask. A manual classification of aerial images was employed to separate the canopy segments of deciduous trees from those of coniferous trees. Finally, linear discriminant analysis was applied to the correctly classified canopy segments of deciduous trees to classify them into segments belonging to aspen and those belonging to other deciduous trees. The independent variables used in the classification were obtained from the first pulse ALS point data. The accuracy of discrimination between aspen and other deciduous trees was 78.6%. The independent variables in the classification function were the proportion of vegetation hits, the standard deviation of in pulse heights, accumulated intensity at the 90th percentile and the proportion of laser points reflected at the 60th height percentile. The accuracy of classification corresponded to the validation results of earlier ALS-based studies on the classification of individual deciduous trees to tree species. PMID:27873799

Object-Based Classification and Change Detection of Hokkaido, Japan

NASA Astrophysics Data System (ADS)

Park, J. G.; Harada, I.; Kwak, Y.

2016-06-01

Topography and geology are factors to characterize the distribution of natural vegetation. Topographic contour is particularly influential on the living conditions of plants such as soil moisture, sunlight, and windiness. Vegetation associations having similar characteristics are present in locations having similar topographic conditions unless natural disturbances such as landslides and forest fires or artificial disturbances such as deforestation and man-made plantation bring about changes in such conditions. We developed a vegetation map of Japan using an object-based segmentation approach with topographic information (elevation, slope, slope direction) that is closely related to the distribution of vegetation. The results found that the object-based classification is more effective to produce a vegetation map than the pixel-based classification.

The effect of spatial, spectral and radiometric factors on classification accuracy using thematic mapper data

NASA Technical Reports Server (NTRS)

Wrigley, R. C.; Acevedo, W.; Alexander, D.; Buis, J.; Card, D.

1984-01-01

An experiment of a factorial design was conducted to test the effects on classification accuracy of land cover types due to the improved spatial, spectral and radiometric characteristics of the Thematic Mapper (TM) in comparison to the Multispectral Scanner (MSS). High altitude aircraft scanner data from the Airborne Thematic Mapper instrument was acquired over central California in August, 1983 and used to simulate Thematic Mapper data as well as all combinations of the three characteristics for eight data sets in all. Results for the training sites (field center pixels) showed better classification accuracies for MSS spatial resolution, TM spectral bands and TM radiometry in order of importance.

Spatial Mutual Information Based Hyperspectral Band Selection for Classification

PubMed Central

2015-01-01

The amount of information involved in hyperspectral imaging is large. Hyperspectral band selection is a popular method for reducing dimensionality. Several information based measures such as mutual information have been proposed to reduce information redundancy among spectral bands. Unfortunately, mutual information does not take into account the spatial dependency between adjacent pixels in images thus reducing its robustness as a similarity measure. In this paper, we propose a new band selection method based on spatial mutual information. As validation criteria, a supervised classification method using support vector machine (SVM) is used. Experimental results of the classification of hyperspectral datasets show that the proposed method can achieve more accurate results. PMID:25918742

Lossless Compression of Classification-Map Data

NASA Technical Reports Server (NTRS)

Hua, Xie; Klimesh, Matthew

2009-01-01

A lossless image-data-compression algorithm intended specifically for application to classification-map data is based on prediction, context modeling, and entropy coding. The algorithm was formulated, in consideration of the differences between classification maps and ordinary images of natural scenes, so as to be capable of compressing classification- map data more effectively than do general-purpose image-data-compression algorithms. Classification maps are typically generated from remote-sensing images acquired by instruments aboard aircraft (see figure) and spacecraft. A classification map is a synthetic image that summarizes information derived from one or more original remote-sensing image(s) of a scene. The value assigned to each pixel in such a map is the index of a class that represents some type of content deduced from the original image data for example, a type of vegetation, a mineral, or a body of water at the corresponding location in the scene. When classification maps are generated onboard the aircraft or spacecraft, it is desirable to compress the classification-map data in order to reduce the volume of data that must be transmitted to a ground station.

Classification of weld defect based on information fusion technology for radiographic testing system

DOE Office of Scientific and Technical Information (OSTI.GOV)

Jiang, Hongquan; Liang, Zeming, E-mail: heavenlzm@126.com; Gao, Jianmin

Improving the efficiency and accuracy of weld defect classification is an important technical problem in developing the radiographic testing system. This paper proposes a novel weld defect classification method based on information fusion technology, Dempster–Shafer evidence theory. First, to characterize weld defects and improve the accuracy of their classification, 11 weld defect features were defined based on the sub-pixel level edges of radiographic images, four of which are presented for the first time in this paper. Second, we applied information fusion technology to combine different features for weld defect classification, including a mass function defined based on the weld defectmore » feature information and the quartile-method-based calculation of standard weld defect class which is to solve a sample problem involving a limited number of training samples. A steam turbine weld defect classification case study is also presented herein to illustrate our technique. The results show that the proposed method can increase the correct classification rate with limited training samples and address the uncertainties associated with weld defect classification.« less

Classification of weld defect based on information fusion technology for radiographic testing system.

PubMed

Jiang, Hongquan; Liang, Zeming; Gao, Jianmin; Dang, Changying

2016-03-01

Improving the efficiency and accuracy of weld defect classification is an important technical problem in developing the radiographic testing system. This paper proposes a novel weld defect classification method based on information fusion technology, Dempster-Shafer evidence theory. First, to characterize weld defects and improve the accuracy of their classification, 11 weld defect features were defined based on the sub-pixel level edges of radiographic images, four of which are presented for the first time in this paper. Second, we applied information fusion technology to combine different features for weld defect classification, including a mass function defined based on the weld defect feature information and the quartile-method-based calculation of standard weld defect class which is to solve a sample problem involving a limited number of training samples. A steam turbine weld defect classification case study is also presented herein to illustrate our technique. The results show that the proposed method can increase the correct classification rate with limited training samples and address the uncertainties associated with weld defect classification.

Ocean Thermal Feature Recognition, Discrimination and Tracking Using Infrared Satellite Imagery

DTIC Science & Technology

1991-06-01

rejected if the temperature in the mapped area exceeds classification criteria ............................... 17 viii 2.6 Ideal feature space mapping from...in seconds, and 1P is the side dimension of the pixel in meters. Figure 2.6: Ideal feature space mapping from pattern tile - search tile comparison. 20

A HYBRID HIGH RESOLUTION IMAGE CLASSIFICATION METHOD FOR MAPPING EELGRASS DISTRIBUTIONS IN YAQUINA BAY ESTUARY, OREGON

EPA Science Inventory

False-color infrared aerial photography of the Yaquina Bay Estuary, Oregon was acquired at extreme low tides and digitally orthorectified with a ground pixel resolution of 20 cm to provide data for intertidal vegetation mapping. Submerged, semi-exposed and exposed eelgrass mead...

Mapping forest tree species over large areas with partially cloudy Landsat imagery

NASA Astrophysics Data System (ADS)

Turlej, K.; Radeloff, V.

2017-12-01

Forests provide numerous services to natural systems and humankind, but which services forest provide depends greatly on their tree species composition. That makes it important to track not only changes in forest extent, something that remote sensing excels in, but also to map tree species. The main goal of our work was to map tree species with Landsat imagery, and to identify how to maximize mapping accuracy by including partially cloudy imagery. Our study area covered one Landsat footprint (26/28) in Northern Wisconsin, USA, with temperate and boreal forests. We selected this area because it contains numerous tree species and variable forest composition providing an ideal study area to test the limits of Landsat data. We quantified how species-level classification accuracy was affected by a) the number of acquisitions, b) the seasonal distribution of observations, and c) the amount of cloud contamination. We classified a single year stack of Landsat-7, and -8 images data with a decision tree algorithm to generate a map of dominant tree species at the pixel- and stand-level. We obtained three important results. First, we achieved producer's accuracies in the range 70-80% and user's accuracies in range 80-90% for the most abundant tree species in our study area. Second, classification accuracy improved with more acquisitions, when observations were available from all seasons, and is the best when images with up to 40% cloud cover are included. Finally, classifications for pure stands were 10 to 30 percentage points better than those for mixed stands. We conclude that including partially cloudy Landsat imagery allows to map forest tree species with accuracies that were previously only possible for rare years with many cloud-free observations. Our approach thus provides important information for both forest management and science.

Assessment of landscape diversity and determination of landscape hotspots - a case of Slovenia

NASA Astrophysics Data System (ADS)

Perko, Drago; Ciglič, Rok; Hrvatin, Mauro

2017-04-01

Areas with high landscape diversity can be regarded as landscape hotspots, and vice versa areas with low landscape diversity can be marked as landscape coldspots. The main purpose of this paper is to use quantitative geoinformatical approach and identify parts of our test area (the country of Slovenia) that can be described as very diverse according to natural landscapes and natural elements. We used different digital raster data of natural elements and landscape classifications and defined landscape diversity and landscape hotspots. We defined diversity for each raster pixel by counting the number of different unique types of landscape elements and types of landscapes in its neighborhood. Namely, the method was used separately to define diversity according to natural elements (types of relief forms, rocks, and vegetation) and diversity according to existing geographical landscape classifications of Slovenia (types of landscapes). In both cases one-tenth of Slovenia's surface with the highest landscape diversity was defined as landscape hotspots. The same applies to the coldspots. Additionally we tested the same method of counting different types of landscapes in certain radius also for the area of Europe in order to find areas that are more diverse at continental level. By doing so we were able to find areas that have similar level of diversity as Slovenia according to different European landscape classifications. Areas with landscape diversity may have an advantage in economic development, especially in tourism. Such areas are also important for biodiversity, habitat, and species diversity. On the other hand, localities where various natural influences mix can also be areas where it is hard to transfer best practices from one place to another because of the varying responses of the landscapes to human intervention. Thus it is important to know where areas with high landscape diversity are.

A higher order conditional random field model for simultaneous classification of land cover and land use

NASA Astrophysics Data System (ADS)

Albert, Lena; Rottensteiner, Franz; Heipke, Christian

2017-08-01

We propose a new approach for the simultaneous classification of land cover and land use considering spatial as well as semantic context. We apply a Conditional Random Fields (CRF) consisting of a land cover and a land use layer. In the land cover layer of the CRF, the nodes represent super-pixels; in the land use layer, the nodes correspond to objects from a geospatial database. Intra-layer edges of the CRF model spatial dependencies between neighbouring image sites. All spatially overlapping sites in both layers are connected by inter-layer edges, which leads to higher order cliques modelling the semantic relation between all land cover and land use sites in the clique. A generic formulation of the higher order potential is proposed. In order to enable efficient inference in the two-layer higher order CRF, we propose an iterative inference procedure in which the two classification tasks mutually influence each other. We integrate contextual relations between land cover and land use in the classification process by using contextual features describing the complex dependencies of all nodes in a higher order clique. These features are incorporated in a discriminative classifier, which approximates the higher order potentials during the inference procedure. The approach is designed for input data based on aerial images. Experiments are carried out on two test sites to evaluate the performance of the proposed method. The experiments show that the classification results are improved compared to the results of a non-contextual classifier. For land cover classification, the result is much more homogeneous and the delineation of land cover segments is improved. For the land use classification, an improvement is mainly achieved for land use objects showing non-typical characteristics or similarities to other land use classes. Furthermore, we have shown that the size of the super-pixels has an influence on the level of detail of the classification result, but also on the degree of smoothing induced by the segmentation method, which is especially beneficial for land cover classes covering large, homogeneous areas.

Probabilistic multi-resolution human classification

NASA Astrophysics Data System (ADS)

Tu, Jun; Ran, H.

2006-02-01

Recently there has been some interest in using infrared cameras for human detection because of the sharply decreasing prices of infrared cameras. The training data used in our work for developing the probabilistic template consists images known to contain humans in different poses and orientation but having the same height. Multiresolution templates are performed. They are based on contour and edges. This is done so that the model does not learn the intensity variations among the background pixels and intensity variations among the foreground pixels. Each template at every level is then translated so that the centroid of the non-zero pixels matches the geometrical center of the image. After this normalization step, for each pixel of the template, the probability of it being pedestrian is calculated based on the how frequently it appears as 1 in the training data. We also use periodicity gait to verify the pedestrian in a Bayesian manner for the whole blob in a probabilistic way. The videos had quite a lot of variations in the scenes, sizes of people, amount of occlusions and clutter in the backgrounds as is clearly evident. Preliminary experiments show the robustness.

Parameterization of Shape and Compactness in Object-based Image Classification Using Quickbird-2 Imagery

NASA Astrophysics Data System (ADS)

Tonbul, H.; Kavzoglu, T.

2016-12-01

In recent years, object based image analysis (OBIA) has spread out and become a widely accepted technique for the analysis of remotely sensed data. OBIA deals with grouping pixels into homogenous objects based on spectral, spatial and textural features of contiguous pixels in an image. The first stage of OBIA, named as image segmentation, is the most prominent part of object recognition. In this study, multiresolution segmentation, which is a region-based approach, was employed to construct image objects. In the application of multi-resolution, three parameters, namely shape, compactness and scale must be set by the analyst. Segmentation quality remarkably influences the fidelity of the thematic maps and accordingly the classification accuracy. Therefore, it is of great importance to search and set optimal values for the segmentation parameters. In the literature, main focus has been on the definition of scale parameter, assuming that the effect of shape and compactness parameters is limited in terms of achieved classification accuracy. The aim of this study is to deeply analyze the influence of shape/compactness parameters by varying their values while using the optimal scale parameter determined by the use of Estimation of Scale Parameter (ESP-2) approach. A pansharpened Qickbird-2 image covering Trabzon, Turkey was employed to investigate the objectives of the study. For this purpose, six different combinations of shape/compactness were utilized to make deductions on the behavior of shape and compactness parameters and optimal setting for all parameters as a whole. Objects were assigned to classes using nearest neighbor classifier in all segmentation observations and equal number of pixels was randomly selected to calculate accuracy metrics. The highest overall accuracy (92.3%) was achieved by setting the shape/compactness criteria to 0.3/0.3. The results of this study indicate that shape/compactness parameters can have significant effect on classification accuracy with 4% change in overall accuracy. Also, statistical significance of differences in accuracy was tested using the McNemar's test and found that the difference between poor and optimal setting of shape/compactness parameters was statistically significant, suggesting a search for optimal parameterization instead of default setting.

Computational Short-cutting the Big Data Classification Bottleneck: Using the MODIS Land Cover Product to Derive a Consistent 30 m Landsat Land Cover Product of the Conterminous United States

NASA Astrophysics Data System (ADS)

Zhang, H.; Roy, D. P.

2016-12-01

Classification is a fundamental process in remote sensing used to relate pixel values to land cover classes present on the surface. The state of the practice for large area land cover classification is to classify satellite time series metrics with a supervised (i.e., training data dependent) non-parametric classifier. Classification accuracy generally increases with training set size. However, training data collection is expensive and the optimal training distribution over large areas is unknown. The MODIS 500 m land cover product is available globally on an annual basis and so provides a potentially very large source of land cover training data. A novel methodology to classify large volume Landsat data using high quality training data derived automatically from the MODIS land cover product is demonstrated for all of the Conterminous United States (CONUS). The known misclassification accuracy of the MODIS land cover product and the scale difference between the 500 m MODIS and 30 m Landsat data are accommodated for by a novel MODIS product filtering, Landsat pixel selection, and iterative training approach to balance the proportion of local and CONUS training data used. Three years of global Web-enabled Landsat data (WELD) data for all of the CONUS are classified using a random forest classifier and the results assessed using random forest `out-of-bag' training samples. The global WELD data are corrected to surface nadir BRDF-Adjusted Reflectance and are defined in 158 × 158 km tiles in the same projection and nested to the MODIS land cover products. This reduces the need to pre-process the considerable Landsat data volume (more than 14,000 Landsat 5 and 7 scenes per year over the CONUS covering 11,000 million 30 m pixels). The methodology is implemented in a parallel manner on WELD tile by tile basis but provides a wall-to-wall seamless 30 m land cover product. Detailed tile and CONUS results are presented and the potential for global production using the recently available global WELD products are discussed.

Early breast tumor and late SARS detections using space-variant multispectral infrared imaging at a single pixel

NASA Astrophysics Data System (ADS)

Szu, Harold H.; Buss, James R.; Kopriva, Ivica

2004-04-01

We proposed the physics approach to solve a physical inverse problem, namely to choose the unique equilibrium solution (at the minimum free energy: H= E - ToS, including the Wiener, l.m.s E, and ICA, Max S, as special cases). The "unsupervised classification" presumes that required information must be learned and derived directly and solely from the data alone, in consistence with the classical Duda-Hart ATR definition of the "unlabelled data". Such truly unsupervised methodology is presented for space-variant imaging processing for a single pixel in the real world case of remote sensing, early tumor detections and SARS. The indeterminacy of the multiple solutions of the inverse problem is regulated or selected by means of the absolute minimum of isothermal free energy as the ground truth of local equilibrium condition at the single-pixel foot print.

Object-Based Random Forest Classification of Land Cover from Remotely Sensed Imagery for Industrial and Mining Reclamation

NASA Astrophysics Data System (ADS)

Chen, Y.; Luo, M.; Xu, L.; Zhou, X.; Ren, J.; Zhou, J.

2018-04-01

The RF method based on grid-search parameter optimization could achieve a classification accuracy of 88.16 % in the classification of images with multiple feature variables. This classification accuracy was higher than that of SVM and ANN under the same feature variables. In terms of efficiency, the RF classification method performs better than SVM and ANN, it is more capable of handling multidimensional feature variables. The RF method combined with object-based analysis approach could highlight the classification accuracy further. The multiresolution segmentation approach on the basis of ESP scale parameter optimization was used for obtaining six scales to execute image segmentation, when the segmentation scale was 49, the classification accuracy reached the highest value of 89.58 %. The classification accuracy of object-based RF classification was 1.42 % higher than that of pixel-based classification (88.16 %), and the classification accuracy was further improved. Therefore, the RF classification method combined with object-based analysis approach could achieve relatively high accuracy in the classification and extraction of land use information for industrial and mining reclamation areas. Moreover, the interpretation of remotely sensed imagery using the proposed method could provide technical support and theoretical reference for remotely sensed monitoring land reclamation.

Classifying environmentally significant urban land uses with satellite imagery.

PubMed

Park, Mi-Hyun; Stenstrom, Michael K

2008-01-01

We investigated Bayesian networks to classify urban land use from satellite imagery. Landsat Enhanced Thematic Mapper Plus (ETM(+)) images were used for the classification in two study areas: (1) Marina del Rey and its vicinity in the Santa Monica Bay Watershed, CA and (2) drainage basins adjacent to the Sweetwater Reservoir in San Diego, CA. Bayesian networks provided 80-95% classification accuracy for urban land use using four different classification systems. The classifications were robust with small training data sets with normal and reduced radiometric resolution. The networks needed only 5% of the total data (i.e., 1500 pixels) for sample size and only 5- or 6-bit information for accurate classification. The network explicitly showed the relationship among variables from its structure and was also capable of utilizing information from non-spectral data. The classification can be used to provide timely and inexpensive land use information over large areas for environmental purposes such as estimating stormwater pollutant loads.

Exclusion of overlapping symptoms in DSM-5 mixed features specifier: heuristic diagnostic and treatment implications.

PubMed

Malhi, Gin S; Byrow, Yulisha; Outhred, Tim; Fritz, Kristina

2017-04-01

This article focuses on the controversial decision to exclude the overlapping symptoms of distractibility, irritability, and psychomotor agitation (DIP) with the introduction of the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) mixed features specifier. In order to understand the placement of mixed states within the current classification system, we first review the evolution of mixed states. Then, using Kraepelin's original classification of mixed states, we compare and contrast his conceptualization with modern day definitions. The DSM-5 workgroup excluded DIP symptoms, arguing that they lack the ability to differentiate between manic and depressive states; however, accumulating evidence suggests that DIP symptoms may be core features of mixed states. We suggest a return to a Kraepelinian approach to classification-with mood, ideation, and activity as key axes-and reintegration of DIP symptoms as features that are expressed across presentations. An inclusive definition of mixed states is urgently needed to resolve confusion in clinical practice and to redirect future research efforts.

Particle tracking with a Timepix based triple GEM detector

NASA Astrophysics Data System (ADS)

George, S. P.; Murtas, F.; Alozy, J.; Curioni, A.; Rosenfeld, A. B.; Silari, M.

2015-11-01

This paper details the response of a triple GEM detector with a 55 μmetre pitch pixelated ASIC for readout. The detector is operated as a micro TPC with 9.5 cm3 sensitive volume and characterized with a mixed beam of 120 GeV protons and positive pions. A process for reconstruction of incident particle tracks from individual ionization clusters is described and scans of the gain and drift fields are performed. The angular resolution of the measured tracks is characterized. Also, the readout was operated in a mixed mode where some pixels measure drift time and others charge. This was used to measure the energy deposition in the detector and the charge cloud size as a function of interaction depth. The future uses of the device, including in microdosimetry are discussed.

Toward VIP-PIX: A Low Noise Readout ASIC for Pixelated CdTe Gamma-Ray Detectors for Use in the Next Generation of PET Scanners.

PubMed

Macias-Montero, Jose-Gabriel; Sarraj, Maher; Chmeissani, Mokhtar; Puigdengoles, Carles; Lorenzo, Gianluca De; Martínez, Ricardo

2013-08-01

VIP-PIX will be a low noise and low power pixel readout electronics with digital output for pixelated Cadmium Telluride (CdTe) detectors. The proposed pixel will be part of a 2D pixel-array detector for various types of nuclear medicine imaging devices such as positron-emission tomography (PET) scanners, Compton gamma cameras, and positron-emission mammography (PEM) scanners. Each pixel will include a SAR ADC that provides the energy deposited with 10-bit resolution. Simultaneously, the self-triggered pixel which will be connected to a global time-to-digital converter (TDC) with 1 ns resolution will provide the event's time stamp. The analog part of the readout chain and the ADC have been fabricated with TSMC 0.25 μ m mixed-signal CMOS technology and characterized with an external test pulse. The power consumption of these parts is 200 μ W from a 2.5 V supply. It offers 4 switchable gains from ±10 mV/fC to ±40 mV/fC and an input charge dynamic range of up to ±70 fC for the minimum gain for both polarities. Based on noise measurements, the expected equivalent noise charge (ENC) is 65 e - RMS at room temperature.

Terrestrial hyperspectral image shadow restoration through fusion with terrestrial lidar

NASA Astrophysics Data System (ADS)

Hartzell, Preston J.; Glennie, Craig L.; Finnegan, David C.; Hauser, Darren L.

2017-05-01

Recent advances in remote sensing technology have expanded the acquisition and fusion of active lidar and passive hyperspectral imagery (HSI) from exclusively airborne observations to include terrestrial modalities. In contrast to airborne collection geometry, hyperspectral imagery captured from terrestrial cameras is prone to extensive solar shadowing on vertical surfaces leading to reductions in pixel classification accuracies or outright removal of shadowed areas from subsequent analysis tasks. We demonstrate the use of lidar spatial information for sub-pixel HSI shadow detection and the restoration of shadowed pixel spectra via empirical methods that utilize sunlit and shadowed pixels of similar material composition. We examine the effectiveness of radiometrically calibrated lidar intensity in identifying these similar materials in sun and shade conditions and further evaluate a restoration technique that leverages ratios derived from the overlapping lidar laser and HSI wavelengths. Simulations of multiple lidar wavelengths, i.e., multispectral lidar, indicate the potential for HSI spectral restoration that is independent of the complexity and costs associated with rigorous radiometric transfer models, which have yet to be developed for horizontal-viewing terrestrial HSI sensors. The spectral restoration performance of shadowed HSI pixels is quantified for imagery of a geologic outcrop through improvements in spectral shape, spectral scale, and HSI band correlation.

Decoding brain responses to pixelized images in the primary visual cortex: implications for visual cortical prostheses

PubMed Central

Guo, Bing-bing; Zheng, Xiao-lin; Lu, Zhen-gang; Wang, Xing; Yin, Zheng-qin; Hou, Wen-sheng; Meng, Ming

2015-01-01

Visual cortical prostheses have the potential to restore partial vision. Still limited by the low-resolution visual percepts provided by visual cortical prostheses, implant wearers can currently only “see” pixelized images, and how to obtain the specific brain responses to different pixelized images in the primary visual cortex (the implant area) is still unknown. We conducted a functional magnetic resonance imaging experiment on normal human participants to investigate the brain activation patterns in response to 18 different pixelized images. There were 100 voxels in the brain activation pattern that were selected from the primary visual cortex, and voxel size was 4 mm × 4 mm × 4 mm. Multi-voxel pattern analysis was used to test if these 18 different brain activation patterns were specific. We chose a Linear Support Vector Machine (LSVM) as the classifier in this study. The results showed that the classification accuracies of different brain activation patterns were significantly above chance level, which suggests that the classifier can successfully distinguish the brain activation patterns. Our results suggest that the specific brain activation patterns to different pixelized images can be obtained in the primary visual cortex using a 4 mm × 4 mm × 4 mm voxel size and a 100-voxel pattern. PMID:26692860

The realization of an SVGA OLED-on-silicon microdisplay driving circuit

NASA Astrophysics Data System (ADS)

Bohua, Zhao; Ran, Huang; Fei, Ma; Guohua, Xie; Zhensong, Zhang; Huan, Du; Jiajun, Luo; Yi, Zhao

2012-03-01

An 800 × 600 pixel organic light-emitting diode-on-silicon (OLEDoS) driving circuit is proposed. The pixel cell circuit utilizes a subthreshold-voltage-scaling structure which can modulate the pixel current between 170 pA and 11.4 nA. In order to keep the voltage of the column bus at a relatively high level, the sample-and-hold circuits adopt a ping-pong operation. The driving circuit is fabricated in a commercially available 0.35 μm two-poly four-metal 3.3 V mixed-signal CMOS process. The pixel cell area is 15 × 15 μm2 and the total chip occupies 15.5 × 12.3 mm2. Experimental results show that the chip can work properly at a frame frequency of 60 Hz and has a 64 grayscale (monochrome) display. The total power consumption of the chip is about 85 mW with a 3.3V supply voltage.

Recent development of feature extraction and classification multispectral/hyperspectral images: a systematic literature review

NASA Astrophysics Data System (ADS)

Setiyoko, A.; Dharma, I. G. W. S.; Haryanto, T.

2017-01-01

Multispectral data and hyperspectral data acquired from satellite sensor have the ability in detecting various objects on the earth ranging from low scale to high scale modeling. These data are increasingly being used to produce geospatial information for rapid analysis by running feature extraction or classification process. Applying the most suited model for this data mining is still challenging because there are issues regarding accuracy and computational cost. This research aim is to develop a better understanding regarding object feature extraction and classification applied for satellite image by systematically reviewing related recent research projects. A method used in this research is based on PRISMA statement. After deriving important points from trusted sources, pixel based and texture-based feature extraction techniques are promising technique to be analyzed more in recent development of feature extraction and classification.

Hyperspectral Image Classification With Markov Random Fields and a Convolutional Neural Network

NASA Astrophysics Data System (ADS)

Cao, Xiangyong; Zhou, Feng; Xu, Lin; Meng, Deyu; Xu, Zongben; Paisley, John

2018-05-01

This paper presents a new supervised classification algorithm for remotely sensed hyperspectral image (HSI) which integrates spectral and spatial information in a unified Bayesian framework. First, we formulate the HSI classification problem from a Bayesian perspective. Then, we adopt a convolutional neural network (CNN) to learn the posterior class distributions using a patch-wise training strategy to better use the spatial information. Next, spatial information is further considered by placing a spatial smoothness prior on the labels. Finally, we iteratively update the CNN parameters using stochastic gradient decent (SGD) and update the class labels of all pixel vectors using an alpha-expansion min-cut-based algorithm. Compared with other state-of-the-art methods, the proposed classification method achieves better performance on one synthetic dataset and two benchmark HSI datasets in a number of experimental settings.

Comparison of Danish dichotomous and BI-RADS classifications of mammographic density.

PubMed

Hodge, Rebecca; Hellmann, Sophie Sell; von Euler-Chelpin, My; Vejborg, Ilse; Andersen, Zorana Jovanovic

2014-06-01

In the Copenhagen mammography screening program from 1991 to 2001, mammographic density was classified either as fatty or mixed/dense. This dichotomous mammographic density classification system is unique internationally, and has not been validated before. To compare the Danish dichotomous mammographic density classification system from 1991 to 2001 with the density BI-RADS classifications, in an attempt to validate the Danish classification system. The study sample consisted of 120 mammograms taken in Copenhagen in 1991-2001, which tested false positive, and which were in 2012 re-assessed and classified according to the BI-RADS classification system. We calculated inter-rater agreement between the Danish dichotomous mammographic classification as fatty or mixed/dense and the four-level BI-RADS classification by the linear weighted Kappa statistic. Of the 120 women, 32 (26.7%) were classified as having fatty and 88 (73.3%) as mixed/dense mammographic density, according to Danish dichotomous classification. According to BI-RADS density classification, 12 (10.0%) women were classified as having predominantly fatty (BI-RADS code 1), 46 (38.3%) as having scattered fibroglandular (BI-RADS code 2), 57 (47.5%) as having heterogeneously dense (BI-RADS 3), and five (4.2%) as having extremely dense (BI-RADS code 4) mammographic density. The inter-rater variability assessed by weighted kappa statistic showed a substantial agreement (0.75). The dichotomous mammographic density classification system utilized in early years of Copenhagen's mammographic screening program (1991-2001) agreed well with the BI-RADS density classification system.

Determination of Classification Accuracy for Land Use/cover Types Using Landsat-Tm Spot-Mss and Multipolarized and Multi-Channel Synthetic Aperture Radar

NASA Astrophysics Data System (ADS)

Dondurur, Mehmet

The primary objective of this study was to determine the degree to which modern SAR systems can be used to obtain information about the Earth's vegetative resources. Information obtainable from microwave synthetic aperture radar (SAR) data was compared with that obtainable from LANDSAT-TM and SPOT data. Three hypotheses were tested: (a) Classification of land cover/use from SAR data can be accomplished on a pixel-by-pixel basis with the same overall accuracy as from LANDSAT-TM and SPOT data. (b) Classification accuracy for individual land cover/use classes will differ between sensors. (c) Combining information derived from optical and SAR data into an integrated monitoring system will improve overall and individual land cover/use class accuracies. The study was conducted with three data sets for the Sleeping Bear Dunes test site in the northwestern part of Michigan's lower peninsula, including an October 1982 LANDSAT-TM scene, a June 1989 SPOT scene and C-, L- and P-Band radar data from the Jet Propulsion Laboratory AIRSAR. Reference data were derived from the Michigan Resource Information System (MIRIS) and available color infrared aerial photos. Classification and rectification of data sets were done using ERDAS Image Processing Programs. Classification algorithms included Maximum Likelihood, Mahalanobis Distance, Minimum Spectral Distance, ISODATA, Parallelepiped, and Sequential Cluster Analysis. Classified images were rectified as necessary so that all were at the same scale and oriented north-up. Results were analyzed with contingency tables and percent correctly classified (PCC) and Cohen's Kappa (CK) as accuracy indices using CSLANT and ImagePro programs developed for this study. Accuracy analyses were based upon a 1.4 by 6.5 km area with its long axis east-west. Reference data for this subscene total 55,770 15 by 15 m pixels with sixteen cover types, including seven level III forest classes, three level III urban classes, two level II range classes, two water classes, one wetland class and one agriculture class. An initial analysis was made without correcting the 1978 MIRIS reference data to the different dates of the TM, SPOT and SAR data sets. In this analysis, highest overall classification accuracy (PCC) was 87% with the TM data set, with both SPOT and C-Band SAR at 85%, a difference statistically significant at the 0.05 level. When the reference data were corrected for land cover change between 1978 and 1991, classification accuracy with the C-Band SAR data increased to 87%. Classification accuracy differed from sensor to sensor for individual land cover classes, Combining sensors into hypothetical multi-sensor systems resulted in higher accuracies than for any single sensor. Combining LANDSAT -TM and C-Band SAR yielded an overall classification accuracy (PCC) of 92%. The results of this study indicate that C-Band SAR data provide an acceptable substitute for LANDSAT-TM or SPOT data when land cover information is desired of areas where cloud cover obscures the terrain. Even better results can be obtained by integrating TM and C-Band SAR data into a multi-sensor system.

Spectral reflectance characteristics of different snow and snow-covered land surface objects and mixed spectrum fitting

USGS Publications Warehouse

Zhang, J.-H.; Zhou, Z.-M.; Wang, P.-J.; Yao, F.-M.; Yang, L.

2011-01-01

The field spectroradiometer was used to measure spectra of different snow and snow-covered land surface objects in Beijing area. The result showed that for a pure snow spectrum, the snow reflectance peaks appeared from visible to 800 nm band locations; there was an obvious absorption valley of snow spectrum near 1030 nm wavelength. Compared with fresh snow, the reflection peaks of the old snow and melting snow showed different degrees of decline in the ranges of 300~1300, 1700~1800 and 2200~2300 nm, the lowest was from the compacted snow and frozen ice. For the vegetation and snow mixed spectral characteristics, it was indicated that the spectral reflectance increased for the snow-covered land types(including pine leaf with snow and pine leaf on snow background), due to the influence of snow background in the range of 350~1300 nm. However, the spectrum reflectance of mixed pixel remained a vegetation spectral characteristic. In the end, based on the spectrum analysis of snow, vegetation, and mixed snow/vegetation pixels, the mixed spectral fitting equations were established, and the results showed that there was good correlation between spectral curves by simulation fitting and observed ones(correlation coefficient R2=0.9509).

Impact of atmospheric correction and image filtering on hyperspectral classification of tree species using support vector machine

NASA Astrophysics Data System (ADS)

Shahriari Nia, Morteza; Wang, Daisy Zhe; Bohlman, Stephanie Ann; Gader, Paul; Graves, Sarah J.; Petrovic, Milenko

2015-01-01

Hyperspectral images can be used to identify savannah tree species at the landscape scale, which is a key step in measuring biomass and carbon, and tracking changes in species distributions, including invasive species, in these ecosystems. Before automated species mapping can be performed, image processing and atmospheric correction is often performed, which can potentially affect the performance of classification algorithms. We determine how three processing and correction techniques (atmospheric correction, Gaussian filters, and shade/green vegetation filters) affect the prediction accuracy of classification of tree species at pixel level from airborne visible/infrared imaging spectrometer imagery of longleaf pine savanna in Central Florida, United States. Species classification using fast line-of-sight atmospheric analysis of spectral hypercubes (FLAASH) atmospheric correction outperformed ATCOR in the majority of cases. Green vegetation (normalized difference vegetation index) and shade (near-infrared) filters did not increase classification accuracy when applied to large and continuous patches of specific species. Finally, applying a Gaussian filter reduces interband noise and increases species classification accuracy. Using the optimal preprocessing steps, our classification accuracy of six species classes is about 75%.

Training set size, scale, and features in Geographic Object-Based Image Analysis of very high resolution unmanned aerial vehicle imagery

NASA Astrophysics Data System (ADS)

Ma, Lei; Cheng, Liang; Li, Manchun; Liu, Yongxue; Ma, Xiaoxue

2015-04-01

Unmanned Aerial Vehicle (UAV) has been used increasingly for natural resource applications in recent years due to their greater availability and the miniaturization of sensors. In addition, Geographic Object-Based Image Analysis (GEOBIA) has received more attention as a novel paradigm for remote sensing earth observation data. However, GEOBIA generates some new problems compared with pixel-based methods. In this study, we developed a strategy for the semi-automatic optimization of object-based classification, which involves an area-based accuracy assessment that analyzes the relationship between scale and the training set size. We found that the Overall Accuracy (OA) increased as the training set ratio (proportion of the segmented objects used for training) increased when the Segmentation Scale Parameter (SSP) was fixed. The OA increased more slowly as the training set ratio became larger and a similar rule was obtained according to the pixel-based image analysis. The OA decreased as the SSP increased when the training set ratio was fixed. Consequently, the SSP should not be too large during classification using a small training set ratio. By contrast, a large training set ratio is required if classification is performed using a high SSP. In addition, we suggest that the optimal SSP for each class has a high positive correlation with the mean area obtained by manual interpretation, which can be summarized by a linear correlation equation. We expect that these results will be applicable to UAV imagery classification to determine the optimal SSP for each class.

Fusion of shallow and deep features for classification of high-resolution remote sensing images

NASA Astrophysics Data System (ADS)

Gao, Lang; Tian, Tian; Sun, Xiao; Li, Hang

2018-02-01

Effective spectral and spatial pixel description plays a significant role for the classification of high resolution remote sensing images. Current approaches of pixel-based feature extraction are of two main kinds: one includes the widelyused principal component analysis (PCA) and gray level co-occurrence matrix (GLCM) as the representative of the shallow spectral and shape features, and the other refers to the deep learning-based methods which employ deep neural networks and have made great promotion on classification accuracy. However, the former traditional features are insufficient to depict complex distribution of high resolution images, while the deep features demand plenty of samples to train the network otherwise over fitting easily occurs if only limited samples are involved in the training. In view of the above, we propose a GLCM-based convolution neural network (CNN) approach to extract features and implement classification for high resolution remote sensing images. The employment of GLCM is able to represent the original images and eliminate redundant information and undesired noises. Meanwhile, taking shallow features as the input of deep network will contribute to a better guidance and interpretability. In consideration of the amount of samples, some strategies such as L2 regularization and dropout methods are used to prevent over-fitting. The fine-tuning strategy is also used in our study to reduce training time and further enhance the generalization performance of the network. Experiments with popular data sets such as PaviaU data validate that our proposed method leads to a performance improvement compared to individual involved approaches.

A new Fourier transform based CBIR scheme for mammographic mass classification: a preliminary invariance assessment

NASA Astrophysics Data System (ADS)

Gundreddy, Rohith Reddy; Tan, Maxine; Qui, Yuchen; Zheng, Bin

2015-03-01

The purpose of this study is to develop and test a new content-based image retrieval (CBIR) scheme that enables to achieve higher reproducibility when it is implemented in an interactive computer-aided diagnosis (CAD) system without significantly reducing lesion classification performance. This is a new Fourier transform based CBIR algorithm that determines image similarity of two regions of interest (ROI) based on the difference of average regional image pixel value distribution in two Fourier transform mapped images under comparison. A reference image database involving 227 ROIs depicting the verified soft-tissue breast lesions was used. For each testing ROI, the queried lesion center was systematically shifted from 10 to 50 pixels to simulate inter-user variation of querying suspicious lesion center when using an interactive CAD system. The lesion classification performance and reproducibility as the queried lesion center shift were assessed and compared among the three CBIR schemes based on Fourier transform, mutual information and Pearson correlation. Each CBIR scheme retrieved 10 most similar reference ROIs and computed a likelihood score of the queried ROI depicting a malignant lesion. The experimental results shown that three CBIR schemes yielded very comparable lesion classification performance as measured by the areas under ROC curves with the p-value greater than 0.498. However, the CBIR scheme using Fourier transform yielded the highest invariance to both queried lesion center shift and lesion size change. This study demonstrated the feasibility of improving robustness of the interactive CAD systems by adding a new Fourier transform based image feature to CBIR schemes.

The performance improvement of automatic classification among obstructive lung diseases on the basis of the features of shape analysis, in addition to texture analysis at HRCT

NASA Astrophysics Data System (ADS)

Lee, Youngjoo; Kim, Namkug; Seo, Joon Beom; Lee, JuneGoo; Kang, Suk Ho

2007-03-01

In this paper, we proposed novel shape features to improve classification performance of differentiating obstructive lung diseases, based on HRCT (High Resolution Computerized Tomography) images. The images were selected from HRCT images, obtained from 82 subjects. For each image, two experienced radiologists selected rectangular ROIs with various sizes (16x16, 32x32, and 64x64 pixels), representing each disease or normal lung parenchyma. Besides thirteen textural features, we employed additional seven shape features; cluster shape features, and Top-hat transform features. To evaluate the contribution of shape features for differentiation of obstructive lung diseases, several experiments were conducted with two different types of classifiers and various ROI sizes. For automated classification, the Bayesian classifier and support vector machine (SVM) were implemented. To assess the performance and cross-validation of the system, 5-folding method was used. In comparison to employing only textural features, adding shape features yields significant enhancement of overall sensitivity(5.9, 5.4, 4.4% in the Bayesian and 9.0, 7.3, 5.3% in the SVM), in the order of ROI size 16x16, 32x32, 64x64 pixels, respectively (t-test, p<0.01). Moreover, this enhancement was largely due to the improvement on class-specific sensitivity of mild centrilobular emphysema and bronchiolitis obliterans which are most hard to differentiate for radiologists. According to these experimental results, adding shape features to conventional texture features is much useful to improve classification performance of obstructive lung diseases in both Bayesian and SVM classifiers.

Analysis of AIS Data of the Recluse Oil Field, Recluse, Wyoming

NASA Technical Reports Server (NTRS)

Dykstra, J. D.; Segal, D. B.

1985-01-01

Airborne Imaging Spectrometer (AIS) data were flown over the Recluse, Wyoming oil field on September 9, 1984. Processing software was developed at Earth Satellite Corporation (EarthSat) for interactive analysis of the AIS data. EarthSat's AIS processing capabilities include destriping, solar irradiance corrections, residual calculations, geometric resampling, equal energy normalization, interactive spectral classifications and a variety of compressive algorithms to reduce the data to 8-bit format with a minimum of information loss. The in-house photolab facilities of EarthSat can routinely produce high-quality color renditions of the enhanced AIS data. A total of 80 lithologic samples were collected under the AIS flight lines. Correlation (within the atmospheric windows) between the laboratory and the AIS spectra of sample sites was generally poor. Reasonable correlation was only possible in large, freshly plowed fields. Mixed pixels and contrast between the natural and sample's surfaces were believed responsible for the poor correlation. Finally, a drift of approximately three channels was observed in the diffraction grating position within the 1.8 to 2.1 micron quadrant.

Multi-temporal sub-pixel landsat ETM+ classification of isolated wetlands in Cuyahoga County, Ohio, USA

EPA Science Inventory

The goal of this project was to determine the utility of subpixel processing of multi-temporal Landsat Enhanced Thematic Mapper Plus (ETM+) data for the detection of isolated wetlands greater than 0.50 acres in Cuyahoga County, located in the Erie Drift Plains ecoregion of northe...

Trophic classification of Tennessee Valley area reservoirs derived from LANDSAT multispectral scanner data. [Alabama, Georgia, Kentucky, Tennessee, and North Carolina

NASA Technical Reports Server (NTRS)

Meinert, D. L.; Malone, D. L.; Voss, A. W. (Principal Investigator); Scarpace, F. L.

1980-01-01

LANDSAT MSS data from four different dates were extracted from computer tapes using a semiautomated digital data handling and analysis system. Reservoirs were extracted from the surrounding land matrix by using a Band 7 density level slice of 3; and descriptive statistics to include mean, variance, and ratio between bands for each of the four bands were calculated. Significant correlations ( 0.80) were identified between the MSS statistics and many trophic indicators from ground truth water quality data collected at 35 reservoirs in the greater Tennessee Valley region. Regression models were developed which gave significant estimates of each reservoir's trophic state as defined by its trophic state index and explained in all four LANDSAT frames at least 85 percent of the variability in the data. To illustrate the spatial variations within reservoirs as well as the relative variations between reservoirs, a table look up elliptical classification was used in conjunction with each reservoir's trophic state index to classify each reservoir on a pixel by pixel basis and produce color coded thematic representations.

Per-point and per-field contextual classification of multipolarization and multiple incidence angle aircraft L-band radar data

NASA Technical Reports Server (NTRS)

Hoffer, Roger M.; Hussin, Yousif Ali

1989-01-01

Multipolarized aircraft L-band radar data are classified using two different image classification algorithms: (1) a per-point classifier, and (2) a contextual, or per-field, classifier. Due to the distinct variations in radar backscatter as a function of incidence angle, the data are stratified into three incidence-angle groupings, and training and test data are defined for each stratum. A low-pass digital mean filter with varied window size (i.e., 3x3, 5x5, and 7x7 pixels) is applied to the data prior to the classification. A predominately forested area in northern Florida was the study site. The results obtained by using these image classifiers are then presented and discussed.

The Pediatric Home Care/Expenditure Classification Model (P/ECM): A Home Care Case-Mix Model for Children Facing Special Health Care Challenges.

PubMed

Phillips, Charles D

2015-01-01

Case-mix classification and payment systems help assure that persons with similar needs receive similar amounts of care resources, which is a major equity concern for consumers, providers, and programs. Although health service programs for adults regularly use case-mix payment systems, programs providing health services to children and youth rarely use such models. This research utilized Medicaid home care expenditures and assessment data on 2,578 children receiving home care in one large state in the USA. Using classification and regression tree analyses, a case-mix model for long-term pediatric home care was developed. The Pediatric Home Care/Expenditure Classification Model (P/ECM) grouped children and youth in the study sample into 24 groups, explaining 41% of the variance in annual home care expenditures. The P/ECM creates the possibility of a more equitable, and potentially more effective, allocation of home care resources among children and youth facing serious health care challenges.

The Pediatric Home Care/Expenditure Classification Model (P/ECM): A Home Care Case-Mix Model for Children Facing Special Health Care Challenges

PubMed Central

Phillips, Charles D.

2015-01-01

Case-mix classification and payment systems help assure that persons with similar needs receive similar amounts of care resources, which is a major equity concern for consumers, providers, and programs. Although health service programs for adults regularly use case-mix payment systems, programs providing health services to children and youth rarely use such models. This research utilized Medicaid home care expenditures and assessment data on 2,578 children receiving home care in one large state in the USA. Using classification and regression tree analyses, a case-mix model for long-term pediatric home care was developed. The Pediatric Home Care/Expenditure Classification Model (P/ECM) grouped children and youth in the study sample into 24 groups, explaining 41% of the variance in annual home care expenditures. The P/ECM creates the possibility of a more equitable, and potentially more effective, allocation of home care resources among children and youth facing serious health care challenges. PMID:26740744

Regional shape-based feature space for segmenting biomedical images using neural networks

NASA Astrophysics Data System (ADS)

Sundaramoorthy, Gopal; Hoford, John D.; Hoffman, Eric A.

1993-07-01

In biomedical images, structure of interest, particularly the soft tissue structures, such as the heart, airways, bronchial and arterial trees often have grey-scale and textural characteristics similar to other structures in the image, making it difficult to segment them using only gray- scale and texture information. However, these objects can be visually recognized by their unique shapes and sizes. In this paper we discuss, what we believe to be, a novel, simple scheme for extracting features based on regional shapes. To test the effectiveness of these features for image segmentation (classification), we use an artificial neural network and a statistical cluster analysis technique. The proposed shape-based feature extraction algorithm computes regional shape vectors (RSVs) for all pixels that meet a certain threshold criteria. The distance from each such pixel to a boundary is computed in 8 directions (or in 26 directions for a 3-D image). Together, these 8 (or 26) values represent the pixel's (or voxel's) RSV. All RSVs from an image are used to train a multi-layered perceptron neural network which uses these features to 'learn' a suitable classification strategy. To clearly distinguish the desired object from other objects within an image, several examples from inside and outside the desired object are used for training. Several examples are presented to illustrate the strengths and weaknesses of our algorithm. Both synthetic and actual biomedical images are considered. Future extensions to this algorithm are also discussed.

Superpixel-based classification of gastric chromoendoscopy images

NASA Astrophysics Data System (ADS)

Boschetto, Davide; Grisan, Enrico

2017-03-01

Chromoendoscopy (CH) is a gastroenterology imaging modality that involves the staining of tissues with methylene blue, which reacts with the internal walls of the gastrointestinal tract, improving the visual contrast in mucosal surfaces and thus enhancing a doctor's ability to screen precancerous lesions or early cancer. This technique helps identify areas that can be targeted for biopsy or treatment and in this work we will focus on gastric cancer detection. Gastric chromoendoscopy for cancer detection has several taxonomies available, one of which classifies CH images into three classes (normal, metaplasia, dysplasia) based on color, shape and regularity of pit patterns. Computer-assisted diagnosis is desirable to help us improve the reliability of the tissue classification and abnormalities detection. However, traditional computer vision methodologies, mainly segmentation, do not translate well to the specific visual characteristics of a gastroenterology imaging scenario. We propose the exploitation of a first unsupervised segmentation via superpixel, which groups pixels into perceptually meaningful atomic regions, used to replace the rigid structure of the pixel grid. For each superpixel, a set of features is extracted and then fed to a random forest based classifier, which computes a model used to predict the class of each superpixel. The average general accuracy of our model is 92.05% in the pixel domain (86.62% in the superpixel domain), while detection accuracies on the normal and abnormal class are respectively 85.71% and 95%. Eventually, the whole image class can be predicted image through a majority vote on each superpixel's predicted class.

An Investigation of Automatic Change Detection for Topographic Map Updating

NASA Astrophysics Data System (ADS)

Duncan, P.; Smit, J.

2012-08-01

Changes to the landscape are constantly occurring and it is essential for geospatial and mapping organisations that these changes are regularly detected and captured, so that map databases can be updated to reflect the current status of the landscape. The Chief Directorate of National Geospatial Information (CD: NGI), South Africa's national mapping agency, currently relies on manual methods of detecting changes and capturing these changes. These manual methods are time consuming and labour intensive, and rely on the skills and interpretation of the operator. It is therefore necessary to move towards more automated methods in the production process at CD: NGI. The aim of this research is to do an investigation into a methodology for automatic or semi-automatic change detection for the purpose of updating topographic databases. The method investigated for detecting changes is through image classification as well as spatial analysis and is focussed on urban landscapes. The major data input into this study is high resolution aerial imagery and existing topographic vector data. Initial results indicate the traditional pixel-based image classification approaches are unsatisfactory for large scale land-use mapping and that object-orientated approaches hold more promise. Even in the instance of object-oriented image classification generalization of techniques on a broad-scale has provided inconsistent results. A solution may lie with a hybrid approach of pixel and object-oriented techniques.

Improving Spectral Image Classification through Band-Ratio Optimization and Pixel Clustering

NASA Astrophysics Data System (ADS)

O'Neill, M.; Burt, C.; McKenna, I.; Kimblin, C.

2017-12-01

The Underground Nuclear Explosion Signatures Experiment (UNESE) seeks to characterize non-prompt observables from underground nuclear explosions (UNE). As part of this effort, we evaluated the ability of DigitalGlobe's WorldView-3 (WV3) to detect and map UNE signatures. WV3 is the current state-of-the-art, commercial, multispectral imaging satellite; however, it has relatively limited spectral and spatial resolutions. These limitations impede image classifiers from detecting targets that are spatially small and lack distinct spectral features. In order to improve classification results, we developed custom algorithms to reduce false positive rates while increasing true positive rates via a band-ratio optimization and pixel clustering front-end. The clusters resulting from these algorithms were processed with standard spectral image classifiers such as Mixture-Tuned Matched Filter (MTMF) and Adaptive Coherence Estimator (ACE). WV3 and AVIRIS data of Cuprite, Nevada, were used as a validation data set. These data were processed with a standard classification approach using MTMF and ACE algorithms. They were also processed using the custom front-end prior to the standard approach. A comparison of the results shows that the custom front-end significantly increases the true positive rate and decreases the false positive rate.This work was done by National Security Technologies, LLC, under Contract No. DE-AC52-06NA25946 with the U.S. Department of Energy. DOE/NV/25946-3283.

Phase information contained in meter-scale SAR images

NASA Astrophysics Data System (ADS)

Datcu, Mihai; Schwarz, Gottfried; Soccorsi, Matteo; Chaabouni, Houda

2007-10-01

The properties of single look complex SAR satellite images have already been analyzed by many investigators. A common belief is that, apart from inverse SAR methods or polarimetric applications, no information can be gained from the phase of each pixel. This belief is based on the assumption that we obtain uniformly distributed random phases when a sufficient number of small-scale scatterers are mixed in each image pixel. However, the random phase assumption does no longer hold for typical high resolution urban remote sensing scenes, when a limited number of prominent human-made scatterers with near-regular shape and sub-meter size lead to correlated phase patterns. If the pixel size shrinks to a critical threshold of about 1 meter, the reflectance of built-up urban scenes becomes dominated by typical metal reflectors, corner-like structures, and multiple scattering. The resulting phases are hard to model, but one can try to classify a scene based on the phase characteristics of neighboring image pixels. We provide a "cooking recipe" of how to analyze existing phase patterns that extend over neighboring pixels.

Spatial and spectral simulation of LANDSAT images of agricultural areas

NASA Technical Reports Server (NTRS)

Pont, W. F., Jr. (Principal Investigator)

1982-01-01

A LANDSAT scene simulation capability was developed to study the effects of small fields and misregistration on LANDSAT-based crop proportion estimation procedures. The simulation employs a pattern of ground polygons each with a crop ID, planting date, and scale factor. Historical greenness/brightness crop development profiles generate the mean signal values for each polygon. Historical within-field covariances add texture to pixels in each polygon. The planting dates and scale factors create between-field/within-crop variation. Between field and crop variation is achieved by the above and crop profile differences. The LANDSAT point spread function is used to add correlation between nearby pixels. The next effect of the point spread function is to blur the image. Mixed pixels and misregistration are also simulated.

New SOFRADIR 10μm pixel pitch infrared products

NASA Astrophysics Data System (ADS)

Lefoul, X.; Pere-Laperne, N.; Augey, T.; Rubaldo, L.; Aufranc, Sébastien; Decaens, G.; Ricard, N.; Mazaleyrat, E.; Billon-Lanfrey, D.; Gravrand, Olivier; Bisotto, Sylvette

2014-10-01

Recent advances in miniaturization of IR imaging technology have led to a growing market for mini thermal-imaging sensors. In that respect, Sofradir development on smaller pixel pitch has made much more compact products available to the users. When this competitive advantage is mixed with smaller coolers, made possible by HOT technology, we achieved valuable reductions in the size, weight and power of the overall package. At the same time, we are moving towards a global offer based on digital interfaces that provides our customers simplifications at the IR system design process while freeing up more space. This paper discusses recent developments on hot and small pixel pitch technologies as well as efforts made on compact packaging solution developed by SOFRADIR in collaboration with CEA-LETI.

Estimating cropland NPP using national crop inventory and MODIS derived crop specific parameters

NASA Astrophysics Data System (ADS)

Bandaru, V.; West, T. O.; Ricciuto, D. M.

2011-12-01

Estimates of cropland net primary production (NPP) are needed as input for estimates of carbon flux and carbon stock changes. Cropland NPP is currently estimated using terrestrial ecosystem models, satellite remote sensing, or inventory data. All three of these methods have benefits and problems. Terrestrial ecosystem models are often better suited for prognostic estimates rather than diagnostic estimates. Satellite-based NPP estimates often underestimate productivity on intensely managed croplands and are also limited to a few broad crop categories. Inventory-based estimates are consistent with nationally collected data on crop yields, but they lack sub-county spatial resolution. Integrating these methods will allow for spatial resolution consistent with current land cover and land use, while also maintaining total biomass quantities recorded in national inventory data. The main objective of this study was to improve cropland NPP estimates by using a modification of the CASA NPP model with individual crop biophysical parameters partly derived from inventory data and MODIS 8day 250m EVI product. The study was conducted for corn and soybean crops in Iowa and Illinois for years 2006 and 2007. We used EVI as a linear function for fPAR, and used crop land cover data (56m spatial resolution) to extract individual crop EVI pixels. First, we separated mixed pixels of both corn and soybean that occur when MODIS 250m pixel contains more than one crop. Second, we substituted mixed EVI pixels with nearest pure pixel values of the same crop within 1km radius. To get more accurate photosynthetic active radiation (PAR), we applied the Mountain Climate Simulator (MTCLIM) algorithm with the use of temperature and precipitation data from the North American Land Data Assimilation System (NLDAS-2) to generate shortwave radiation data. Finally, county specific light use efficiency (LUE) values of each crop for years 2006 to 2007 were determined by application of mean county inventory NPP and EVI-derived APAR into the Monteith equation. Results indicate spatial variability in LUE values across Iowa and Illinois. Northern regions of both Iowa and Illinois have higher LUE values than southern regions. This trend is reflected in NPP estimates. Results also show that corn has higher LUE values than soybean, resulting in higher NPP for corn than for soybean. Current NPP estimates were compared with NPP estimates from MOD17A3 product and with county inventory-based NPP estimates. Results indicate that current NPP estimates closely agree with inventory-based estimates, and that current NPP estimates are higher than those of the MOD17A3 product. It was also found that when mixed pixels were substituted with nearest pure pixels, revised NPP estimates were improved showing better agreement with inventory-based estimates.

Decision Tree Repository and Rule Set Based Mingjiang River Estuarine Wetlands Classifaction

NASA Astrophysics Data System (ADS)

Zhang, W.; Li, X.; Xiao, W.

2018-05-01

The increasing urbanization and industrialization have led to wetland losses in estuarine area of Mingjiang River over past three decades. There has been increasing attention given to produce wetland inventories using remote sensing and GIS technology. Due to inconsistency training site and training sample, traditionally pixel-based image classification methods can't achieve a comparable result within different organizations. Meanwhile, object-oriented image classification technique shows grate potential to solve this problem and Landsat moderate resolution remote sensing images are widely used to fulfill this requirement. Firstly, the standardized atmospheric correct, spectrally high fidelity texture feature enhancement was conducted before implementing the object-oriented wetland classification method in eCognition. Secondly, we performed the multi-scale segmentation procedure, taking the scale, hue, shape, compactness and smoothness of the image into account to get the appropriate parameters, using the top and down region merge algorithm from single pixel level, the optimal texture segmentation scale for different types of features is confirmed. Then, the segmented object is used as the classification unit to calculate the spectral information such as Mean value, Maximum value, Minimum value, Brightness value and the Normalized value. The Area, length, Tightness and the Shape rule of the image object Spatial features and texture features such as Mean, Variance and Entropy of image objects are used as classification features of training samples. Based on the reference images and the sampling points of on-the-spot investigation, typical training samples are selected uniformly and randomly for each type of ground objects. The spectral, texture and spatial characteristics of each type of feature in each feature layer corresponding to the range of values are used to create the decision tree repository. Finally, with the help of high resolution reference images, the random sampling method is used to conduct the field investigation, achieve an overall accuracy of 90.31 %, and the Kappa coefficient is 0.88. The classification method based on decision tree threshold values and rule set developed by the repository, outperforms the results obtained from the traditional methodology. Our decision tree repository and rule set based object-oriented classification technique was an effective method for producing comparable and consistency wetlands data set.

Lagrange constraint neural network for audio varying BSS

NASA Astrophysics Data System (ADS)

Szu, Harold H.; Hsu, Charles C.

2002-03-01

Lagrange Constraint Neural Network (LCNN) is a statistical-mechanical ab-initio model without assuming the artificial neural network (ANN) model at all but derived it from the first principle of Hamilton and Lagrange Methodology: H(S,A)= f(S)- (lambda) C(s,A(x,t)) that incorporates measurement constraint C(S,A(x,t))= (lambda) ([A]S-X)+((lambda) 0-1)((Sigma) isi -1) using the vector Lagrange multiplier-(lambda) and a- priori Shannon Entropy f(S) = -(Sigma) i si log si as the Contrast function of unknown number of independent sources si. Szu et al. have first solved in 1997 the general Blind Source Separation (BSS) problem for spatial-temporal varying mixing matrix for the real world remote sensing where a large pixel footprint implies the mixing matrix [A(x,t)] necessarily fill with diurnal and seasonal variations. Because the ground truth is difficult to be ascertained in the remote sensing, we have thus illustrated in this paper, each step of the LCNN algorithm for the simulated spatial-temporal varying BSS in speech, music audio mixing. We review and compare LCNN with other popular a-posteriori Maximum Entropy methodologies defined by ANN weight matrix-[W] sigmoid-(sigma) post processing H(Y=(sigma) ([W]X)) by Bell-Sejnowski, Amari and Oja (BSAO) called Independent Component Analysis (ICA). Both are mirror symmetric of the MaxEnt methodologies and work for a constant unknown mixing matrix [A], but the major difference is whether the ensemble average is taken at neighborhood pixel data X's in BASO or at the a priori sources S variables in LCNN that dictates which method works for spatial-temporal varying [A(x,t)] that would not allow the neighborhood pixel average. We expected the success of sharper de-mixing by the LCNN method in terms of a controlled ground truth experiment in the simulation of variant mixture of two music of similar Kurtosis (15 seconds composed of Saint-Saens Swan and Rachmaninov cello concerto).

Design of the small pixel pitch ROIC

NASA Astrophysics Data System (ADS)

Liang, Qinghua; Jiang, Dazhao; Chen, Honglei; Zhai, Yongcheng; Gao, Lei; Ding, Ruijun

2014-11-01

Since the technology trend of the third generation IRFPA towards resolution enhancing has steadily progressed,the pixel pitch of IRFPA has been greatly reduced.A 640×512 readout integrated circuit(ROIC) of IRFPA with 15μm pixel pitch is presented in this paper.The 15μm pixel pitch ROIC design will face many challenges.As we all known,the integrating capacitor is a key performance parameter when considering pixel area,charge capacity and dynamic range,so we adopt the effective method of 2 by 2 pixels sharing an integrating capacitor to solve this problem.The input unit cell architecture will contain two paralleled sample and hold parts,which not only allow the FPA to be operated in full frame snapshot mode but also save relatively unit circuit area.Different applications need more matching input unit circuits. Because the dimension of 2×2 pixels is 30μm×30μm, an input stage based on direct injection (DI) which has medium injection ratio and small layout area is proved to be suitable for middle wave (MW) while BDI with three-transistor cascode amplifier for long wave(LW). By adopting the 0.35μm 2P4M mixed signal process, the circuit architecture can make the effective charge capacity of 7.8Me- per pixel with 2.2V output range for MW and 7.3 Me- per pixel with 2.6V output range for LW. According to the simulation results, this circuit works well under 5V power supply and achieves less than 0.1% nonlinearity.

Supervised nonlinear spectral unmixing using a postnonlinear mixing model for hyperspectral imagery.

PubMed

Altmann, Yoann; Halimi, Abderrahim; Dobigeon, Nicolas; Tourneret, Jean-Yves

2012-06-01

This paper presents a nonlinear mixing model for hyperspectral image unmixing. The proposed model assumes that the pixel reflectances are nonlinear functions of pure spectral components contaminated by an additive white Gaussian noise. These nonlinear functions are approximated using polynomial functions leading to a polynomial postnonlinear mixing model. A Bayesian algorithm and optimization methods are proposed to estimate the parameters involved in the model. The performance of the unmixing strategies is evaluated by simulations conducted on synthetic and real data.

GPU implementation of the simplex identification via split augmented Lagrangian

NASA Astrophysics Data System (ADS)

Sevilla, Jorge; Nascimento, José M. P.

2015-10-01

Hyperspectral imaging can be used for object detection and for discriminating between different objects based on their spectral characteristics. One of the main problems of hyperspectral data analysis is the presence of mixed pixels, due to the low spatial resolution of such images. This means that several spectrally pure signatures (endmembers) are combined into the same mixed pixel. Linear spectral unmixing follows an unsupervised approach which aims at inferring pure spectral signatures and their material fractions at each pixel of the scene. The huge data volumes acquired by such sensors put stringent requirements on processing and unmixing methods. This paper proposes an efficient implementation of a unsupervised linear unmixing method on GPUs using CUDA. The method finds the smallest simplex by solving a sequence of nonsmooth convex subproblems using variable splitting to obtain a constraint formulation, and then applying an augmented Lagrangian technique. The parallel implementation of SISAL presented in this work exploits the GPU architecture at low level, using shared memory and coalesced accesses to memory. The results herein presented indicate that the GPU implementation can significantly accelerate the method's execution over big datasets while maintaining the methods accuracy.

SVGA and XGA active matrix microdisplays for head-mounted applications

NASA Astrophysics Data System (ADS)

Alvelda, Phillip; Bolotski, Michael; Brown, Imani L.

2000-03-01

The MicroDisplay Corporation's liquid crystal on silicon (LCOS) display devices are based on the union of several technologies with the extreme integration capability of conventionally fabricated CMOS substrates. The fast liquid crystal operation modes and new scalable high-performance pixel addressing architectures presented in this paper enable substantially improved color, contrast, and brightness while still satisfying the optical, packaging, and power requirements of portable applications. The entire suite of MicroDisplay's technologies was devised to create a line of mixed-signal application-specific integrated circuits (ASICs) in single-chip display systems. Mixed-signal circuits can integrate computing, memory, and communication circuitry on the same substrate as the display drivers and pixel array for a multifunctional complete system-on-a-chip. System-on-a-chip benefits also include reduced head supported weight requirements through the elimination of off-chip drive electronics.

Techniques for delineation and portrayal of land cover types using ERTS-1 data. [Pennsylvania, Montana, and Texas

NASA Technical Reports Server (NTRS)

Mcmurtry, G. J.; Petersen, G. W. (Principal Investigator); Wilson, A. D.

1974-01-01

The author has identified the following significant results. ERTS data was used to map land cover in agricultural areas, although in some parts of Pennsylvania, with small irregular fields, many of the pixels overlap field boundaries and cause difficulties in classification. Various techniques and devices were used to display the results of these land cover analyses. The most promising approach would be a user-interactive color monitor interfaced with a large computer so that classification results could be displayed on the CRT and these results output by a hard complete copier.

Local neighborhood transition probability estimation and its use in contextual classification

NASA Technical Reports Server (NTRS)

Chittineni, C. B.

1979-01-01

The problem of incorporating spatial or contextual information into classifications is considered. A simple model that describes the spatial dependencies between the neighboring pixels with a single parameter, Theta, is presented. Expressions are derived for updating the posteriori probabilities of the states of nature of the pattern under consideration using information from the neighboring patterns, both for spatially uniform context and for Markov dependencies in terms of Theta. Techniques for obtaining the optimal value of the parameter Theta as a maximum likelihood estimate from the local neighborhood of the pattern under consideration are developed.

Sub-pixel flood inundation mapping from multispectral remotely sensed images based on discrete particle swarm optimization

NASA Astrophysics Data System (ADS)

Li, Linyi; Chen, Yun; Yu, Xin; Liu, Rui; Huang, Chang

2015-03-01

The study of flood inundation is significant to human life and social economy. Remote sensing technology has provided an effective way to study the spatial and temporal characteristics of inundation. Remotely sensed images with high temporal resolutions are widely used in mapping inundation. However, mixed pixels do exist due to their relatively low spatial resolutions. One of the most popular approaches to resolve this issue is sub-pixel mapping. In this paper, a novel discrete particle swarm optimization (DPSO) based sub-pixel flood inundation mapping (DPSO-SFIM) method is proposed to achieve an improved accuracy in mapping inundation at a sub-pixel scale. The evaluation criterion for sub-pixel inundation mapping is formulated. The DPSO-SFIM algorithm is developed, including particle discrete encoding, fitness function designing and swarm search strategy. The accuracy of DPSO-SFIM in mapping inundation at a sub-pixel scale was evaluated using Landsat ETM + images from study areas in Australia and China. The results show that DPSO-SFIM consistently outperformed the four traditional SFIM methods in these study areas. A sensitivity analysis of DPSO-SFIM was also carried out to evaluate its performances. It is hoped that the results of this study will enhance the application of medium-low spatial resolution images in inundation detection and mapping, and thereby support the ecological and environmental studies of river basins.

A Comparison of Local Variance, Fractal Dimension, and Moran's I as Aids to Multispectral Image Classification

NASA Technical Reports Server (NTRS)

Emerson, Charles W.; Sig-NganLam, Nina; Quattrochi, Dale A.

2004-01-01

The accuracy of traditional multispectral maximum-likelihood image classification is limited by the skewed statistical distributions of reflectances from the complex heterogenous mixture of land cover types in urban areas. This work examines the utility of local variance, fractal dimension and Moran's I index of spatial autocorrelation in segmenting multispectral satellite imagery. Tools available in the Image Characterization and Modeling System (ICAMS) were used to analyze Landsat 7 imagery of Atlanta, Georgia. Although segmentation of panchromatic images is possible using indicators of spatial complexity, different land covers often yield similar values of these indices. Better results are obtained when a surface of local fractal dimension or spatial autocorrelation is combined as an additional layer in a supervised maximum-likelihood multispectral classification. The addition of fractal dimension measures is particularly effective at resolving land cover classes within urbanized areas, as compared to per-pixel spectral classification techniques.

Updating and testing of a Finnish method for mixed municipal solid waste composition studies.

PubMed

Liikanen, M; Sahimaa, O; Hupponen, M; Havukainen, J; Sorvari, J; Horttanainen, M

2016-06-01

More efficient recycling of municipal solid waste (MSW) is an essential precondition for turning Europe into a circular economy. Thus, the recycling of MSW must increase significantly in several member states, including Finland. This has increased the interest in the composition of mixed MSW. Due to increased information needs, a method for mixed MSW composition studies was introduced in Finland in order to improve the national comparability of composition study results. The aim of this study was to further develop the method so that it corresponds to the information needed about the composition of mixed MSW and still works in practice. A survey and two mixed MSW composition studies were carried out in the study. According to the responses of the survey, the intensification of recycling, the landfill ban on organic waste and the producer responsibility for packaging waste have particularly influenced the need for information about the composition of mixed MSW. The share of biowaste in mixed MSW interested the respondents most. Additionally, biowaste proved to be the largest waste fraction in mixed MSW in the composition studies. It constituted over 40% of mixed MSW in both composition studies. For these reasons, the classification system of the method was updated by further defining the classifications of biowaste. The classifications of paper as well as paperboard and cardboard were also updated. The updated classification system provides more information on the share of avoidable food waste and waste materials suitable for recycling in mixed MSW. The updated method and the information gained from the composition studies are important in ensuring that the method will be adopted by municipal waste management companies and thus used widely in Finland. Copyright © 2016 Elsevier Ltd. All rights reserved.

Analysing land cover and land use change in the Matobo National Park and surroundings in Zimbabwe

NASA Astrophysics Data System (ADS)

Scharsich, Valeska; Mtata, Kupakwashe; Hauhs, Michael; Lange, Holger; Bogner, Christina

2016-04-01

Natural forests are threatened worldwide, therefore their protection in National Parks is essential. Here, we investigate how this protection status affects the land cover. To answer this question, we analyse the surface reflectance of three Landsat images of Matobo National Park and surrounding in Zimbabwe from 1989, 1998 and 2014 to detect changes in land cover in this region. To account for the rolling countryside and the resulting prominent shadows, a topographical correction of the surface reflectance was required. To infer land cover changes it is not only necessary to have some ground data for the current satellite images but also for the old ones. In particular for the older images no recent field study could help to reconstruct these data reliably. In our study we follow the idea that land cover classes of pixels in current images can be transferred to the equivalent pixels of older ones if no changes occurred meanwhile. Therefore we combine unsupervised clustering with supervised classification as follows. At first, we produce a land cover map for 2014. Secondly, we cluster the images with clara, which is similar to k-means, but suitable for large data sets. Whereby the best number of classes were determined to be 4. Thirdly, we locate unchanged pixels with change vector analysis in the images of 1989 and 1998. For these pixels we transfer the corresponding cluster label from 2014 to 1989 and 1998. Subsequently, the classified pixels serve as training data for supervised classification with random forest, which is carried out for each image separately. Finally, we derive land cover classes from the Landsat image in 2014, photographs and Google Earth and transfer them to the other two images. The resulting classes are shrub land; forest/shallow waters; bare soils/fields with some trees/shrubs; and bare light soils/rocks, fields and settlements. Subsequently the three different classifications are compared and land changes are mapped. The main changes are observable in the surroundings of the National Park, especially the common lands have lost their clear boundaries with time. In the National Park, the area of forest increases from 1989 to 2014 from 58% to 61% whereas the area of shrub land decreases by the same amount. The amount of each of the other two classes remains constant. These changes indicate an actual effect of the protection status of the National Park. In our study remote sensing data are the main source to evaluate the effects and the benefits of a protected area without on-side studies. This could be important for regions, where field studies are not possible because of insecure political conditions and only remote sensing data are available.

42 CFR 412.620 - Patient classification system.

Code of Federal Regulations, 2013 CFR

2013-10-01

... weighting factors to reflect changes in— (1) Treatment patterns; (2) Technology; (3) Number of discharges... 42 Public Health 2 2013-10-01 2013-10-01 false Patient classification system. 412.620 Section 412... rehabilitation facilities into mutually exclusive case-mix groups. (2) For purposes of this subpart, case-mix...

42 CFR 412.620 - Patient classification system.

Code of Federal Regulations, 2012 CFR

2012-10-01

... weighting factors to reflect changes in— (1) Treatment patterns; (2) Technology; (3) Number of discharges... 42 Public Health 2 2012-10-01 2012-10-01 false Patient classification system. 412.620 Section 412... rehabilitation facilities into mutually exclusive case-mix groups. (2) For purposes of this subpart, case-mix...

42 CFR 412.620 - Patient classification system.

Code of Federal Regulations, 2014 CFR

2014-10-01

... weighting factors to reflect changes in— (1) Treatment patterns; (2) Technology; (3) Number of discharges... 42 Public Health 2 2014-10-01 2014-10-01 false Patient classification system. 412.620 Section 412... rehabilitation facilities into mutually exclusive case-mix groups. (2) For purposes of this subpart, case-mix...

Acoustic⁻Seismic Mixed Feature Extraction Based on Wavelet Transform for Vehicle Classification in Wireless Sensor Networks.

PubMed

Zhang, Heng; Pan, Zhongming; Zhang, Wenna

2018-06-07

An acoustic⁻seismic mixed feature extraction method based on the wavelet coefficient energy ratio (WCER) of the target signal is proposed in this study for classifying vehicle targets in wireless sensor networks. The signal was decomposed into a set of wavelet coefficients using the à trous algorithm, which is a concise method used to implement the wavelet transform of a discrete signal sequence. After the wavelet coefficients of the target acoustic and seismic signals were obtained, the energy ratio of each layer coefficient was calculated as the feature vector of the target signals. Subsequently, the acoustic and seismic features were merged into an acoustic⁻seismic mixed feature to improve the target classification accuracy after the acoustic and seismic WCER features of the target signal were simplified using the hierarchical clustering method. We selected the support vector machine method for classification and utilized the data acquired from a real-world experiment to validate the proposed method. The calculated results show that the WCER feature extraction method can effectively extract the target features from target signals. Feature simplification can reduce the time consumption of feature extraction and classification, with no effect on the target classification accuracy. The use of acoustic⁻seismic mixed features effectively improved target classification accuracy by approximately 12% compared with either acoustic signal or seismic signal alone.

14 CFR Sec. 19-4 - Service classes.

Code of Federal Regulations, 2010 CFR

2010-01-01

... a composite of first class, coach, and mixed passenger/cargo service. The following classifications... integral part of services performed pursuant to published flight schedules. The following classifications... Classifications Sec. 19-4 Service classes. The statistical classifications are designed to reflect the operating...

14 CFR Sec. 19-4 - Service classes.

Code of Federal Regulations, 2011 CFR

2011-01-01

... a composite of first class, coach, and mixed passenger/cargo service. The following classifications... integral part of services performed pursuant to published flight schedules. The following classifications... Classifications Sec. 19-4 Service classes. The statistical classifications are designed to reflect the operating...

Analysis of lithology: Vegetation mixes in multispectral images

NASA Technical Reports Server (NTRS)

Adams, J. B.; Smith, M.; Adams, J. D.

1982-01-01

Discrimination and identification of lithologies from multispectral images is discussed. Rock/soil identification can be facilitated by removing the component of the signal in the images that is contributed by the vegetation. Mixing models were developed to predict the spectra of combinations of pure end members, and those models were refined using laboratory measurements of real mixtures. Models in use include a simple linear (checkerboard) mix, granular mixing, semi-transparent coatings, and combinations of the above. The use of interactive computer techniques that allow quick comparison of the spectrum of a pixel stack (in a multiband set) with laboratory spectra is discussed.

A multi-temporal fusion-based approach for land cover mapping in support of nuclear incident response

NASA Astrophysics Data System (ADS)

Sah, Shagan

An increasingly important application of remote sensing is to provide decision support during emergency response and disaster management efforts. Land cover maps constitute one such useful application product during disaster events; if generated rapidly after any disaster, such map products can contribute to the efficacy of the response effort. In light of recent nuclear incidents, e.g., after the earthquake/tsunami in Japan (2011), our research focuses on constructing rapid and accurate land cover maps of the impacted area in case of an accidental nuclear release. The methodology involves integration of results from two different approaches, namely coarse spatial resolution multi-temporal and fine spatial resolution imagery, to increase classification accuracy. Although advanced methods have been developed for classification using high spatial or temporal resolution imagery, only a limited amount of work has been done on fusion of these two remote sensing approaches. The presented methodology thus involves integration of classification results from two different remote sensing modalities in order to improve classification accuracy. The data used included RapidEye and MODIS scenes over the Nine Mile Point Nuclear Power Station in Oswego (New York, USA). The first step in the process was the construction of land cover maps from freely available, high temporal resolution, low spatial resolution MODIS imagery using a time-series approach. We used the variability in the temporal signatures among different land cover classes for classification. The time series-specific features were defined by various physical properties of a pixel, such as variation in vegetation cover and water content over time. The pixels were classified into four land cover classes - forest, urban, water, and vegetation - using Euclidean and Mahalanobis distance metrics. On the other hand, a high spatial resolution commercial satellite, such as RapidEye, can be tasked to capture images over the affected area in the case of a nuclear event. This imagery served as a second source of data to augment results from the time series approach. The classifications from the two approaches were integrated using an a posteriori probability-based fusion approach. This was done by establishing a relationship between the classes, obtained after classification of the two data sources. Despite the coarse spatial resolution of MODIS pixels, acceptable accuracies were obtained using time series features. The overall accuracies using the fusion-based approach were in the neighborhood of 80%, when compared with GIS data sets from New York State. This fusion thus contributed to classification accuracy refinement, with a few additional advantages, such as correction for cloud cover and providing for an approach that is robust against point-in-time seasonal anomalies, due to the inclusion of multi-temporal data. We concluded that this approach is capable of generating land cover maps of acceptable accuracy and rapid turnaround, which in turn can yield reliable estimates of crop acreage of a region. The final algorithm is part of an automated software tool, which can be used by emergency response personnel to generate a nuclear ingestion pathway information product within a few hours of data collection.

Automatic image analysis and spot classification for detection of pathogenic Escherichia coli on glass slide DNA microarrays

USDA-ARS?s Scientific Manuscript database

A computer algorithm was created to inspect scanned images from DNA microarray slides developed to rapidly detect and genotype E. Coli O157 virulent strains. The algorithm computes centroid locations for signal and background pixels in RGB space and defines a plane perpendicular to the line connect...

Improved discrimination among similar agricultural plots using red-and-green-based pseudo-colour imaging

NASA Astrophysics Data System (ADS)

Doi, Ryoichi

2016-04-01

The effects of a pseudo-colour imaging method were investigated by discriminating among similar agricultural plots in remote sensing images acquired using the Airborne Visible/Infrared Imaging Spectrometer (Indiana, USA) and the Landsat 7 satellite (Fergana, Uzbekistan), and that provided by GoogleEarth (Toyama, Japan). From each dataset, red (R)-green (G)-R-G-blue yellow (RGrgbyB), and RGrgby-1B pseudo-colour images were prepared. From each, cyan, magenta, yellow, key black, L*, a*, and b* derivative grayscale images were generated. In the Airborne Visible/Infrared Imaging Spectrometer image, pixels were selected for corn no tillage (29 pixels), corn minimum tillage (27), and soybean (34) plots. Likewise, in the Landsat 7 image, pixels representing corn (73 pixels), cotton (110), and wheat (112) plots were selected, and in the GoogleEarth image, those representing soybean (118 pixels) and rice (151) were selected. When the 14 derivative grayscale images were used together with an RGB yellow grayscale image, the overall classification accuracy improved from 74 to 94% (Airborne Visible/Infrared Imaging Spectrometer), 64 to 83% (Landsat), or 77 to 90% (GoogleEarth). As an indicator of discriminatory power, the kappa significance improved 1018-fold (Airborne Visible/Infrared Imaging Spectrometer) or greater. The derivative grayscale images were found to increase the dimensionality and quantity of data. Herein, the details of the increases in dimensionality and quantity are further analysed and discussed.

ASIC Readout Circuit Architecture for Large Geiger Photodiode Arrays

NASA Technical Reports Server (NTRS)

Vasile, Stefan; Lipson, Jerold

2012-01-01

The objective of this work was to develop a new class of readout integrated circuit (ROIC) arrays to be operated with Geiger avalanche photodiode (GPD) arrays, by integrating multiple functions at the pixel level (smart-pixel or active pixel technology) in 250-nm CMOS (complementary metal oxide semiconductor) processes. In order to pack a maximum of functions within a minimum pixel size, the ROIC array is a full, custom application-specific integrated circuit (ASIC) design using a mixed-signal CMOS process with compact primitive layout cells. The ROIC array was processed to allow assembly in bump-bonding technology with photon-counting infrared detector arrays into 3-D imaging cameras (LADAR). The ROIC architecture was designed to work with either common- anode Si GPD arrays or common-cathode InGaAs GPD arrays. The current ROIC pixel design is hardwired prior to processing one of the two GPD array configurations, and it has the provision to allow soft reconfiguration to either array (to be implemented into the next ROIC array generation). The ROIC pixel architecture implements the Geiger avalanche quenching, bias, reset, and time to digital conversion (TDC) functions in full-digital design, and uses time domain over-sampling (vernier) to allow high temporal resolution at low clock rates, increased data yield, and improved utilization of the laser beam.

Documentation and Detection of Colour Changes of Bas Relieves Using Close Range Photogrammetry

NASA Astrophysics Data System (ADS)

Malinverni, E. S.; Pierdicca, R.; Sturari, M.; Colosi, F.; Orazi, R.

2017-05-01

The digitization of complex buildings, findings or bas relieves can strongly facilitate the work of archaeologists, mainly for in depth analysis tasks. Notwithstanding, whether new visualization techniques ease the study phase, a classical naked-eye approach for determining changes or surface alteration could bring towards several drawbacks. The research work described in these pages is aimed at providing experts with a workflow for the evaluation of alterations (e.g. color decay or surface alterations), allowing a more rapid and objective monitoring of monuments. More in deep, a pipeline of work has been tested in order to evaluate the color variation between surfaces acquired at different époques. The introduction of reliable tools of change detection in the archaeological domain is needful; in fact, the most widespread practice, among archaeologists and practitioners, is to perform a traditional monitoring of surfaces that is made of three main steps: production of a hand-made map based on a subjective analysis, selection of a sub-set of regions of interest, removal of small portion of surface for in depth analysis conducted in laboratory. To overcome this risky and time consuming process, digital automatic change detection procedure represents a turning point. To do so, automatic classification has been carried out according to two approaches: a pixel-based and an object-based method. Pixel-based classification aims to identify the classes by means of the spectral information provided by each pixel belonging to the original bands. The object-based approach operates on sets of pixels (objects/regions) grouped together by means of an image segmentation technique. The methodology was tested by studying the bas-relieves of a temple located in Peru, named Huaca de la Luna. Despite the data sources were collected with unplanned surveys, the workflow proved to be a valuable solution useful to understand which are the main changes over time.

Bayesian Network Structure Learning for Urban Land Use Classification from Landsat ETM+ and Ancillary Data

NASA Astrophysics Data System (ADS)

Park, M.; Stenstrom, M. K.

2004-12-01

Recognizing urban information from the satellite imagery is problematic due to the diverse features and dynamic changes of urban landuse. The use of Landsat imagery for urban land use classification involves inherent uncertainty due to its spatial resolution and the low separability among land uses. To resolve the uncertainty problem, we investigated the performance of Bayesian networks to classify urban land use since Bayesian networks provide a quantitative way of handling uncertainty and have been successfully used in many areas. In this study, we developed the optimized networks for urban land use classification from Landsat ETM+ images of Marina del Rey area based on USGS land cover/use classification level III. The networks started from a tree structure based on mutual information between variables and added the links to improve accuracy. This methodology offers several advantages: (1) The network structure shows the dependency relationships between variables. The class node value can be predicted even with particular band information missing due to sensor system error. The missing information can be inferred from other dependent bands. (2) The network structure provides information of variables that are important for the classification, which is not available from conventional classification methods such as neural networks and maximum likelihood classification. In our case, for example, bands 1, 5 and 6 are the most important inputs in determining the land use of each pixel. (3) The networks can be reduced with those input variables important for classification. This minimizes the problem without considering all possible variables. We also examined the effect of incorporating ancillary data: geospatial information such as X and Y coordinate values of each pixel and DEM data, and vegetation indices such as NDVI and Tasseled Cap transformation. The results showed that the locational information improved overall accuracy (81%) and kappa coefficient (76%), and lowered the omission and commission errors compared with using only spectral data (accuracy 71%, kappa coefficient 62%). Incorporating DEM data did not significantly improve overall accuracy (74%) and kappa coefficient (66%) but lowered the omission and commission errors. Incorporating NDVI did not much improve the overall accuracy (72%) and k coefficient (65%). Including Tasseled Cap transformation reduced the accuracy (accuracy 70%, kappa 61%). Therefore, additional information from the DEM and vegetation indices was not useful as locational ancillary data.

Analysis of the changes in the tarcrete layer on the desert surface of Kuwait using satellite imagery and cell-based modeling

NASA Astrophysics Data System (ADS)

Al-Doasari, Ahmad E.

The 1991 Gulf War caused massive environmental damage in Kuwait. Deposition of oil and soot droplets from hundreds of burning oil-wells created a layer of tarcrete on the desert surface covering over 900 km2. This research investigates the spatial change in the tarcrete extent from 1991 to 1998 using Landsat Thematic Mapper (TM) imagery and statistical modeling techniques. The pixel structure of TM data allows the spatial analysis of the change in tarcrete extent to be conducted at the pixel (cell) level within a geographical information system (GIS). There are two components to this research. The first is a comparison of three remote sensing classification techniques used to map the tarcrete layer. The second is a spatial-temporal analysis and simulation of tarcrete changes through time. The analysis focuses on an area of 389 km2 located south of the Al-Burgan oil field. Five TM images acquired in 1991, 1993, 1994, 1995, and 1998 were geometrically and atmospherically corrected. These images were classified into six classes: oil lakes; heavy, intermediate, light, and traces of tarcrete; and sand. The classification methods tested were unsupervised, supervised, and neural network supervised (fuzzy ARTMAP). Field data of tarcrete characteristics were collected to support the classification process and to evaluate the classification accuracies. Overall, the neural network method is more accurate (60 percent) than the other two methods; both the unsupervised and the supervised classification accuracy assessments resulted in 46 percent accuracy. The five classifications were used in a lagged autologistic model to analyze the spatial changes of the tarcrete through time. The autologistic model correctly identified overall tarcrete contraction between 1991--1993 and 1995--1998. However, tarcrete contraction between 1993--1994 and 1994--1995 was less well marked, in part because of classification errors in the maps from these time periods. Initial simulations of tarcrete contraction with a cellular automaton model were not very successful. However, more accurate classifications could improve the simulations. This study illustrates how an empirical investigation using satellite images, field data, GIS, and spatial statistics can simulate dynamic land-cover change through the use of a discrete statistical and cellular automaton model.

Resampling approach for anomalous change detection

NASA Astrophysics Data System (ADS)

Theiler, James; Perkins, Simon

2007-04-01

We investigate the problem of identifying pixels in pairs of co-registered images that correspond to real changes on the ground. Changes that are due to environmental differences (illumination, atmospheric distortion, etc.) or sensor differences (focus, contrast, etc.) will be widespread throughout the image, and the aim is to avoid these changes in favor of changes that occur in only one or a few pixels. Formal outlier detection schemes (such as the one-class support vector machine) can identify rare occurrences, but will be confounded by pixels that are "equally rare" in both images: they may be anomalous, but they are not changes. We describe a resampling scheme we have developed that formally addresses both of these issues, and reduces the problem to a binary classification, a problem for which a large variety of machine learning tools have been developed. In principle, the effects of misregistration will manifest themselves as pervasive changes, and our method will be robust against them - but in practice, misregistration remains a serious issue.

Fluorescence imaging to quantify crop residue cover

NASA Technical Reports Server (NTRS)

Daughtry, C. S. T.; Mcmurtrey, J. E., III; Chappelle, E. W.

1994-01-01

Crop residues, the portion of the crop left in the field after harvest, can be an important management factor in controlling soil erosion. Methods to quantify residue cover are needed that are rapid, accurate, and objective. Scenes with known amounts of crop residue were illuminated with long wave ultraviolet (UV) radiation and fluorescence images were recorded with an intensified video camera fitted with a 453 to 488 nm band pass filter. A light colored soil and a dark colored soil were used as background for the weathered soybean stems. Residue cover was determined by counting the proportion of the pixels in the image with fluorescence values greater than a threshold. Soil pixels had the lowest gray levels in the images. The values of the soybean residue pixels spanned nearly the full range of the 8-bit video data. Classification accuracies typically were within 3(absolute units) of measured cover values. Video imaging can provide an intuitive understanding of the fraction of the soil covered by residue.

Shape, Illumination, and Reflectance from Shading

DTIC Science & Technology

2013-05-29

the global entropy of log-reflectance. 3) An “absolute” prior on reflectance which prefers to paint the scene with some colors ( white , gray, green...in log- RGB from pixel i to pixel j, and c (· ;α, σ) is the negative log-likelihood of a discrete univariate Gaussian scale mixture (GSM), parametrized...gs(R) = ∑ i ∑ j∈N(i) C (Ri −Rj ;αR, σR,ΣR) (6) Where Ri−Rj is now a 3-vector of the log- RGB differ- ences, α are mixing coefficients, σ are the

A Method for Qualitative Mapping of Thick Oil Spills Using Imaging Spectroscopy

USGS Publications Warehouse

Clark, Roger N.; Swayze, Gregg A.; Leifer, Ira; Livo, K. Eric; Lundeen, Sarah; Eastwood, Michael; Green, Robert O.; Kokaly, Raymond F.; Hoefen, Todd; Sarture, Charles; McCubbin, Ian; Roberts, Dar; Steele, Denis; Ryan, Thomas; Dominguez, Roseanne; Pearson, Neil; ,

2010-01-01

A method is described to create qualitative images of thick oil in oil spills on water using near-infrared imaging spectroscopy data. The method uses simple 'three-point-band depths' computed for each pixel in an imaging spectrometer image cube using the organic absorption features due to chemical bonds in aliphatic hydrocarbons at 1.2, 1.7, and 2.3 microns. The method is not quantitative because sub-pixel mixing and layering effects are not considered, which are necessary to make a quantitative volume estimate of oil.

Identification of cortex in magnetic resonance images

NASA Astrophysics Data System (ADS)

VanMeter, John W.; Sandon, Peter A.

1992-06-01

The overall goal of the work described here is to make available to the neurosurgeon in the operating room an on-line, three-dimensional, anatomically labeled model of the patient brain, based on pre-operative magnetic resonance (MR) images. A stereotactic operating microscope is currently in experimental use, which allows structures that have been manually identified in MR images to be made available on-line. We have been working to enhance this system by combining image processing techniques applied to the MR data with an anatomically labeled 3-D brain model developed from the Talairach and Tournoux atlas. Here we describe the process of identifying cerebral cortex in the patient MR images. MR images of brain tissue are reasonably well described by material mixture models, which identify each pixel as corresponding to one of a small number of materials, or as being a composite of two materials. Our classification algorithm consists of three steps. First, we apply hierarchical, adaptive grayscale adjustments to correct for nonlinearities in the MR sensor. The goal of this preprocessing step, based on the material mixture model, is to make the grayscale distribution of each tissue type constant across the entire image. Next, we perform an initial classification of all tissue types according to gray level. We have used a sum of Gaussian's approximation of the histogram to perform this classification. Finally, we identify pixels corresponding to cortex, by taking into account the spatial patterns characteristic of this tissue. For this purpose, we use a set of matched filters to identify image locations having the appropriate configuration of gray matter (cortex), cerebrospinal fluid and white matter, as determined by the previous classification step.

Influence of pansharpening techniques in obtaining accurate vegetation thematic maps

NASA Astrophysics Data System (ADS)

Ibarrola-Ulzurrun, Edurne; Gonzalo-Martin, Consuelo; Marcello-Ruiz, Javier

2016-10-01

In last decades, there have been a decline in natural resources, becoming important to develop reliable methodologies for their management. The appearance of very high resolution sensors has offered a practical and cost-effective means for a good environmental management. In this context, improvements are needed for obtaining higher quality of the information available in order to get reliable classified images. Thus, pansharpening enhances the spatial resolution of the multispectral band by incorporating information from the panchromatic image. The main goal in the study is to implement pixel and object-based classification techniques applied to the fused imagery using different pansharpening algorithms and the evaluation of thematic maps generated that serve to obtain accurate information for the conservation of natural resources. A vulnerable heterogenic ecosystem from Canary Islands (Spain) was chosen, Teide National Park, and Worldview-2 high resolution imagery was employed. The classes considered of interest were set by the National Park conservation managers. 7 pansharpening techniques (GS, FIHS, HCS, MTF based, Wavelet `à trous' and Weighted Wavelet `à trous' through Fractal Dimension Maps) were chosen in order to improve the data quality with the goal to analyze the vegetation classes. Next, different classification algorithms were applied at pixel-based and object-based approach, moreover, an accuracy assessment of the different thematic maps obtained were performed. The highest classification accuracy was obtained applying Support Vector Machine classifier at object-based approach in the Weighted Wavelet `à trous' through Fractal Dimension Maps fused image. Finally, highlight the difficulty of the classification in Teide ecosystem due to the heterogeneity and the small size of the species. Thus, it is important to obtain accurate thematic maps for further studies in the management and conservation of natural resources.

Quantifying the Availability of Tidewater Glacial Ice as Habitat for Harbor Seals in a Tidewater Glacial Fjord in Alaska Using Object-Based Image Analysis of Airborne Visible Imagery

NASA Astrophysics Data System (ADS)

Prakash, A.; Haselwimmer, C. E.; Gens, R.; Womble, J. N.; Ver Hoef, J.

2013-12-01

Tidewater glaciers are prominent landscape features that play a significant role in landscape and ecosystem processes along the southeastern and southcentral coasts of Alaska. Tidewater glaciers calve large icebergs that serve as an important substrate for harbor seals (Phoca vitulina richardii) for resting, pupping, nursing young, molting, and avoiding predators. Many of the tidewater glaciers in Alaska are retreating, which may influence harbor seal populations. Our objectives are to investigate the relationship between ice conditions and harbor seal distributions, which are poorly understood, in John's Hopkins Inlet, Glacier Bay National Park, Alaska, using a combination of airborne remote sensing and statistical modeling techniques. We present an overview of some results from Object-Based Image Analysis (OBIA) for classification of a time series of very high spatial resolution (4 cm pixels) airborne imagery acquired over John's Hopkins Inlet during the harbor seal pupping season in June and during the molting season in August from 2007 - 2012. Using OBIA we have developed a workflow to automate processing of the large volumes (~1250 images/survey) of airborne visible imagery for 1) classification of ice products (e.g. percent ice cover, percent brash ice, percent ice bergs) at a range of scales, and 2) quantitative determination of ice morphological properties such as iceberg size, roundness, and texture that are not found in traditional per-pixel classification approaches. These ice classifications and morphological variables are then used in statistical models to assess relationships with harbor seal abundance and distribution. Ultimately, understanding these relationships may provide novel perspectives on the spatial and temporal variation of harbor seals in tidewater glacial fjords.

A comparative study of linear and nonlinear anomaly detectors for hyperspectral imagery

NASA Astrophysics Data System (ADS)

Goldberg, Hirsh; Nasrabadi, Nasser M.

2007-04-01

In this paper we implement various linear and nonlinear subspace-based anomaly detectors for hyperspectral imagery. First, a dual window technique is used to separate the local area around each pixel into two regions - an inner-window region (IWR) and an outer-window region (OWR). Pixel spectra from each region are projected onto a subspace which is defined by projection bases that can be generated in several ways. Here we use three common pattern classification techniques (Principal Component Analysis (PCA), Fisher Linear Discriminant (FLD) Analysis, and the Eigenspace Separation Transform (EST)) to generate projection vectors. In addition to these three algorithms, the well-known Reed-Xiaoli (RX) anomaly detector is also implemented. Each of the four linear methods is then implicitly defined in a high- (possibly infinite-) dimensional feature space by using a nonlinear mapping associated with a kernel function. Using a common machine-learning technique known as the kernel trick all dot products in the feature space are replaced with a Mercer kernel function defined in terms of the original input data space. To determine how anomalous a given pixel is, we then project the current test pixel spectra and the spectral mean vector of the OWR onto the linear and nonlinear projection vectors in order to exploit the statistical differences between the IWR and OWR pixels. Anomalies are detected if the separation of the projection of the current test pixel spectra and the OWR mean spectra are greater than a certain threshold. Comparisons are made using receiver operating characteristics (ROC) curves.

Angular difference feature extraction for urban scene classification using ZY-3 multi-angle high-resolution satellite imagery

NASA Astrophysics Data System (ADS)

Huang, Xin; Chen, Huijun; Gong, Jianya

2018-01-01

Spaceborne multi-angle images with a high-resolution are capable of simultaneously providing spatial details and three-dimensional (3D) information to support detailed and accurate classification of complex urban scenes. In recent years, satellite-derived digital surface models (DSMs) have been increasingly utilized to provide height information to complement spectral properties for urban classification. However, in such a way, the multi-angle information is not effectively exploited, which is mainly due to the errors and difficulties of the multi-view image matching and the inaccuracy of the generated DSM over complex and dense urban scenes. Therefore, it is still a challenging task to effectively exploit the available angular information from high-resolution multi-angle images. In this paper, we investigate the potential for classifying urban scenes based on local angular properties characterized from high-resolution ZY-3 multi-view images. Specifically, three categories of angular difference features (ADFs) are proposed to describe the angular information at three levels (i.e., pixel, feature, and label levels): (1) ADF-pixel: the angular information is directly extrapolated by pixel comparison between the multi-angle images; (2) ADF-feature: the angular differences are described in the feature domains by comparing the differences between the multi-angle spatial features (e.g., morphological attribute profiles (APs)). (3) ADF-label: label-level angular features are proposed based on a group of urban primitives (e.g., buildings and shadows), in order to describe the specific angular information related to the types of primitive classes. In addition, we utilize spatial-contextual information to refine the multi-level ADF features using superpixel segmentation, for the purpose of alleviating the effects of salt-and-pepper noise and representing the main angular characteristics within a local area. The experiments on ZY-3 multi-angle images confirm that the proposed ADF features can effectively improve the accuracy of urban scene classification, with a significant increase in overall accuracy (3.8-11.7%) compared to using the spectral bands alone. Furthermore, the results indicated the superiority of the proposed ADFs in distinguishing between the spectrally similar and complex man-made classes, including roads and various types of buildings (e.g., high buildings, urban villages, and residential apartments).

Mapping burn severity, pine beetle infestation, and their interaction at the High Park Fire

NASA Astrophysics Data System (ADS)

Stone, Brandon

North America's western forests are experiencing wildfire and mountain pine beetle (MPB) disturbances that are unprecedented in the historic record, but it remains unclear whether and how MPB infestation influences post-infestation fire behavior. The 2012 High Park Fire burned in an area that's estimated to have begun a MPB outbreak cycle within five years before the wildfire, resulting in a landscape in which disturbance interactions can be studied. A first step in studying these interactions is mapping regions of beetle infestation and post-fire disturbance. We implemented an approach for mapping beetle infestation and burn severity using as source data three 5 m resolution RapidEye satellite images (two pre-fire, one post-fire). A two-tiered methodology was developed to overcome the spatial limitations of many classification approaches through explicit analyses at both pixel and plot level. Major land cover classes were photo-interpreted at the plot-level and their spectral signature used to classify 5 m images. A new image was generated at 25 m resolution by tabulating the fraction of coincident 5 m pixels in each cover class. The original photo interpretation was then used to train a second classification using as its source image the new 25 m image. Maps were validated using k-fold analysis of the original photo interpretation, field data collected immediately post-fire, and publicly available classifications. To investigate the influence of pre-fire beetle infestation on burn severity within the High Park Fire, we fit a log-linear model of conditional independence to our thematic maps after controlling for forest cover class and slope aspect. Our analysis revealed a high co-occurrence of severe burning and beetle infestation within high elevation lodgepole pine stands, but did not find statistically significant evidence that infected stands were more likely to burn severely than similar uninfected stands. Through an inspection of the year-to-year changes in the class fraction signatures of pixels classified as MPB infestation, we were able to observe increases in infection extent and intensity in the year before the fire. The resulting maps will help to increase our understanding of the process that contributed to the High Park Fire, and we believe that the novel classification approach will allow for improved characterization of forest disturbances.

WND-CHARM: Multi-purpose image classification using compound image transforms

PubMed Central

Orlov, Nikita; Shamir, Lior; Macura, Tomasz; Johnston, Josiah; Eckley, D. Mark; Goldberg, Ilya G.

2008-01-01

We describe a multi-purpose image classifier that can be applied to a wide variety of image classification tasks without modifications or fine-tuning, and yet provide classification accuracy comparable to state-of-the-art task-specific image classifiers. The proposed image classifier first extracts a large set of 1025 image features including polynomial decompositions, high contrast features, pixel statistics, and textures. These features are computed on the raw image, transforms of the image, and transforms of transforms of the image. The feature values are then used to classify test images into a set of pre-defined image classes. This classifier was tested on several different problems including biological image classification and face recognition. Although we cannot make a claim of universality, our experimental results show that this classifier performs as well or better than classifiers developed specifically for these image classification tasks. Our classifier’s high performance on a variety of classification problems is attributed to (i) a large set of features extracted from images; and (ii) an effective feature selection and weighting algorithm sensitive to specific image classification problems. The algorithms are available for free download from openmicroscopy.org. PMID:18958301

Real-time, resource-constrained object classification on a micro-air vehicle

NASA Astrophysics Data System (ADS)

Buck, Louis; Ray, Laura

2013-12-01

A real-time embedded object classification algorithm is developed through the novel combination of binary feature descriptors, a bag-of-visual-words object model and the cortico-striatal loop (CSL) learning algorithm. The BRIEF, ORB and FREAK binary descriptors are tested and compared to SIFT descriptors with regard to their respective classification accuracies, execution times, and memory requirements when used with CSL on a 12.6 g ARM Cortex embedded processor running at 800 MHz. Additionally, the effect of x2 feature mapping and opponent-color representations used with these descriptors is examined. These tests are performed on four data sets of varying sizes and difficulty, and the BRIEF descriptor is found to yield the best combination of speed and classification accuracy. Its use with CSL achieves accuracies between 67% and 95% of those achieved with SIFT descriptors and allows for the embedded classification of a 128x192 pixel image in 0.15 seconds, 60 times faster than classification with SIFT. X2 mapping is found to provide substantial improvements in classification accuracy for all of the descriptors at little cost, while opponent-color descriptors are offer accuracy improvements only on colorful datasets.

Reducing uncertainty on satellite image classification through spatiotemporal reasoning

NASA Astrophysics Data System (ADS)

Partsinevelos, Panagiotis; Nikolakaki, Natassa; Psillakis, Periklis; Miliaresis, George; Xanthakis, Michail

2014-05-01

The natural habitat constantly endures both inherent natural and human-induced influences. Remote sensing has been providing monitoring oriented solutions regarding the natural Earth surface, by offering a series of tools and methodologies which contribute to prudent environmental management. Processing and analysis of multi-temporal satellite images for the observation of the land changes include often classification and change-detection techniques. These error prone procedures are influenced mainly by the distinctive characteristics of the study areas, the remote sensing systems limitations and the image analysis processes. The present study takes advantage of the temporal continuity of multi-temporal classified images, in order to reduce classification uncertainty, based on reasoning rules. More specifically, pixel groups that temporally oscillate between classes are liable to misclassification or indicate problematic areas. On the other hand, constant pixel group growth indicates a pressure prone area. Computational tools are developed in order to disclose the alterations in land use dynamics and offer a spatial reference to the pressures that land use classes endure and impose between them. Moreover, by revealing areas that are susceptible to misclassification, we propose specific target site selection for training during the process of supervised classification. The underlying objective is to contribute to the understanding and analysis of anthropogenic and environmental factors that influence land use changes. The developed algorithms have been tested upon Landsat satellite image time series, depicting the National Park of Ainos in Kefallinia, Greece, where the unique in the world Abies cephalonica grows. Along with the minor changes and pressures indicated in the test area due to harvesting and other human interventions, the developed algorithms successfully captured fire incidents that have been historically confirmed. Overall, the results have shown that the use of the suggested procedures can contribute to the reduction of the classification uncertainty and support the existing knowledge regarding the pressure among land-use changes.

Detection of Coastline Deformation Using Remote Sensing and Geodetic Surveys

NASA Astrophysics Data System (ADS)

Sabuncu, A.; Dogru, A.; Ozener, H.; Turgut, B.

2016-06-01

The coastal areas are being destroyed due to the usage that effect the natural balance. Unconsciously sand mining from the sea for nearshore nourishment and construction uses are the main ones. Physical interferences for mining of sand cause an ecologic threat to the coastal environment. However, use of marine sand is inevitable because of economic reasons or unobtainable land-based sand resources. The most convenient solution in such a protection-usage dilemma is to reduce negative impacts of sand production from marine. This depends on the accurate determination of criteriaon production place, style, and amount of sand. With this motivation, nearshore geodedic surveying studies performed on Kilyos Campus of Bogazici University located on the Black Sea coast, north of Istanbul, Turkey between 2001-2002. The study area extends 1 km in the longshore. Geodetic survey was carried out in the summer of 2001 to detect the initial condition for the shoreline. Long-term seasonal changes in shoreline positions were determined biannually. The coast was measured with post-processed kinematic GPS. Besides, shoreline change has studied using Landsat imagery between the years 1986-2015. The data set of Landsat 5 imageries were dated 05.08.1986 and 31.08.2007 and Landsat 7 imageries were dated 21.07.2001 and 28.07.2015. Landcover types in the study area were analyzed on the basis of pixel based classification method. Firstly, unsupervised classification based on ISODATA (Iterative Self Organizing Data Analysis Technique) has been applied and spectral clusters have been determined that gives prior knowledge about the study area. In the second step, supervised classification was carried out by using the three different approaches which are minimum-distance, parallelepiped and maximum-likelihood. All pixel based classification processes were performed with ENVI 4.8 image processing software. Results of geodetic studies and classification outputs will be presented in this paper.

Lunar and Planetary Science XXXVI, Part 12

NASA Technical Reports Server (NTRS)

2005-01-01

Topics discussed include: The Ancient Lakes in Hellas Basin Region as Seen Through the First Year of Mars Express HRSC-Camera; DISR Observations of Craters at Titan at the Huygens Landing Site: Insights Anticipated; The Sun s Dust Disk - Discovery Potential of the New Horizons Mission During Interplanetary Cruise; Evidence for Aqueously Precipitated Sulfates in Northeast Meridiani Using THEMIS and TES Data; Integrated Spectroscopic Studies of Anhydrous Sulfate Minerals; Venusian Channel Formation as a Subsurface Process; Reexamination of Quartz Grains from the Permian-Triassic Boundary Section at Graphite Peak, Antarctica; Observations of Calcium Sulfate Deposits at High Latitudes by OMEGA/Mex at Km/Pixel Resolutions; Observations of the North Permanent Cap of Mars in Mid-Summer by OMEGA/MEX at km per Pixel Resolutions; Classification and Distribution of Patterned Ground in the Southern Hemisphere of Mars Genesis: Removing Contamination from Sample Collectors; Thermal Characterization of Fe3O4 Nanoparticles Formed from Poorly Crystalline Siderite; Hydrogen Abundances in Metal Grains from the Hammadah Al Hamra (HaH) 237 Metal-rich Chondrite: A Test of the Nebular-Formation Theory; REE and Some Other Trace Elements Distributions of Mineral Separates in Atlanta (EL6); The Composition and Origin of the Dewar Geochemical Anomaly; Asteroid Modal Mineralogy Using Hapke Mixing Models: Testing the Utility of Spectral Lookup Tables; and The Huygens Mission at Titan: Results Highlights. (sup 182)Hf-(sup 182)W Chronometry and an Early Differentiation in the Parent Body of Ureilites Ground Penetrating Radar in Sedimentary Rocks Mars, Always Cold, Sometimes Wet: New Constraints on Mars Denudation Rates and Climate Evolution from Analog Studies at Haughton Crater, Devon Island, High Arctic Europa s Porous Ice Rheology and Implications for Ice-penetrating Radar Scattering Loss Surface Generated Cracks on Europa

Sparsely-sampled hyperspectral stimulated Raman scattering microscopy: a theoretical investigation

NASA Astrophysics Data System (ADS)

Lin, Haonan; Liao, Chien-Sheng; Wang, Pu; Huang, Kai-Chih; Bouman, Charles A.; Kong, Nan; Cheng, Ji-Xin

2017-02-01

A hyperspectral image corresponds to a data cube with two spatial dimensions and one spectral dimension. Through linear un-mixing, hyperspectral images can be decomposed into spectral signatures of pure components as well as their concentration maps. Due to this distinct advantage on component identification, hyperspectral imaging becomes a rapidly emerging platform for engineering better medicine and expediting scientific discovery. Among various hyperspectral imaging techniques, hyperspectral stimulated Raman scattering (HSRS) microscopy acquires data in a pixel-by-pixel scanning manner. Nevertheless, current image acquisition speed for HSRS is insufficient to capture the dynamics of freely moving subjects. Instead of reducing the pixel dwell time to achieve speed-up, which would inevitably decrease signal-to-noise ratio (SNR), we propose to reduce the total number of sampled pixels. Location of sampled pixels are carefully engineered with triangular wave Lissajous trajectory. Followed by a model-based image in-painting algorithm, the complete data is recovered for linear unmixing. Simulation results show that by careful selection of trajectory, a fill rate as low as 10% is sufficient to generate accurate linear unmixing results. The proposed framework applies to any hyperspectral beam-scanning imaging platform which demands high acquisition speed.

Numerical trials of HISSE

NASA Technical Reports Server (NTRS)

Peters, C.; Kampe, F. (Principal Investigator)

1980-01-01

The mathematical description and implementation of the statistical estimation procedure known as the Houston integrated spatial/spectral estimator (HISSE) is discussed. HISSE is based on a normal mixture model and is designed to take advantage of spectral and spatial information of LANDSAT data pixels, utilizing the initial classification and clustering information provided by the AMOEBA algorithm. The HISSE calculates parametric estimates of class proportions which reduce the error inherent in estimates derived from typical classify and count procedures common to nonparametric clustering algorithms. It also singles out spatial groupings of pixels which are most suitable for labeling classes. These calculations are designed to aid the analyst/interpreter in labeling patches with a crop class label. Finally, HISSE's initial performance on an actual LANDSAT agricultural ground truth data set is reported.

Binarization of Gray-Scaled Digital Images Via Fuzzy Reasoning

NASA Technical Reports Server (NTRS)

Dominquez, Jesus A.; Klinko, Steve; Voska, Ned (Technical Monitor)

2002-01-01

A new fast-computational technique based on fuzzy entropy measure has been developed to find an optimal binary image threshold. In this method, the image pixel membership functions are dependent on the threshold value and reflect the distribution of pixel values in two classes; thus, this technique minimizes the classification error. This new method is compared with two of the best-known threshold selection techniques, Otsu and Huang-Wang. The performance of the proposed method supersedes the performance of Huang- Wang and Otsu methods when the image consists of textured background and poor printing quality. The three methods perform well but yield different binarization approaches if the background and foreground of the image have well-separated gray-level ranges.

Binarization of Gray-Scaled Digital Images Via Fuzzy Reasoning

NASA Technical Reports Server (NTRS)

Dominquez, Jesus A.; Klinko, Steve; Voska, Ned (Technical Monitor)

2002-01-01

A new fast-computational technique based on fuzzy entropy measure has been developed to find an optimal binary image threshold. In this method, the image pixel membership functions are dependent on the threshold value and reflect the distribution of pixel values in two classes; thus, this technique minimizes the classification error. This new method is compared with two of the best-known threshold selection techniques, Otsu and Huang-Wang. The performance of the proposed method supersedes the performance of Huang-Wang and Otsu methods when the image consists of textured background and poor printing quality. The three methods perform well but yield different binarization approaches if the background and foreground of the image have well-separated gray-level ranges.

Fast microcalcification detection in ultrasound images using image enhancement and threshold adjacency statistics

NASA Astrophysics Data System (ADS)

Cho, Baek Hwan; Chang, Chuho; Lee, Jong-Ha; Ko, Eun Young; Seong, Yeong Kyeong; Woo, Kyoung-Gu

2013-02-01

The existence of microcalcifications (MCs) is an important marker of malignancy in breast cancer. In spite of the benefits in mass detection for dense breasts, ultrasonography is believed that it might not reliably detect MCs. For computer aided diagnosis systems, however, accurate detection of MCs has the possibility of improving the performance in both Breast Imaging-Reporting and Data System (BI-RADS) lexicon description for calcifications and malignancy classification. We propose a new efficient and effective method for MC detection using image enhancement and threshold adjacency statistics (TAS). The main idea of TAS is to threshold an image and to count the number of white pixels with a given number of adjacent white pixels. Our contribution is to adopt TAS features and apply image enhancement to facilitate MC detection in ultrasound images. We employed fuzzy logic, tophat filter, and texture filter to enhance images for MCs. Using a total of 591 images, the classification accuracy of the proposed method in MC detection showed 82.75%, which is comparable to that of Haralick texture features (81.38%). When combined, the performance was as high as 85.11%. In addition, our method also showed the ability in mass classification when combined with existing features. In conclusion, the proposed method exploiting image enhancement and TAS features has the potential to deal with MC detection in ultrasound images efficiently and extend to the real-time localization and visualization of MCs.

Probabilistic detection of volcanic ash using a Bayesian approach

NASA Astrophysics Data System (ADS)

Mackie, Shona; Watson, Matthew

2014-03-01

Airborne volcanic ash can pose a hazard to aviation, agriculture, and both human and animal health. It is therefore important that ash clouds are monitored both day and night, even when they travel far from their source. Infrared satellite data provide perhaps the only means of doing this, and since the hugely expensive ash crisis that followed the 2010 Eyjafjalljökull eruption, much research has been carried out into techniques for discriminating ash in such data and for deriving key properties. Such techniques are generally specific to data from particular sensors, and most approaches result in a binary classification of pixels into "ash" and "ash free" classes with no indication of the classification certainty for individual pixels. Furthermore, almost all operational methods rely on expert-set thresholds to determine what constitutes "ash" and can therefore be criticized for being subjective and dependent on expertise that may not remain with an institution. Very few existing methods exploit available contemporaneous atmospheric data to inform the detection, despite the sensitivity of most techniques to atmospheric parameters. The Bayesian method proposed here does exploit such data and gives a probabilistic, physically based classification. We provide an example of the method's implementation for a scene containing both land and sea observations, and a large area of desert dust (often misidentified as ash by other methods). The technique has already been successfully applied to other detection problems in remote sensing, and this work shows that it will be a useful and effective tool for ash detection.

Evolutionary image simplification for lung nodule classification with convolutional neural networks.

PubMed

Lückehe, Daniel; von Voigt, Gabriele

2018-05-29

Understanding decisions of deep learning techniques is important. Especially in the medical field, the reasons for a decision in a classification task are as crucial as the pure classification results. In this article, we propose a new approach to compute relevant parts of a medical image. Knowing the relevant parts makes it easier to understand decisions. In our approach, a convolutional neural network is employed to learn structures of images of lung nodules. Then, an evolutionary algorithm is applied to compute a simplified version of an unknown image based on the learned structures by the convolutional neural network. In the simplified version, irrelevant parts are removed from the original image. In the results, we show simplified images which allow the observer to focus on the relevant parts. In these images, more than 50% of the pixels are simplified. The simplified pixels do not change the meaning of the images based on the learned structures by the convolutional neural network. An experimental analysis shows the potential of the approach. Besides the examples of simplified images, we analyze the run time development. Simplified images make it easier to focus on relevant parts and to find reasons for a decision. The combination of an evolutionary algorithm employing a learned convolutional neural network is well suited for the simplification task. From a research perspective, it is interesting which areas of the images are simplified and which parts are taken as relevant.

Spatiotemporal analysis of land use and land cover change in the Brazilian Amazon

PubMed Central

Li, Guiying; Moran, Emilio; Hetrick, Scott

2013-01-01

This paper provides a comparative analysis of land use and land cover (LULC) changes among three study areas with different biophysical environments in the Brazilian Amazon at multiple scales, from per-pixel, polygon, census sector, to study area. Landsat images acquired in the years of 1990/1991, 1999/2000, and 2008/2010 were used to examine LULC change trajectories with the post-classification comparison approach. A classification system composed of six classes – forest, savanna, other-vegetation (secondary succession and plantations), agro-pasture, impervious surface, and water, was designed for this study. A hierarchical-based classification method was used to classify Landsat images into thematic maps. This research shows different spatiotemporal change patterns, composition and rates among the three study areas and indicates the importance of analyzing LULC change at multiple scales. The LULC change analysis over time for entire study areas provides an overall picture of change trends, but detailed change trajectories and their spatial distributions can be better examined at a per-pixel scale. The LULC change at the polygon scale provides the information of the changes in patch sizes over time, while the LULC change at census sector scale gives new insights on how human-induced activities (e.g., urban expansion, roads, and land use history) affect LULC change patterns and rates. This research indicates the necessity to implement change detection at multiple scales for better understanding the mechanisms of LULC change patterns and rates. PMID:24127130

Classification and analysis of the Rudaki's Area

NASA Astrophysics Data System (ADS)

Zambon, F.; De sanctis, M.; Capaccioni, F.; Filacchione, G.; Carli, C.; Ammannito, E.; Frigeri, A.

2011-12-01

During the first two MESSENGER flybys the Mercury Dual Imaging System (MDIS) has mapped 90% of the Mercury's surface. An effective way to study the different terrain on planetary surfaces is to apply classification methods. These are based on clustering algorithms and they can be divided in two categories: unsupervised and supervised. The unsupervised classifiers do not require the analyst feedback and the algorithm automatically organizes pixels values into classes. In the supervised method, instead, the analyst must choose the "training area" that define the pixels value of a given class. We applied an unsupervised classifier, ISODATA, to the WAC filter images of the Rudaki's area where several kind of terrain have been identified showing differences in albedo, topography and crater density. ISODATA classifier divides this region in four classes: 1) shadow regions, 2) rough regions, 3) smooth plane, 4) highest reflectance area. ISODATA can not distinguish the high albedo regions from highly reflective illuminated edge of the craters, however the algorithm identify four classes that can be considered different units mainly on the basis of their reflectances at the various wavelengths. Is not possible, instead, to extrapolate compositional information because of the absence of clear spectral features. An additional analysis was made using ISODATA to choose the "training area" for further supervised classifications. These approach would allow, for example, to separate more accurately the edge of the craters from the high reflectance areas and the low reflectance regions from the shadow areas.

[Research on identification of cabbages and weeds combining spectral imaging technology and SAM taxonomy].

PubMed

Zu, Qin; Zhang, Shui-fa; Cao, Yang; Zhao, Hui-yi; Dang, Chang-qing

2015-02-01

Weeds automatic identification is the key technique and also the bottleneck for implementation of variable spraying and precision pesticide. Therefore, accurate, rapid and non-destructive automatic identification of weeds has become a very important research direction for precision agriculture. Hyperspectral imaging system was used to capture the hyperspectral images of cabbage seedlings and five kinds of weeds such as pigweed, barnyard grass, goosegrass, crabgrass and setaria with the wavelength ranging from 1000 to 2500 nm. In ENVI, by utilizing the MNF rotation to implement the noise reduction and de-correlation of hyperspectral data and reduce the band dimensions from 256 to 11, and extracting the region of interest to get the spectral library as standard spectra, finally, using the SAM taxonomy to identify cabbages and weeds, the classification effect was good when the spectral angle threshold was set as 0. 1 radians. In HSI Analyzer, after selecting the training pixels to obtain the standard spectrum, the SAM taxonomy was used to distinguish weeds from cabbages. Furthermore, in order to measure the recognition accuracy of weeds quantificationally, the statistical data of the weeds and non-weeds were obtained by comparing the SAM classification image with the best classification effects to the manual classification image. The experimental results demonstrated that, when the parameters were set as 5-point smoothing, 0-order derivative and 7-degree spectral angle, the best classification result was acquired and the recognition rate of weeds, non-weeds and overall samples was 80%, 97.3% and 96.8% respectively. The method that combined the spectral imaging technology and the SAM taxonomy together took full advantage of fusion information of spectrum and image. By applying the spatial classification algorithms to establishing training sets for spectral identification, checking the similarity among spectral vectors in the pixel level, integrating the advantages of spectra and images meanwhile considering their accuracy and rapidity and improving weeds detection range in the full range that could detect weeds between and within crop rows, the above method contributes relevant analysis tools and means to the application field requiring the accurate information of plants in agricultural precision management

A New Item Selection Procedure for Mixed Item Type in Computerized Classification Testing.

ERIC Educational Resources Information Center

Lau, C. Allen; Wang, Tianyou

This paper proposes a new Information-Time index as the basis for item selection in computerized classification testing (CCT) and investigates how this new item selection algorithm can help improve test efficiency for item pools with mixed item types. It also investigates how practical constraints such as item exposure rate control, test…

A comparison of FIA plot data derived from image pixels and image objects

Treesearch

Charles E. Werstak

2012-01-01

The use of Forest Inventory and Analysis (FIA) plot data for producing continuous and thematic maps of forest attributes (e.g., forest type, canopy cover, volume, and biomass) at the regional level from satellite imagery can be challenging due to differences in scale. Specifically, classification errors that may result from assumptions made between what the field data...

Monitoring urban tree cover using object-based image analysis and public domain remotely sensed data

Treesearch

L. Monika Moskal; Diane M. Styers; Meghan Halabisky

2011-01-01

Urban forest ecosystems provide a range of social and ecological services, but due to the heterogeneity of these canopies their spatial extent is difficult to quantify and monitor. Traditional per-pixel classification methods have been used to map urban canopies, however, such techniques are not generally appropriate for assessing these highly variable landscapes....

Classification of large-scale fundus image data sets: a cloud-computing framework.

PubMed

Roychowdhury, Sohini

2016-08-01

Large medical image data sets with high dimensionality require substantial amount of computation time for data creation and data processing. This paper presents a novel generalized method that finds optimal image-based feature sets that reduce computational time complexity while maximizing overall classification accuracy for detection of diabetic retinopathy (DR). First, region-based and pixel-based features are extracted from fundus images for classification of DR lesions and vessel-like structures. Next, feature ranking strategies are used to distinguish the optimal classification feature sets. DR lesion and vessel classification accuracies are computed using the boosted decision tree and decision forest classifiers in the Microsoft Azure Machine Learning Studio platform, respectively. For images from the DIARETDB1 data set, 40 of its highest-ranked features are used to classify four DR lesion types with an average classification accuracy of 90.1% in 792 seconds. Also, for classification of red lesion regions and hemorrhages from microaneurysms, accuracies of 85% and 72% are observed, respectively. For images from STARE data set, 40 high-ranked features can classify minor blood vessels with an accuracy of 83.5% in 326 seconds. Such cloud-based fundus image analysis systems can significantly enhance the borderline classification performances in automated screening systems.

Modeling the Effect of Grain Size Mixing on Thermal Inertia Values Derived from Diurnal and Seasonal THEMIS Measurements

NASA Astrophysics Data System (ADS)

McCarty, C.; Moersch, J.

2017-12-01

Sedimentary processes have slowed over Mars' geologic history. Analysis of the surface today can provide insight into the processes that may have affected it over its history. Sub-resolved checkerboard mixtures of materials with different thermal inertias (and therefore different grain sizes) can lead to differences in thermal inertia values inferred from night and day radiance observations. Information about the grain size distribution of a surface can help determine the degree of sorting it has experienced or it's geologic maturity. Standard methods for deriving thermal inertia from measurements made with THEMIS can give values for the same location that vary by as much as 20% between scenes. Such methods make the assumption that each THEMIS pixel contains material that has uniform thermophysical properties. Here we propose that if a mixture of small and large particles is present within a pixel, the inferred thermal inertia will be strongly dominated by whichever particle is warmer at the time of the measurement because the power radiated by a surface is proportional (by the Stefan-Boltzmann law) to the fourth power of its temperature. This effect will result in a change in thermal inertia values inferred from measurements taken at different times of day and night. Therefore, we expect to see correlation between the magnitude of diurnal variations in inferred thermal inertia values and the degree of grain size mixing for a given pixel location. Preliminary work has shown that the magnitude of such diurnal variation in inferred thermal inertias is sufficient to detect geologically useful differences in grain size distributions. We hypothesize that at least some of the 20% variability in thermal inertias inferred from multiple scenes for a given location could be attributed to sub-pixel grain size mixing rather than uncertainty inherent to the experiment, as previously thought. Mapping the difference in inferred thermal inertias from day and night THEMIS observations may prove to be a new way of distinguishing surfaces that have relatively uniform grain sizes from those that have mixed grain sizes. Assessing the effects of different geologic processes can be aided by noting variations in grain size distributions, so this method may be useful as a new way to extract geologic interpretations from the THEMIS thermal data set.

Landsat Thematic Mapper studies of land cover spatial variability related to hydrology

NASA Technical Reports Server (NTRS)

Wharton, S.; Ormsby, J.; Salomonson, V.; Mulligan, P.

1984-01-01

Past accomplishments involving remote sensing based land-cover analysis for hydrologic applications are reviewed. Ongoing research in exploiting the increased spatial, radiometric, and spectral capabilities afforded by the TM on Landsats 4 and 5 is considered. Specific studies to compare MSS and TM for urbanizing watersheds, wetlands, and floodplain mapping situations show that only a modest improvement in classification accuracy is achieved via statistical per pixel multispectral classifiers. The limitations of current approaches to multispectral classification are illustrated. The objectives, background, and progress in the development of an alternative analysis approach for defining inputs to urban hydrologic models using TM are discussed.

Landscape object-based analysis of wetland plant functional types: the effects of spatial scale, vegetation classes and classifier methods

NASA Astrophysics Data System (ADS)

Dronova, I.; Gong, P.; Wang, L.; Clinton, N.; Fu, W.; Qi, S.

2011-12-01

Remote sensing-based vegetation classifications representing plant function such as photosynthesis and productivity are challenging in wetlands with complex cover and difficult field access. Recent advances in object-based image analysis (OBIA) and machine-learning algorithms offer new classification tools; however, few comparisons of different algorithms and spatial scales have been discussed to date. We applied OBIA to delineate wetland plant functional types (PFTs) for Poyang Lake, the largest freshwater lake in China and Ramsar wetland conservation site, from 30-m Landsat TM scene at the peak of spring growing season. We targeted major PFTs (C3 grasses, C3 forbs and different types of C4 grasses and aquatic vegetation) that are both key players in system's biogeochemical cycles and critical providers of waterbird habitat. Classification results were compared among: a) several object segmentation scales (with average object sizes 900-9000 m2); b) several families of statistical classifiers (including Bayesian, Logistic, Neural Network, Decision Trees and Support Vector Machines) and c) two hierarchical levels of vegetation classification, a generalized 3-class set and more detailed 6-class set. We found that classification benefited from object-based approach which allowed including object shape, texture and context descriptors in classification. While a number of classifiers achieved high accuracy at the finest pixel-equivalent segmentation scale, the highest accuracies and best agreement among algorithms occurred at coarser object scales. No single classifier was consistently superior across all scales, although selected algorithms of Neural Network, Logistic and K-Nearest Neighbors families frequently provided the best discrimination of classes at different scales. The choice of vegetation categories also affected classification accuracy. The 6-class set allowed for higher individual class accuracies but lower overall accuracies than the 3-class set because individual classes differed in scales at which they were best discriminated from others. Main classification challenges included a) presence of C3 grasses in C4-grass areas, particularly following harvesting of C4 reeds and b) mixtures of emergent, floating and submerged aquatic plants at sub-object and sub-pixel scales. We conclude that OBIA with advanced statistical classifiers offers useful instruments for landscape vegetation analyses, and that spatial scale considerations are critical in mapping PFTs, while multi-scale comparisons can be used to guide class selection. Future work will further apply fuzzy classification and field-collected spectral data for PFT analysis and compare results with MODIS PFT products.

Design methodology: edgeless 3D ASICs with complex in-pixel processing for pixel detectors

DOE Office of Scientific and Technical Information (OSTI.GOV)

Fahim Farah, Fahim Farah; Deptuch, Grzegorz W.; Hoff, James R.

The design methodology for the development of 3D integrated edgeless pixel detectors with in-pixel processing using Electronic Design Automation (EDA) tools is presented. A large area 3 tier 3D detector with one sensor layer and two ASIC layers containing one analog and one digital tier, is built for x-ray photon time of arrival measurement and imaging. A full custom analog pixel is 65μm x 65μm. It is connected to a sensor pixel of the same size on one side, and on the other side it has approximately 40 connections to the digital pixel. A 32 x 32 edgeless array withoutmore » any peripheral functional blocks constitutes a sub-chip. The sub-chip is an indivisible unit, which is further arranged in a 6 x 6 array to create the entire 1.248cm x 1.248cm ASIC. Each chip has 720 bump-bond I/O connections, on the back of the digital tier to the ceramic PCB. All the analog tier power and biasing is conveyed through the digital tier from the PCB. The assembly has no peripheral functional blocks, and hence the active area extends to the edge of the detector. This was achieved by using a few flavors of almost identical analog pixels (minimal variation in layout) to allow for peripheral biasing blocks to be placed within pixels. The 1024 pixels within a digital sub-chip array have a variety of full custom, semi-custom and automated timing driven functional blocks placed together. The methodology uses a modified mixed-mode on-top digital implementation flow to not only harness the tool efficiency for timing and floor-planning but also to maintain designer control over compact parasitically aware layout. The methodology uses the Cadence design platform, however it is not limited to this tool.« less

Design methodology: edgeless 3D ASICs with complex in-pixel processing for pixel detectors

NASA Astrophysics Data System (ADS)

Fahim, Farah; Deptuch, Grzegorz W.; Hoff, James R.; Mohseni, Hooman

2015-08-01

The design methodology for the development of 3D integrated edgeless pixel detectors with in-pixel processing using Electronic Design Automation (EDA) tools is presented. A large area 3 tier 3D detector with one sensor layer and two ASIC layers containing one analog and one digital tier, is built for x-ray photon time of arrival measurement and imaging. A full custom analog pixel is 65μm x 65μm. It is connected to a sensor pixel of the same size on one side, and on the other side it has approximately 40 connections to the digital pixel. A 32 x 32 edgeless array without any peripheral functional blocks constitutes a sub-chip. The sub-chip is an indivisible unit, which is further arranged in a 6 x 6 array to create the entire 1.248cm x 1.248cm ASIC. Each chip has 720 bump-bond I/O connections, on the back of the digital tier to the ceramic PCB. All the analog tier power and biasing is conveyed through the digital tier from the PCB. The assembly has no peripheral functional blocks, and hence the active area extends to the edge of the detector. This was achieved by using a few flavors of almost identical analog pixels (minimal variation in layout) to allow for peripheral biasing blocks to be placed within pixels. The 1024 pixels within a digital sub-chip array have a variety of full custom, semi-custom and automated timing driven functional blocks placed together. The methodology uses a modified mixed-mode on-top digital implementation flow to not only harness the tool efficiency for timing and floor-planning but also to maintain designer control over compact parasitically aware layout. The methodology uses the Cadence design platform, however it is not limited to this tool.

Bringing the Coastal Zone into Finer Focus

NASA Astrophysics Data System (ADS)

Guild, L. S.; Hooker, S. B.; Kudela, R. M.; Morrow, J. H.; Torres-Perez, J. L.; Palacios, S. L.; Negrey, K.; Dungan, J. L.

2015-12-01

Measurements over extents from submeter to 10s of meters are critical science requirements for the design and integration of remote sensing instruments for coastal zone research. Various coastal ocean phenomena operate at different scales (e.g. meters to kilometers). For example, river plumes and algal blooms have typical extents of 10s of meters and therefore can be resolved with satellite data, however, shallow benthic ecosystem (e.g., coral, seagrass, and kelp) biodiversity and change are best studied at resolutions of submeter to meter, below the pixel size of typical satellite products. The delineation of natural phenomena do not fit nicely into gridded pixels and the coastal zone is complicated by mixed pixels at the land-sea interface with a range of bio-optical signals from terrestrial and water components. In many standard satellite products, these coastal mixed pixels are masked out because they confound algorithms for the ocean color parameter suite. In order to obtain data at the land/sea interface, finer spatial resolution satellite data can be achieved yet spectral resolution is sacrificed. This remote sensing resolution challenge thwarts the advancement of research in the coastal zone. Further, remote sensing of benthic ecosystems and shallow sub-surface phenomena are challenged by the requirements to sense through the sea surface and through a water column with varying light conditions from the open ocean to the water's edge. For coastal waters, >80% of the remote sensing signal is scattered/absorbed due to the atmospheric constituents, sun glint from the sea surface, and water column components. In addition to in-water measurements from various platforms (e.g., ship, glider, mooring, and divers), low altitude aircraft outfitted with high quality bio-optical radiometer sensors and targeted channels matched with in-water sensors and higher altitude platform sensors for ocean color products, bridge the sea-truth measurements to the pixels acquired from satellite and high altitude platforms. We highlight a novel NASA airborne calibration, validation, and research capability for addressing the coastal remote sensing resolution challenge.

The distance function effect on k-nearest neighbor classification for medical datasets.

PubMed

Hu, Li-Yu; Huang, Min-Wei; Ke, Shih-Wen; Tsai, Chih-Fong

2016-01-01

K-nearest neighbor (k-NN) classification is conventional non-parametric classifier, which has been used as the baseline classifier in many pattern classification problems. It is based on measuring the distances between the test data and each of the training data to decide the final classification output. Since the Euclidean distance function is the most widely used distance metric in k-NN, no study examines the classification performance of k-NN by different distance functions, especially for various medical domain problems. Therefore, the aim of this paper is to investigate whether the distance function can affect the k-NN performance over different medical datasets. Our experiments are based on three different types of medical datasets containing categorical, numerical, and mixed types of data and four different distance functions including Euclidean, cosine, Chi square, and Minkowsky are used during k-NN classification individually. The experimental results show that using the Chi square distance function is the best choice for the three different types of datasets. However, using the cosine and Euclidean (and Minkowsky) distance function perform the worst over the mixed type of datasets. In this paper, we demonstrate that the chosen distance function can affect the classification accuracy of the k-NN classifier. For the medical domain datasets including the categorical, numerical, and mixed types of data, K-NN based on the Chi square distance function performs the best.

UNMANNED AERIAL VEHICLE (UAV) HYPERSPECTRAL REMOTE SENSING FOR DRYLAND VEGETATION MONITORING

DOE Office of Scientific and Technical Information (OSTI.GOV)

Nancy F. Glenn; Jessica J. Mitchell; Matthew O. Anderson

2012-06-01

UAV-based hyperspectral remote sensing capabilities developed by the Idaho National Lab and Idaho State University, Boise Center Aerospace Lab, were recently tested via demonstration flights that explored the influence of altitude on geometric error, image mosaicking, and dryland vegetation classification. The test flights successfully acquired usable flightline data capable of supporting classifiable composite images. Unsupervised classification results support vegetation management objectives that rely on mapping shrub cover and distribution patterns. Overall, supervised classifications performed poorly despite spectral separability in the image-derived endmember pixels. Future mapping efforts that leverage ground reference data, ultra-high spatial resolution photos and time series analysis shouldmore » be able to effectively distinguish native grasses such as Sandberg bluegrass (Poa secunda), from invasives such as burr buttercup (Ranunculus testiculatus) and cheatgrass (Bromus tectorum).« less

A robust object-based shadow detection method for cloud-free high resolution satellite images over urban areas and water bodies

NASA Astrophysics Data System (ADS)

Tatar, Nurollah; Saadatseresht, Mohammad; Arefi, Hossein; Hadavand, Ahmad

2018-06-01

Unwanted contrast in high resolution satellite images such as shadow areas directly affects the result of further processing in urban remote sensing images. Detecting and finding the precise position of shadows is critical in different remote sensing processing chains such as change detection, image classification and digital elevation model generation from stereo images. The spectral similarity between shadow areas, water bodies, and some dark asphalt roads makes the development of robust shadow detection algorithms challenging. In addition, most of the existing methods work on pixel-level and neglect the contextual information contained in neighboring pixels. In this paper, a new object-based shadow detection framework is introduced. In the proposed method a pixel-level shadow mask is built by extending established thresholding methods with a new C4 index which enables to solve the ambiguity of shadow and water bodies. Then the pixel-based results are further processed in an object-based majority analysis to detect the final shadow objects. Four different high resolution satellite images are used to validate this new approach. The result shows the superiority of the proposed method over some state-of-the-art shadow detection method with an average of 96% in F-measure.

Change detection from synthetic aperture radar images based on neighborhood-based ratio and extreme learning machine

NASA Astrophysics Data System (ADS)

Gao, Feng; Dong, Junyu; Li, Bo; Xu, Qizhi; Xie, Cui

2016-10-01

Change detection is of high practical value to hazard assessment, crop growth monitoring, and urban sprawl detection. A synthetic aperture radar (SAR) image is the ideal information source for performing change detection since it is independent of atmospheric and sunlight conditions. Existing SAR image change detection methods usually generate a difference image (DI) first and use clustering methods to classify the pixels of DI into changed class and unchanged class. Some useful information may get lost in the DI generation process. This paper proposed an SAR image change detection method based on neighborhood-based ratio (NR) and extreme learning machine (ELM). NR operator is utilized for obtaining some interested pixels that have high probability of being changed or unchanged. Then, image patches centered at these pixels are generated, and ELM is employed to train a model by using these patches. Finally, pixels in both original SAR images are classified by the pretrained ELM model. The preclassification result and the ELM classification result are combined to form the final change map. The experimental results obtained on three real SAR image datasets and one simulated dataset show that the proposed method is robust to speckle noise and is effective to detect change information among multitemporal SAR images.

Novel approach for image skeleton and distance transformation parallel algorithms

NASA Astrophysics Data System (ADS)

Qing, Kent P.; Means, Robert W.

1994-05-01

Image Understanding is more important in medical imaging than ever, particularly where real-time automatic inspection, screening and classification systems are installed. Skeleton and distance transformations are among the common operations that extract useful information from binary images and aid in Image Understanding. The distance transformation describes the objects in an image by labeling every pixel in each object with the distance to its nearest boundary. The skeleton algorithm starts from the distance transformation and finds the set of pixels that have a locally maximum label. The distance algorithm has to scan the entire image several times depending on the object width. For each pixel, the algorithm must access the neighboring pixels and find the maximum distance from the nearest boundary. It is a computational and memory access intensive procedure. In this paper, we propose a novel parallel approach to the distance transform and skeleton algorithms using the latest VLSI high- speed convolutional chips such as HNC's ViP. The algorithm speed is dependent on the object's width and takes (k + [(k-1)/3]) * 7 milliseconds for a 512 X 512 image with k being the maximum distance of the largest object. All objects in the image will be skeletonized at the same time in parallel.

Semi-Supervised Sparse Representation Based Classification for Face Recognition With Insufficient Labeled Samples

NASA Astrophysics Data System (ADS)

Gao, Yuan; Ma, Jiayi; Yuille, Alan L.

2017-05-01

This paper addresses the problem of face recognition when there is only few, or even only a single, labeled examples of the face that we wish to recognize. Moreover, these examples are typically corrupted by nuisance variables, both linear (i.e., additive nuisance variables such as bad lighting, wearing of glasses) and non-linear (i.e., non-additive pixel-wise nuisance variables such as expression changes). The small number of labeled examples means that it is hard to remove these nuisance variables between the training and testing faces to obtain good recognition performance. To address the problem we propose a method called Semi-Supervised Sparse Representation based Classification (S$^3$RC). This is based on recent work on sparsity where faces are represented in terms of two dictionaries: a gallery dictionary consisting of one or more examples of each person, and a variation dictionary representing linear nuisance variables (e.g., different lighting conditions, different glasses). The main idea is that (i) we use the variation dictionary to characterize the linear nuisance variables via the sparsity framework, then (ii) prototype face images are estimated as a gallery dictionary via a Gaussian Mixture Model (GMM), with mixed labeled and unlabeled samples in a semi-supervised manner, to deal with the non-linear nuisance variations between labeled and unlabeled samples. We have done experiments with insufficient labeled samples, even when there is only a single labeled sample per person. Our results on the AR, Multi-PIE, CAS-PEAL, and LFW databases demonstrate that the proposed method is able to deliver significantly improved performance over existing methods.

Classification of Informal Settlements Through the Integration of 2d and 3d Features Extracted from Uav Data

NASA Astrophysics Data System (ADS)

Gevaert, C. M.; Persello, C.; Sliuzas, R.; Vosselman, G.

2016-06-01

Unmanned Aerial Vehicles (UAVs) are capable of providing very high resolution and up-to-date information to support informal settlement upgrading projects. In order to provide accurate basemaps, urban scene understanding through the identification and classification of buildings and terrain is imperative. However, common characteristics of informal settlements such as small, irregular buildings with heterogeneous roof material and large presence of clutter challenge state-of-the-art algorithms. Especially the dense buildings and steeply sloped terrain cause difficulties in identifying elevated objects. This work investigates how 2D radiometric and textural features, 2.5D topographic features, and 3D geometric features obtained from UAV imagery can be integrated to obtain a high classification accuracy in challenging classification problems for the analysis of informal settlements. It compares the utility of pixel-based and segment-based features obtained from an orthomosaic and DSM with point-based and segment-based features extracted from the point cloud to classify an unplanned settlement in Kigali, Rwanda. Findings show that the integration of 2D and 3D features leads to higher classification accuracies.

"Relative CIR": an image enhancement and visualization technique

USGS Publications Warehouse

Fleming, Michael D.

1993-01-01

Many techniques exist to spectrally and spatially enhance digital multispectral scanner data. One technique enhances an image while keeping the colors as they would appear in a color-infrared (CIR) image. This "relative CIR" technique generates an image that is both spectrally and spatially enhanced, while displaying a maximum range of colors. The technique enables an interpreter to visualize either spectral or land cover classes by their relative CIR characteristics. A relative CIR image is generated by developed spectral statistics for each class in the classifications and then, using a nonparametric approach for spectral enhancement, the means of the classes for each band are ranked. A 3 by 3 pixel smoothing filter is applied to the classification for spatial enhancement and the classes are mapped to the representative rank for each band. Practical applications of the technique include displaying an image classification product as a CIR image that was not derived directly from a spectral image, visualizing how a land cover classification would look as a CIR image, and displaying a spectral classification or intermediate product that will be used to label spectral classes.

Evaluation of several schemes for classification of remotely sensed data: Their parameters and performance. [Foster County, North Dakota; Grant County, Kansas; Iroquois County, Illinois, Tippecanoe County, Indiana; and Pottawattamie and Shelby Counties, Iowa

NASA Technical Reports Server (NTRS)

Scholz, D.; Fuhs, N.; Hixson, M.; Akiyama, T. (Principal Investigator)

1979-01-01

The author has identified the following significant results. Data sets for corn, soybeans, winter wheat, and spring wheat were used to evaluate the following schemes for crop identification: (1) per point Gaussian maximum classifier; (2) per point sum of normal densities classifiers; (3) per point linear classifier; (4) per point Gaussian maximum likelihood decision tree classifiers; and (5) texture sensitive per field Gaussian maximum likelihood classifier. Test site location and classifier both had significant effects on classification accuracy of small grains; classifiers did not differ significantly in overall accuracy, with the majority of the difference among classifiers being attributed to training method rather than to the classification algorithm applied. The complexity of use and computer costs for the classifiers varied significantly. A linear classification rule which assigns each pixel to the class whose mean is closest in Euclidean distance was the easiest for the analyst and cost the least per classification.

A Study of Hand Back Skin Texture Patterns for Personal Identification and Gender Classification

PubMed Central

Xie, Jin; Zhang, Lei; You, Jane; Zhang, David; Qu, Xiaofeng

2012-01-01

Human hand back skin texture (HBST) is often consistent for a person and distinctive from person to person. In this paper, we study the HBST pattern recognition problem with applications to personal identification and gender classification. A specially designed system is developed to capture HBST images, and an HBST image database was established, which consists of 1,920 images from 80 persons (160 hands). An efficient texton learning based method is then presented to classify the HBST patterns. First, textons are learned in the space of filter bank responses from a set of training images using the l1 -minimization based sparse representation (SR) technique. Then, under the SR framework, we represent the feature vector at each pixel over the learned dictionary to construct a representation coefficient histogram. Finally, the coefficient histogram is used as skin texture feature for classification. Experiments on personal identification and gender classification are performed by using the established HBST database. The results show that HBST can be used to assist human identification and gender classification. PMID:23012512

Investigation of SIS Up-Converters for Use in Multi-pixel Receivers

NASA Astrophysics Data System (ADS)

Uzawa, Yoshinori; Kojima, Takafumi; Shan, Wenlei; Gonzalez, Alvaro; Kroug, Matthias

2018-02-01

We propose the use of SIS junctions as a frequency up-converter based on quasiparticle mixing in frequency division multiplexing circuits for multi-pixel heterodyne receivers. Our theoretical calculation showed that SIS junctions have the potential to achieve positive gain and low-noise characteristics in the frequency up-conversion process at local oscillator (LO) frequencies larger than the voltage scale of the dc nonlinearity of the SIS junction. We experimentally observed up-conversion gain in a mixer with four-series Nb-based SIS junctions at the LO frequency of 105 GHz for the first time.

Some spectral and spatial characteristics of LANDSAT data

NASA Technical Reports Server (NTRS)

1982-01-01

Activities are provided for: (1) developing insight into the way in which the LANDSAT MSS produces multispectral data; (2) promoting understanding of what a "pixel" means in a LANDSAT image and the implications of the term "mixed pixel"; (3) explaining the concept of spectral signatures; (4) deriving a simple signature for a class or feature by analysis: of the four band images; (5) understanding the production of false color composites; (6) appreciating the use of color additive techniques; (7) preparing Diazo images; and (8) making quick visual identifications of major land cover types by their characteristic gray tones or colors in LANDSAT images.

Semi-supervised classification tool for DubaiSat-2 multispectral imagery

NASA Astrophysics Data System (ADS)

Al-Mansoori, Saeed

2015-10-01

This paper addresses a semi-supervised classification tool based on a pixel-based approach of the multi-spectral satellite imagery. There are not many studies demonstrating such algorithm for the multispectral images, especially when the image consists of 4 bands (Red, Green, Blue and Near Infrared) as in DubaiSat-2 satellite images. The proposed approach utilizes both unsupervised and supervised classification schemes sequentially to identify four classes in the image, namely, water bodies, vegetation, land (developed and undeveloped areas) and paved areas (i.e. roads). The unsupervised classification concept is applied to identify two classes; water bodies and vegetation, based on a well-known index that uses the distinct wavelengths of visible and near-infrared sunlight that is absorbed and reflected by the plants to identify the classes; this index parameter is called "Normalized Difference Vegetation Index (NDVI)". Afterward, the supervised classification is performed by selecting training homogenous samples for roads and land areas. Here, a precise selection of training samples plays a vital role in the classification accuracy. Post classification is finally performed to enhance the classification accuracy, where the classified image is sieved, clumped and filtered before producing final output. Overall, the supervised classification approach produced higher accuracy than the unsupervised method. This paper shows some current preliminary research results which point out the effectiveness of the proposed technique in a virtual perspective.

On-line analysis of algae in water by discrete three-dimensional fluorescence spectroscopy.

PubMed

Zhao, Nanjing; Zhang, Xiaoling; Yin, Gaofang; Yang, Ruifang; Hu, Li; Chen, Shuang; Liu, Jianguo; Liu, Wenqing

2018-03-19

In view of the problem of the on-line measurement of algae classification, a method of algae classification and concentration determination based on the discrete three-dimensional fluorescence spectra was studied in this work. The discrete three-dimensional fluorescence spectra of twelve common species of algae belonging to five categories were analyzed, the discrete three-dimensional standard spectra of five categories were built, and the recognition, classification and concentration prediction of algae categories were realized by the discrete three-dimensional fluorescence spectra coupled with non-negative weighted least squares linear regression analysis. The results show that similarities between discrete three-dimensional standard spectra of different categories were reduced and the accuracies of recognition, classification and concentration prediction of the algae categories were significantly improved. By comparing with that of the chlorophyll a fluorescence excitation spectra method, the recognition accuracy rate in pure samples by discrete three-dimensional fluorescence spectra is improved 1.38%, and the recovery rate and classification accuracy in pure diatom samples 34.1% and 46.8%, respectively; the recognition accuracy rate of mixed samples by discrete-three dimensional fluorescence spectra is enhanced by 26.1%, the recovery rate of mixed samples with Chlorophyta 37.8%, and the classification accuracy of mixed samples with diatoms 54.6%.

An Active Learning Framework for Hyperspectral Image Classification Using Hierarchical Segmentation

NASA Technical Reports Server (NTRS)

Zhang, Zhou; Pasolli, Edoardo; Crawford, Melba M.; Tilton, James C.

2015-01-01

Augmenting spectral data with spatial information for image classification has recently gained significant attention, as classification accuracy can often be improved by extracting spatial information from neighboring pixels. In this paper, we propose a new framework in which active learning (AL) and hierarchical segmentation (HSeg) are combined for spectral-spatial classification of hyperspectral images. The spatial information is extracted from a best segmentation obtained by pruning the HSeg tree using a new supervised strategy. The best segmentation is updated at each iteration of the AL process, thus taking advantage of informative labeled samples provided by the user. The proposed strategy incorporates spatial information in two ways: 1) concatenating the extracted spatial features and the original spectral features into a stacked vector and 2) extending the training set using a self-learning-based semi-supervised learning (SSL) approach. Finally, the two strategies are combined within an AL framework. The proposed framework is validated with two benchmark hyperspectral datasets. Higher classification accuracies are obtained by the proposed framework with respect to five other state-of-the-art spectral-spatial classification approaches. Moreover, the effectiveness of the proposed pruning strategy is also demonstrated relative to the approaches based on a fixed segmentation.

A semi-supervised classification algorithm using the TAD-derived background as training data

NASA Astrophysics Data System (ADS)

Fan, Lei; Ambeau, Brittany; Messinger, David W.

2013-05-01

In general, spectral image classification algorithms fall into one of two categories: supervised and unsupervised. In unsupervised approaches, the algorithm automatically identifies clusters in the data without a priori information about those clusters (except perhaps the expected number of them). Supervised approaches require an analyst to identify training data to learn the characteristics of the clusters such that they can then classify all other pixels into one of the pre-defined groups. The classification algorithm presented here is a semi-supervised approach based on the Topological Anomaly Detection (TAD) algorithm. The TAD algorithm defines background components based on a mutual k-Nearest Neighbor graph model of the data, along with a spectral connected components analysis. Here, the largest components produced by TAD are used as regions of interest (ROI's),or training data for a supervised classification scheme. By combining those ROI's with a Gaussian Maximum Likelihood (GML) or a Minimum Distance to the Mean (MDM) algorithm, we are able to achieve a semi supervised classification method. We test this classification algorithm against data collected by the HyMAP sensor over the Cooke City, MT area and University of Pavia scene.

Assessment of spruce (Picea obovata) abundance by spectral unmixing algorithm for sustainable forest management in highland Natural Reserve (case study of Zigalga Range, South-Ural State Natural Reserve, Russia).

NASA Astrophysics Data System (ADS)

Mikheeva, Anna; Moiseev, Pavel

2017-04-01

In mountain territories climate change affects forest productivity and growth, which results in the tree line advancing and increasing of the forest density. These changes pose new challenges for forest managers whose responsibilities include forest resources inventory, monitoring and protection of ecosystems, and assessment of forest vulnerability. These activities require a range of sources of information, including exact squares of forested areas, forest densities and species abundances. Picea obovata, dominant tree species in South-Ural State Natural Reserve, Russia has regenerated, propagated and increased its relative cover during the recent 70 years. A remarkable shift of the upper limit of Picea obovata up to 60-80 m upslope was registered by repeating photography, especially on gentle slopes. The stands of Picea obovata are monitored by Reserve inspectors on the test plots to ensure that forests maintain or improve their productivity, these studies also include projective cover measurements. However, it is impossible to cover the entire territory of the Reserve by detailed field observations. Remote sensing data from Terra ASTER imagery provides valuable information for large territories (scene covers an area of 60 x 60 km) and can be used for quantitative mapping of forest and non-forest vegetation at regional scale (spatial resolution is 15-30 m for visible and infrared bands). A case study of estimating Picea obovata abundance was conducted for forest and forest-tundra sites of Zigalga Range, using 9-band ASTER multispectral imagery of 23.08.2007, field data and spectral unmixing algorithm. This type of algorithms intends to derive object and its abundance from a mixed pixel of multispectral imagery which can be further converted to object's projective cover. Atmospheric correction was applied to the imagery prior to spectral unmixing, and then pure spectra of Picea obovata were extracted from the image in 10 points and averaged. These points located in Zigalga Range and were visited in summer 2016. We used Mixture-tuned Match Filtering (MTMF) algorithm, a non-linear subpixel classification technique which allows to separate the spectral mixture containing unknown objects, and to derive only known ones. The results of spectral unmixing classification were abundance maps of Picea obovata. The values were statistically determined (there was only selected abundances with high probabilities of presence and low probabilities of absence) and then constrained to the interval [0; 1]. Verification of maps was made at the sites of Iremel Mountains on the same ASTER image, where projective cover of Picea obovata was measured in the field in 147 points. The correlation coefficient between the spectral unmixing abundances and field-measured abundances was 0.7; not a very high value is due to the low sensitivity of the algorithm to detect abundances less than 0.25. The proposed method provides a tool for defining the Picea obovata boundaries more accurately than per-pixel automatic classification and locating new spruce islands in the mixing tree line environment. The abundances can be obtained for large areas with minimum field work which makes this approach cost-effective in providing timely information to nature reserve managers for adapting forest management actions to climate change.

Impact of sensor's point spread function on land cover characterization: Assessment and deconvolution

USGS Publications Warehouse

Huang, C.; Townshend, J.R.G.; Liang, S.; Kalluri, S.N.V.; DeFries, R.S.

2002-01-01

Measured and modeled point spread functions (PSF) of sensor systems indicate that a significant portion of the recorded signal of each pixel of a satellite image originates from outside the area represented by that pixel. This hinders the ability to derive surface information from satellite images on a per-pixel basis. In this study, the impact of the PSF of the Moderate Resolution Imaging Spectroradiometer (MODIS) 250 m bands was assessed using four images representing different landscapes. Experimental results showed that though differences between pixels derived with and without PSF effects were small on the average, the PSF generally brightened dark objects and darkened bright objects. This impact of the PSF lowered the performance of a support vector machine (SVM) classifier by 5.4% in overall accuracy and increased the overall root mean square error (RMSE) by 2.4% in estimating subpixel percent land cover. An inversion method based on the known PSF model reduced the signals originating from surrounding areas by as much as 53%. This method differs from traditional PSF inversion deconvolution methods in that the PSF was adjusted with lower weighting factors for signals originating from neighboring pixels than those specified by the PSF model. By using this deconvolution method, the lost classification accuracy due to residual impact of PSF effects was reduced to only 1.66% in overall accuracy. The increase in the RMSE of estimated subpixel land cover proportions due to the residual impact of PSF effects was reduced to 0.64%. Spatial aggregation also effectively reduced the errors in estimated land cover proportion images. About 50% of the estimation errors were removed after applying the deconvolution method and aggregating derived proportion images to twice their dimensional pixel size. ?? 2002 Elsevier Science Inc. All rights reserved.

Correlating multispectral imaging and compositional data from the Mars Exploration Rovers and implications for Mars Science Laboratory

NASA Astrophysics Data System (ADS)

Anderson, Ryan B.; Bell, James F.

2013-03-01

In an effort to infer compositional information about distant targets based on multispectral imaging data, we investigated methods of relating Mars Exploration Rover (MER) Pancam multispectral remote sensing observations to in situ alpha particle X-ray spectrometer (APXS)-derived elemental abundances and Mössbauer (MB)-derived abundances of Fe-bearing phases at the MER field sites in Gusev crater and Meridiani Planum. The majority of the partial correlation coefficients between these data sets were not statistically significant. Restricting the targets to those that were abraded by the rock abrasion tool (RAT) led to improved Pearson’s correlations, most notably between the red-blue ratio (673 nm/434 nm) and Fe3+-bearing phases, but partial correlations were not statistically significant. Partial Least Squares (PLS) calculations relating Pancam 11-color visible to near-IR (VNIR; ∼400-1000 nm) “spectra” to APXS and Mössbauer element or mineral abundances showed generally poor performance, although the presence of compositional outliers led to improved PLS results for data from Meridiani. When the Meridiani PLS model for pyroxene was tested by predicting the pyroxene content of Gusev targets, the results were poor, indicating that the PLS models for Meridiani are not applicable to data from other sites. Soft Independent Modeling of Class Analogy (SIMCA) classification of Gusev crater data showed mixed results. Of the 24 Gusev test regions of interest (ROIs) with known classes, 11 had >30% of the pixels in the ROI classified correctly, while others were mis-classified or unclassified. k-Means clustering of APXS and Mössbauer data was used to assign Meridiani targets to compositional classes. The clustering-derived classes corresponded to meaningful geologic and/or color unit differences, and SIMCA classification using these classes was somewhat successful, with >30% of pixels correctly classified in 9 of the 11 ROIs with known classes. This work shows that the relationship between SWIR multispectral imaging data and APXS- and Mössbauer-derived composition/mineralogy is often weak, a perhaps not entirely unexpected result given the different surface sampling depths of SWIR imaging (uppermost few microns) vs. APXS (tens of μm) and MB measurements (hundreds of μm). Results from the upcoming Mars Science Laboratory (MSL) rover’s ChemCam Laser Induced Breakdown Spectroscopy (LIBS) instrument may show a closer relationship to Mastcam SWIR multispectral observations, however, because the initial laser shots onto a target will analyze only the upper few micrometers of the surface. The clustering and classification methods used in this study can be applied to any data set to formalize the definition of classes and identify targets that do not fit in previously defined classes.

Correlating multispectral imaging and compositional data from the Mars Exploration Rovers and implications for Mars Science Laboratory

USGS Publications Warehouse

Anderson, Ryan B.; Bell, James F.

2013-01-01

In an effort to infer compositional information about distant targets based on multispectral imaging data, we investigated methods of relating Mars Exploration Rover (MER) Pancam multispectral remote sensing observations to in situ alpha particle X-ray spectrometer (APXS)-derived elemental abundances and Mössbauer (MB)-derived abundances of Fe-bearing phases at the MER field sites in Gusev crater and Meridiani Planum. The majority of the partial correlation coefficients between these data sets were not statistically significant. Restricting the targets to those that were abraded by the rock abrasion tool (RAT) led to improved Pearson’s correlations, most notably between the red–blue ratio (673 nm/434 nm) and Fe3+-bearing phases, but partial correlations were not statistically significant. Partial Least Squares (PLS) calculations relating Pancam 11-color visible to near-IR (VNIR; ∼400–1000 nm) “spectra” to APXS and Mössbauer element or mineral abundances showed generally poor performance, although the presence of compositional outliers led to improved PLS results for data from Meridiani. When the Meridiani PLS model for pyroxene was tested by predicting the pyroxene content of Gusev targets, the results were poor, indicating that the PLS models for Meridiani are not applicable to data from other sites. Soft Independent Modeling of Class Analogy (SIMCA) classification of Gusev crater data showed mixed results. Of the 24 Gusev test regions of interest (ROIs) with known classes, 11 had >30% of the pixels in the ROI classified correctly, while others were mis-classified or unclassified. k-Means clustering of APXS and Mössbauer data was used to assign Meridiani targets to compositional classes. The clustering-derived classes corresponded to meaningful geologic and/or color unit differences, and SIMCA classification using these classes was somewhat successful, with >30% of pixels correctly classified in 9 of the 11 ROIs with known classes. This work shows that the relationship between SWIR multispectral imaging data and APXS- and Mössbauer-derived composition/mineralogy is often weak, a perhaps not entirely unexpected result given the different surface sampling depths of SWIR imaging (uppermost few microns) vs. APXS (tens of μm) and MB measurements (hundreds of μm). Results from the upcoming Mars Science Laboratory (MSL) rover’s ChemCam Laser Induced Breakdown Spectroscopy (LIBS) instrument may show a closer relationship to Mastcam SWIR multispectral observations, however, because the initial laser shots onto a target will analyze only the upper few micrometers of the surface. The clustering and classification methods used in this study can be applied to any data set to formalize the definition of classes and identify targets that do not fit in previously defined classes.

Remotely Sensed Thermal Anomalies in Western Colorado

DOE Data Explorer

Khalid Hussein

2012-02-01

This layer contains the areas identified as areas of anomalous surface temperature from Landsat satellite imagery in Western Colorado. Data was obtained for two different dates. The digital numbers of each Landsat scene were converted to radiance and the temperature was calculated in degrees Kelvin and then converted to degrees Celsius for each land cover type using the emissivity of that cover type. And this process was repeated for each of the land cover types (open water, barren, deciduous forest and evergreen forest, mixed forest, shrub/scrub, grassland/herbaceous, pasture hay, and cultivated crops). The temperature of each pixel within each scene was calculated using the thermal band. In order to calculate the temperature an average emissivity value was used for each land cover type within each scene. The NLCD 2001 land cover classification raster data of the zones that cover Colorado were downloaded from USGS site and used to identify the land cover types within each scene. Areas that had temperature residual greater than 2o, and areas with temperature equal to 1o to 2o, were considered Landsat modeled very warm and warm surface exposures (thermal anomalies), respectively. Note: 'o' is used in this description to represent lowercase sigma.

A Comparative Object-Based Sugarcane Classification from Sentinel-2 Data Using Random Forests and Support Vector Machines

NASA Astrophysics Data System (ADS)

Chen, C. R.; Chen, C. F.; Nguyen, S. T.; Lau, K.; Lay, J. G.

2016-12-01

Sugarcane mostly grown in tropical and subtropical regions is one of the important commercial crops worldwide, providing significant employment, foreign exchange earnings, and other social and environmental benefits. The sugar industry is a vital component of Belize's economy as it provides employment to 15% of the country's population and 60% of the national agricultural exports. Sugarcane mapping is thus an important task due to official initiatives to provide reliable information on sugarcane-growing areas in respect to improved accuracy in monitoring sugarcane production and yield estimates. Policymakers need such monitoring information to formulate timely plans to ensure sustainably socioeconomic development. Sugarcane monitoring in Belize is traditionally carried out through time-consuming and costly field surveys. Remote sensing is an indispensable tool for crop monitoring on national, regional and global scales. The use of high and low resolution satellites for sugarcane monitoring in Belize is often restricted due to cost limitations and mixed pixel problems because sugarcane fields are small and fragmental. With the launch of Sentinel-2 satellite, it is possible to collectively map small patches of sugarcane fields over a large region as the data are free of charge and have high spectral, spatial, and temporal resolutions. This study aims to develop an object-based classification approach to comparatively map sugarcane fields in Belize from Sentinel-2 data using random forests (RF) and support vector machines (SVM). The data were processed through four main steps: (1) data pre-processing, (2) image segmentation, (3) sugarcane classification, and (4) accuracy assessment. The mapping results compared with the ground reference data indicated satisfactory results. The overall accuracies and Kappa coefficients were generally higher than 80% and 0.7, in both cases. The RF produced slightly more accurate mapping results than SVM. This study demonstrates the realization of the potential application of Sentinel-2 data for sugarcane mapping in Belize with the aid of RF and SVM methods. The methods are thus proposed for monitoring purposes in the country.

Nursing home resident assessment classification and focused care.

PubMed

Boondas, J

1991-06-01

Boondas provides the reader with an interesting history of the development of nursing homes as a prelude to discussion of reimbursement issues and case mix classification for today's residents. The need for minimal data sets and patient classification systems is stressed.

Multi-Resolution Analysis of MODIS and ASTER Satellite Data for Water Classification

DTIC Science & Technology

2006-09-01

spectral bands, but also with different pixel resolutions . The overall goal... the total water surface. Due to the constraint that high spatial resolution satellite images are low temporal resolution , one needs a reliable method...at 15 m resolution , were processed. We used MODIS reflectance data from MOD02 Level 1B data. Even the spatial resolution of the 1240 nm

Study on some useful Operators for Graph-theoretic Image Processing

NASA Astrophysics Data System (ADS)

Moghani, Ali; Nasiri, Parviz

2010-11-01

In this paper we describe a human perception based approach to pixel color segmentation which applied in color reconstruction by numerical method associated with graph-theoretic image processing algorithm typically in grayscale. Fuzzy sets defined on the Hue, Saturation and Value components of the HSV color space, provide a fuzzy logic model that aims to follow the human intuition of color classification.

Mapping trees outside forests using high-resolution aerial imagery: a comparison of pixel- and object based classification approaches

Treesearch

Dacia M. Meneguzzo; Greg C. Liknes; Mark D. Nelson

2013-01-01

Discrete trees and small groups of trees in nonforest settings are considered an essential resource around the world and are collectively referred to as trees outside forests (ToF). ToF provide important functions across the landscape, such as protecting soil and water resources, providing wildlife habitat, and improving farmstead energy efficiency and aesthetics....

Three-Dimensional Road Network by Fusion of Polarimetric and Interferometric SAR Data

NASA Technical Reports Server (NTRS)

Gamba, P.; Houshmand, B.

1998-01-01

In this paper a fuzzy classification procedure is applied to polarimetric radar measurements, and street pixels are detected. These data are successively grouped into consistent roads by means of a dynamic programming approach based on the fuzzy membership function values. Further fusion of the 2D road network extracted and 3D TOPSAR measurements provides a powerful way to analyze urban infrastructures.

Local Histograms for Per-Pixel Classification

DTIC Science & Technology

2012-03-01

few axioms for such models are presented. These axioms are shown to be satisfied using the convergence of random wavelet expansions. The authors of...pathologists can accurately and consistently identify and delineate tissues and their pathologies , it is an expensive and time-consuming task, therefore...Automatic Identification and Delineation of Tissues and Pathologies in H&E Stained Images. PhD Thesis. Carnegie Mellon University, Pittsburgh, PA (September

Integrated Efforts for Analysis of Geophysical Measurements and Models.

DTIC Science & Technology

1997-09-26

12b. DISTRIBUTION CODE 13. ABSTRACT ( Maximum 200 words) This contract supported investigations of integrated applications of physics, ephemerides...REGIONS AND GPS DATA VALIDATIONS 20 2.5 PL-SCINDA: VISUALIZATION AND ANALYSIS TECHNIQUES 22 2.5.1 View Controls 23 2.5.2 Map Selection...and IR data, about cloudy pixels. Clustering and maximum likelihood classification algorithms categorize up to four cloud layers into stratiform or

Mimicking human expert interpretation of remotely sensed raster imagery by using a novel segmentation analysis within ArcGIS

NASA Astrophysics Data System (ADS)

Le Bas, Tim; Scarth, Anthony; Bunting, Peter

2015-04-01

Traditional computer based methods for the interpretation of remotely sensed imagery use each pixel individually or the average of a small window of pixels to calculate a class or thematic value, which provides an interpretation. However when a human expert interprets imagery, the human eye is excellent at finding coherent and homogenous areas and edge features. It may therefore be advantageous for computer analysis to mimic human interpretation. A new toolbox for ArcGIS 10.x will be presented that segments the data layers into a set of polygons. Each polygon is defined by a K-means clustering and region growing algorithm, thus finding areas, their edges and any lineations in the imagery. Attached to each polygon are the characteristics of the imagery such as mean and standard deviation of the pixel values, within the polygon. The segmentation of imagery into a jigsaw of polygons also has the advantage that the human interpreter does not need to spend hours digitising the boundaries. The segmentation process has been taken from the RSGIS library of analysis and classification routines (Bunting et al., 2014). These routines are freeware and have been modified to be available in the ArcToolbox under the Windows (v7) operating system. Input to the segmentation process is a multi-layered raster image, for example; a Landsat image, or a set of raster datasets made up from derivatives of topography. The size and number of polygons are set by the user and are dependent on the imagery used. Examples will be presented of data from the marine environment utilising bathymetric depth, slope, rugosity and backscatter from a multibeam system. Meaningful classification of the polygons using their numerical characteristics is the next goal. Object based image analysis (OBIA) should help this workflow. Fully calibrated imagery systems will allow numerical classification to be translated into more readily understandable terms. Peter Bunting, Daniel Clewley, Richard M. Lucas and Sam Gillingham. 2014. The Remote Sensing and GIS Software Library (RSGISLib), Computers & Geosciences. Volume 62, Pages 216-226 http://dx.doi.org/10.1016/j.cageo.2013.08.007.

Detection by hyperspectral imaging of shiga toxin-producing Escherichia coli serogroups O26, O45, O103, O111, O121, and O145 on rainbow agar.

PubMed

Windham, William R; Yoon, Seung-Chul; Ladely, Scott R; Haley, Jennifer A; Heitschmidt, Jerry W; Lawrence, Kurt C; Park, Bosoon; Narrang, Neelam; Cray, William C

2013-07-01

The U.S. Department of Agriculture, Food Safety Inspection Service has determined that six non-O157 Shiga toxin-producing Escherichia coli (STEC) serogroups (O26, O45, O103, O111, O121, and O145) are adulterants in raw beef. Isolate and phenotypic discrimination of non-O157 STEC is problematic due to the lack of suitable agar media. The lack of distinct phenotypic color variation among non-O157serogroups cultured on chromogenic agar poses a challenge in selecting colonies for confirmation. In this study, visible and near-infrared hyperspectral imaging and chemometrics were used to detect and classify non-O157 STEC serogroups grown on Rainbow agar O157. The method was first developed by building spectral libraries for each serogroup obtained from ground-truth regions of interest representing the true identity of each pixel and thus each pure culture colony in the hyperspectral agar-plate image. The spectral library for the pure-culture non-O157 STEC consisted of 2,171 colonies, with spectra derived from 124,347 of pixels. The classification models for each serogroup were developed with a k nearest-neighbor classifier. The overall classification training accuracy at the colony level was 99%. The classifier was validated with ground beef enrichments artificially inoculated with 10, 50, and 100 CFU/ml STEC. The validation ground-truth regions of interest of the STEC target colonies consisted of 606 colonies, with 3,030 pixels of spectra. The overall classification accuracy was 98%. The average specificity of the method was 98% due to the low false-positive rate of 1.2%. The sensitivity ranged from 78 to 100% due to the false-negative rates of 22, 7, and 8% for O145, O45, and O26, respectively. This study showed the potential of visible and near-infrared hyperspectral imaging for detecting and classifying colonies of the six non-O157 STEC serogroups. The technique needs to be validated with bacterial cultures directly extracted from meat products and positive identification of colonies by using confirmatory tests such as latex agglutination tests or PCR.

Comparison of SAM and OBIA as Tools for Lava Morphology Classification - A Case Study in Krafla, NE Iceland

NASA Astrophysics Data System (ADS)

Aufaristama, Muhammad; Hölbling, Daniel; Höskuldsson, Ármann; Jónsdóttir, Ingibjörg

2017-04-01

The Krafla volcanic system is part of the Icelandic North Volcanic Zone (NVZ). During Holocene, two eruptive events occurred in Krafla, 1724-1729 and 1975-1984. The last eruptive episode (1975-1984), known as the "Krafla Fires", resulted in nine volcanic eruption episodes. The total area covered by the lavas from this eruptive episode is 36 km2 and the volume is about 0.25-0.3 km3. Lava morphology is related to the characteristics of the surface morphology of a lava flow after solidification. The typical morphology of lava can be used as primary basis for the classification of lava flows when rheological properties cannot be directly observed during emplacement, and also for better understanding the behavior of lava flow models. Although mapping of lava flows in the field is relatively accurate such traditional methods are time consuming, especially when the lava covers large areas such as it is the case in Krafla. Semi-automatic mapping methods that make use of satellite remote sensing data allow for an efficient and fast mapping of lava morphology. In this study, two semi-automatic methods for lava morphology classification are presented and compared using Landsat 8 (30 m spatial resolution) and SPOT-5 (10 m spatial resolution) satellite images. For assessing the classification accuracy, the results from semi-automatic mapping were compared to the respective results from visual interpretation. On the one hand, the Spectral Angle Mapper (SAM) classification method was used. With this method an image is classified according to the spectral similarity between the image reflectance spectrums and the reference reflectance spectra. SAM successfully produced detailed lava surface morphology maps. However, the pixel-based approach partly leads to a salt-and-pepper effect. On the other hand, we applied the Random Forest (RF) classification method within an object-based image analysis (OBIA) framework. This statistical classifier uses a randomly selected subset of training samples to produce multiple decision trees. For final classification of pixels or - in the present case - image objects, the average of the class assignments probability predicted by the different decision trees is used. While the resulting OBIA classification of lava morphology types shows a high coincidence with the reference data, the approach is sensitive to the segmentation-derived image objects that constitute the base units for classification. Both semi-automatic methods produce reasonable results in the Krafla lava field, even if the identification of different pahoehoe and aa types of lava appeared to be difficult. The use of satellite remote sensing data shows a high potential for fast and efficient classification of lava morphology, particularly over large and inaccessible areas.

Wavelet-based multicomponent denoising on GPU to improve the classification of hyperspectral images

NASA Astrophysics Data System (ADS)

Quesada-Barriuso, Pablo; Heras, Dora B.; Argüello, Francisco; Mouriño, J. C.

2017-10-01

Supervised classification allows handling a wide range of remote sensing hyperspectral applications. Enhancing the spatial organization of the pixels over the image has proven to be beneficial for the interpretation of the image content, thus increasing the classification accuracy. Denoising in the spatial domain of the image has been shown as a technique that enhances the structures in the image. This paper proposes a multi-component denoising approach in order to increase the classification accuracy when a classification method is applied. It is computed on multicore CPUs and NVIDIA GPUs. The method combines feature extraction based on a 1Ddiscrete wavelet transform (DWT) applied in the spectral dimension followed by an Extended Morphological Profile (EMP) and a classifier (SVM or ELM). The multi-component noise reduction is applied to the EMP just before the classification. The denoising recursively applies a separable 2D DWT after which the number of wavelet coefficients is reduced by using a threshold. Finally, inverse 2D-DWT filters are applied to reconstruct the noise free original component. The computational cost of the classifiers as well as the cost of the whole classification chain is high but it is reduced achieving real-time behavior for some applications through their computation on NVIDIA multi-GPU platforms.

Applying Support Vector Machine in classifying satellite images for the assessment of urban sprawl

NASA Astrophysics Data System (ADS)

murgante, Beniamino; Nolè, Gabriele; Lasaponara, Rosa; Lanorte, Antonio; Calamita, Giuseppe

2013-04-01

In last decades the spreading of new buildings, road infrastructures and a scattered proliferation of houses in zones outside urban areas, produced a countryside urbanization with no rules, consuming soils and impoverishing the landscape. Such a phenomenon generated a huge environmental impact, diseconomies and a decrease in life quality. This study analyzes processes concerning land use change, paying particular attention to urban sprawl phenomenon. The application is based on the integration of Geographic Information Systems and Remote Sensing adopting open source technologies. The objective is to understand size distribution and dynamic expansion of urban areas in order to define a methodology useful to both identify and monitor the phenomenon. In order to classify "urban" pixels, over time monitoring of settlements spread, understanding trends of artificial territories, classifications of satellite images at different dates have been realized. In order to obtain these classifications, supervised classification algorithms have been adopted. More particularly, Support Vector Machine (SVM) learning algorithm has been applied to multispectral remote data. One of the more interesting features in SVM is the possibility to obtain good results also adopting few classification pixels of training areas. SVM has several interesting features, such as the capacity to obtain good results also adopting few classification pixels of training areas, a high possibility of configuration parameters and the ability to discriminate pixels with similar spectral responses. Multi-temporal ASTER satellite data at medium resolution have been adopted because are very suitable in evaluating such phenomena. The application is based on the integration of Geographic Information Systems and Remote Sensing technologies by means of open source software. Tools adopted in managing and processing data are GRASS GIS, Quantum GIS and R statistical project. The area of interest is located south of Bari, in south eastern Italy (Puglia region). Bari, one of the major cities of southern Italy, is characterized by a considerable urban sprawl. The analysis is focused on a rectangular shaped region covering the urban area of three different cities, namely Polignano a Mare and Monopoli (and Conversano minority part) which, in 2011, had a population density comprised in the range of 140-319 people per Km2(istat ). The area of interest has a surface of approximately 253 Km2 , is characterized by three urban areas (Polignano a Mare, Conversano and Monopoli) and has a coastline of almost 17 Km. References Lanorte, A., Danese M., Lasaponara R., Murgante B. (2011) "Multiscale mapping of burn area and severity using multisensor satellite data and spatial autocorrelation analysis" International Journal of Applied Earth Observation and Geoinformation, Elsevier, doi:10.1016/j.jag.2011.09.005 Murgante B. Danese M. (2011) "Urban versus Rural: the decrease of agricultural areas and the development of urban zones analyzed with spatial statistics" Special Issue on "Environmental and agricultural data processing for water and territory management" International Journal of Agricultural and Environmental Information Systems (IJAEIS) volume 2(2) pp. 16-28 IGI Global, ISSN 1947-3192, DOI: 10.4018/jaeis.2011070102. Murgante, B., Las Casas, G., Danese, M., (2012), "Analyzing Neighbourhoods Suitable for Urban Renewal Programs with Autocorrelation Techniques" In Burian J. (Eds.) "Advances in Spatial Planning" InTech - Open Access DOI: 10.5772/33747 ISBN:978-953-51-0377-6 Nolè G., Danese M., Murgante B., Lasaponara R., Lanorte, A., (2012) "Using Spatial Autocorrelation Techniques and Multi-temporal Satellite Data for Analyzing Urban Sprawl" Lecture Notes in Computer Science vol. 7335, pp. 512-527. Springer-Verlag, Berlin. ISSN: 0302-9743, doi: 10.1007/978-3-642-31137-6_39

Shade images of forested areas obtained from LANDSAT MSS data

NASA Technical Reports Server (NTRS)

Shimabukuro, Yosio Edemir; Smith, James A.

1989-01-01

The pixel size in the present day Remote Sensing systems is large enough to include different types of land cover. Depending upon the target area, several components may be present within the pixel. In forested areas, generally, three main components are present: tree canopy, soil (understory), and shadow. The objective is to generate a shade (shadow) image of forested areas from multispectral measurements of LANDSAT MSS (Multispectral Scanner) data by implementing a linear mixing model, where shadow is considered as one of the primary components in a pixel. The shade images are related to the observed variation in forest structure, i.e., the proportion of inferred shadow in a pixel is related to different forest ages, forest types, and tree crown cover. The Constrained Least Squares (CLS) method is used to generate shade images for forest of eucalyptus and vegetation of cerrado using LANDSAT MSS imagery over Itapeva study area in Brazil. The resulted shade images may explain the difference on ages for forest of eucalyptus and the difference on three crown cover for vegetation of cerrado.

Artificial brains. A million spiking-neuron integrated circuit with a scalable communication network and interface.

PubMed

Merolla, Paul A; Arthur, John V; Alvarez-Icaza, Rodrigo; Cassidy, Andrew S; Sawada, Jun; Akopyan, Filipp; Jackson, Bryan L; Imam, Nabil; Guo, Chen; Nakamura, Yutaka; Brezzo, Bernard; Vo, Ivan; Esser, Steven K; Appuswamy, Rathinakumar; Taba, Brian; Amir, Arnon; Flickner, Myron D; Risk, William P; Manohar, Rajit; Modha, Dharmendra S

2014-08-08

Inspired by the brain's structure, we have developed an efficient, scalable, and flexible non-von Neumann architecture that leverages contemporary silicon technology. To demonstrate, we built a 5.4-billion-transistor chip with 4096 neurosynaptic cores interconnected via an intrachip network that integrates 1 million programmable spiking neurons and 256 million configurable synapses. Chips can be tiled in two dimensions via an interchip communication interface, seamlessly scaling the architecture to a cortexlike sheet of arbitrary size. The architecture is well suited to many applications that use complex neural networks in real time, for example, multiobject detection and classification. With 400-pixel-by-240-pixel video input at 30 frames per second, the chip consumes 63 milliwatts. Copyright © 2014, American Association for the Advancement of Science.

On the use of administrative databases to support planning activities: the case of the evaluation of neonatal case-mix in the Emilia-Romagna region using DRG and APR-DRG classification systems.

PubMed

Fantini, M P; Cisbani, L; Manzoli, L; Vertrees, J; Lorenzoni, L

2003-06-01

There are several versions of the Diagnosis Related Group (DRG) classification systems that are used for case-mix analysis, utilization review, prospective payment, and planning applications. The objective of this study was to assess the adequacy of two of these DRG systems--Medicare DRG and All Patient Refined DRG--to classify neonatal patients. The first part of the paper contains a descriptive analysis that outlines the major differences between the two systems in terms of classification logic and variables used in the assignment process. The second part examines the statistical performance of each system on the basis of the administrative data collected in all public hospitals of the Emilia-Romagna region relating to neonates discharged in 1997 and 1998. The Medicare DRG are less developed in terms of classification structure and yield a poorer statistical performance in terms of reduction in variance for length of stay. This is important because, for specific areas, a more refined system can prove useful at regional level to remove systematic biases in the measurement of case-mix due to the structural characteristics of the Medicare DRGs classification system.

Vegetation spatial variability and its effect on vegetation indices

NASA Technical Reports Server (NTRS)

Ormsby, J. P.; Choudhury, B. J.; Owe, M.

1987-01-01

Landsat MSS data were used to simulate low resolution satellite data, such as NOAA AVHRR, to quantify the fractional vegetation cover within a pixel and relate the fractional cover to the normalized difference vegetation index (NDVI) and the simple ratio (SR). The MSS data were converted to radiances from which the NDVI and SR values for the simulated pixels were determined. Each simulated pixel was divided into clusters using an unsupervised classification program. Spatial and spectral analysis provided a means of combining clusters representing similar surface characteristics into vegetated and non-vegetated areas. Analysis showed an average error of 12.7 per cent in determining these areas. NDVI values less than 0.3 represented fractional vegetated areas of 5 per cent or less, while a value of 0.7 or higher represented fractional vegetated areas greater than 80 per cent. Regression analysis showed a strong linear relation between fractional vegetation area and the NDVI and SR values; correlation values were 0.89 and 0.95 respectively. The range of NDVI values calculated from the MSS data agrees well with field studies.

a Study of the Impact of Insolation on Remote Sensing-Based Landcover and Landuse Data Extraction

NASA Astrophysics Data System (ADS)

Becek, K.; Borkowski, A.; Mekik, Ç.

2016-06-01

We examined the dependency of the pixel reflectance of hyperspectral imaging spectrometer data (HISD) on a normalized total insolation index (NTII). The NTII was estimated using a light detection and ranging (LiDAR)-derived digital surface model (DSM). The NTII and the pixel reflectance were dependent, to various degrees, on the band considered, and on the properties of the objects. The findings could be used to improve land cover (LC)/land use (LU) classification, using indices constructed from the spectral bands of imaging spectrometer data (ISD). To study this possibility, we investigated the normalized difference vegetation index (NDVI) at various NTII levels. The results also suggest that the dependency of the pixel reflectance and NTII could be used to mitigate the shadows in ISD. This project was carried out using data provided by the Hyperspectral Image Analysis Group and the NSF-funded Centre for Airborne Laser Mapping (NCALM), University of Houston, for the purpose of organizing the 2013 Data Fusion Contest (IEEE 2014). This contest was organized by the IEEE GRSS Data Fusion Technical Committee.

Estimating Mixed Broadleaves Forest Stand Volume Using Dsm Extracted from Digital Aerial Images

NASA Astrophysics Data System (ADS)

Sohrabi, H.

2012-07-01

In mixed old growth broadleaves of Hyrcanian forests, it is difficult to estimate stand volume at plot level by remotely sensed data while LiDar data is absent. In this paper, a new approach has been proposed and tested for estimating stand forest volume. The approach is based on this idea that forest volume can be estimated by variation of trees height at plots. In the other word, the more the height variation in plot, the more the stand volume would be expected. For testing this idea, 120 circular 0.1 ha sample plots with systematic random design has been collected in Tonekaon forest located in Hyrcanian zone. Digital surface model (DSM) measure the height values of the first surface on the ground including terrain features, trees, building etc, which provides a topographic model of the earth's surface. The DSMs have been extracted automatically from aerial UltraCamD images so that ground pixel size for extracted DSM varied from 1 to 10 m size by 1m span. DSMs were checked manually for probable errors. Corresponded to ground samples, standard deviation and range of DSM pixels have been calculated. For modeling, non-linear regression method was used. The results showed that standard deviation of plot pixels with 5 m resolution was the most appropriate data for modeling. Relative bias and RMSE of estimation was 5.8 and 49.8 percent, respectively. Comparing to other approaches for estimating stand volume based on passive remote sensing data in mixed broadleaves forests, these results are more encouraging. One big problem in this method occurs when trees canopy cover is totally closed. In this situation, the standard deviation of height is low while stand volume is high. In future studies, applying forest stratification could be studied.

Une nouvelle méthode de cartographie de la région d'Oran (Algérie) à l'aide de la télédétection multispectrale

NASA Astrophysics Data System (ADS)

Laoufi, Fatiha; Belbachir, Ahmed-Hafid; Benabadji, Noureddine; Zanoun, Abdelouahab

2011-10-01

We have mapped the region of Oran, Algeria, using multispectral remote sensing with different resolutions. For the identification of objects on the ground using their spectral signatures, two methods were applied to images from SPOT, LANDSAT, IRS-1 C and ASTER. The first one is called Base Rule method (BR method) and is based on a set of rules that must be met at each pixel in the different bands reflectance calibrated and henceforth it is assigned to a given class. The construction of these rules is based on the spectral profiles of popular classes in the scene studied. The second one is called Spectral Angle Mapper method (SAM method) and is based on the direct calculation of the spectral angle between the target vector representing the spectral profile of the desired class and the pixel vector whose components are numbered accounts in the different bands of the calibrated image reflectance. This new method was performed using PCSATWIN software developed by our own laboratory LAAR. After collecting a library of spectral signatures with multiple libraries, a detailed study of the principles and physical processes that can influence the spectral signature has been conducted. The final goal is to establish the range of variation of a spectral profile of a well-defined class and therefore to get precise bases for spectral rules. From the results we have obtained, we find that the supervised classification of these pixels by BR method derived from spectral signatures reduces the uncertainty associated with identifying objects by enhancing significantly the percentage of correct classification with very distinct classes.

Supervised classification of aerial imagery and multi-source data fusion for flood assessment

NASA Astrophysics Data System (ADS)

Sava, E.; Harding, L.; Cervone, G.

2015-12-01

Floods are among the most devastating natural hazards and the ability to produce an accurate and timely flood assessment before, during, and after an event is critical for their mitigation and response. Remote sensing technologies have become the de-facto approach for observing the Earth and its environment. However, satellite remote sensing data are not always available. For these reasons, it is crucial to develop new techniques in order to produce flood assessments during and after an event. Recent advancements in data fusion techniques of remote sensing with near real time heterogeneous datasets have allowed emergency responders to more efficiently extract increasingly precise and relevant knowledge from the available information. This research presents a fusion technique using satellite remote sensing imagery coupled with non-authoritative data such as Civil Air Patrol (CAP) and tweets. A new computational methodology is proposed based on machine learning algorithms to automatically identify water pixels in CAP imagery. Specifically, wavelet transformations are paired with multiple classifiers, run in parallel, to build models discriminating water and non-water regions. The learned classification models are first tested against a set of control cases, and then used to automatically classify each image separately. A measure of uncertainty is computed for each pixel in an image proportional to the number of models classifying the pixel as water. Geo-tagged tweets are continuously harvested and stored on a MongoDB and queried in real time. They are fused with CAP classified data, and with satellite remote sensing derived flood extent results to produce comprehensive flood assessment maps. The final maps are then compared with FEMA generated flood extents to assess their accuracy. The proposed methodology is applied on two test cases, relative to the 2013 floods in Boulder CO, and the 2015 floods in Texas.

Classification of Global Urban Centers Using ASTER Data: Preliminary Results From the Urban Environmental Monitoring Program

NASA Astrophysics Data System (ADS)

Stefanov, W. L.; Stefanov, W. L.; Christensen, P. R.

2001-05-01

Land cover and land use changes associated with urbanization are important drivers of global ecologic and climatic change. Quantification and monitoring of these changes are part of the primary mission of the ASTER instrument, and comprise the fundamental research objective of the Urban Environmental Monitoring (UEM) Program. The UEM program will acquire day/night, visible through thermal infrared ASTER data twice per year for 100 global urban centers over the duration of the mission (6 years). Data are currently available for a number of these urban centers and allow for initial comparison of global city structure using spatial variance texture analysis of the 15 m/pixel visible to near infrared ASTER bands. Variance texture analysis highlights changes in pixel edge density as recorded by sharp transitions from bright to dark pixels. In human-dominated landscapes these brightness variations correlate well with urbanized vs. natural land cover and are useful for characterizing the geographic extent and internal structure of cities. Variance texture analysis was performed on twelve urban centers (Albuquerque, Baghdad, Baltimore, Chongqing, Istanbul, Johannesburg, Lisbon, Madrid, Phoenix, Puebla, Riyadh, Vancouver) for which cloud-free daytime ASTER data are available. Image transects through each urban center produce texture profiles that correspond to urban density. These profiles can be used to classify cities into centralized (ex. Baltimore), decentralized (ex. Phoenix), or intermediate (ex. Madrid) structural types. Image texture is one of the primary data inputs (with vegetation indices and visible to thermal infrared image spectra) to a knowledge-based land cover classifier currently under development for application to ASTER UEM data as it is acquired. Collaboration with local investigators is sought to both verify the accuracy of the knowledge-based system and to develop more sophisticated classification models.

Exploring the impact of wavelet-based denoising in the classification of remote sensing hyperspectral images

NASA Astrophysics Data System (ADS)

Quesada-Barriuso, Pablo; Heras, Dora B.; Argüello, Francisco

2016-10-01

The classification of remote sensing hyperspectral images for land cover applications is a very intensive topic. In the case of supervised classification, Support Vector Machines (SVMs) play a dominant role. Recently, the Extreme Learning Machine algorithm (ELM) has been extensively used. The classification scheme previously published by the authors, and called WT-EMP, introduces spatial information in the classification process by means of an Extended Morphological Profile (EMP) that is created from features extracted by wavelets. In addition, the hyperspectral image is denoised in the 2-D spatial domain, also using wavelets and it is joined to the EMP via a stacked vector. In this paper, the scheme is improved achieving two goals. The first one is to reduce the classification time while preserving the accuracy of the classification by using ELM instead of SVM. The second one is to improve the accuracy results by performing not only a 2-D denoising for every spectral band, but also a previous additional 1-D spectral signature denoising applied to each pixel vector of the image. For each denoising the image is transformed by applying a 1-D or 2-D wavelet transform, and then a NeighShrink thresholding is applied. Improvements in terms of classification accuracy are obtained, especially for images with close regions in the classification reference map, because in these cases the accuracy of the classification in the edges between classes is more relevant.

Information extraction with object based support vector machines and vegetation indices

NASA Astrophysics Data System (ADS)

Ustuner, Mustafa; Abdikan, Saygin; Balik Sanli, Fusun

2016-07-01

Information extraction through remote sensing data is important for policy and decision makers as extracted information provide base layers for many application of real world. Classification of remotely sensed data is the one of the most common methods of extracting information however it is still a challenging issue because several factors are affecting the accuracy of the classification. Resolution of the imagery, number and homogeneity of land cover classes, purity of training data and characteristic of adopted classifiers are just some of these challenging factors. Object based image classification has some superiority than pixel based classification for high resolution images since it uses geometry and structure information besides spectral information. Vegetation indices are also commonly used for the classification process since it provides additional spectral information for vegetation, forestry and agricultural areas. In this study, the impacts of the Normalized Difference Vegetation Index (NDVI) and Normalized Difference Red Edge Index (NDRE) on the classification accuracy of RapidEye imagery were investigated. Object based Support Vector Machines were implemented for the classification of crop types for the study area located in Aegean region of Turkey. Results demonstrated that the incorporation of NDRE increase the classification accuracy from 79,96% to 86,80% as overall accuracy, however NDVI decrease the classification accuracy from 79,96% to 78,90%. Moreover it is proven than object based classification with RapidEye data give promising results for crop type mapping and analysis.

Timepix Device Efficiency for Pattern Recognition of Tracks Generated by Ionizing Radiation

NASA Astrophysics Data System (ADS)

Leroy, Claude; Asbah, Nedaa; Gagnon, Louis-Guilaume; Larochelle, Jean-Simon; Pospisil, Stanislav; Soueid, Paul

2014-06-01

A hybrid silicon pixelated TIMEPIX detector (256 × 256 pixels with 55 μm pitch) operated in Time Over Threshold (TOT) mode was exposed to radioactive sources (241Am, 106Ru, 137Cs), protons and alpha-particles after Rutherford Backscattering on a thin gold foil of protons and alpha-particles beams delivered by the Tandem Accelerator of Montreal University. Measurements were also performed with different mixed radiation fields of heavy charged particles (protons and alpha-particles), photons and electrons produced by simultaneous exposure of TIMEPIX to the radioactive sources and to protons beams on top of the radioactive sources. All measurements were performed in vacuum. The TOT mode of operation has allowed the direct measurement of the energy deposited in each pixel. The efficiency of track recognition with this device was tested by comparing the experimental activities (determined from number of tracks measurements) of the radioactive sources with their expected activities. The efficiency of track recognition of incident protons and alpha-particles of different energies as a function of the incidence angle was measured. The operation of TIMEPIX in TOT mode has allowed a 3D mapping of the charge sharing effect in the whole volume of the silicon sensor. The effect of the bias voltage on charge sharing was investigated as the level of charge sharing is related to the local profile of the electric field in the sensor. The results of the present measurements demonstrate the TIMEPIX capability of differentiating between different types of particles species from mixed radiation fields and measuring their energy deposition. Single track analysis gives a good precision (significantly better than the 55 μm size of one detector pixel) on the coordinates of the impact point of protons interacting in the TIMEPIX silicon layer.

Ice Water Classification Using Statistical Distribution Based Conditional Random Fields in RADARSAT-2 Dual Polarization Imagery

NASA Astrophysics Data System (ADS)

Zhang, Y.; Li, F.; Zhang, S.; Hao, W.; Zhu, T.; Yuan, L.; Xiao, F.

2017-09-01

In this paper, Statistical Distribution based Conditional Random Fields (STA-CRF) algorithm is exploited for improving marginal ice-water classification. Pixel level ice concentration is presented as the comparison of methods based on CRF. Furthermore, in order to explore the effective statistical distribution model to be integrated into STA-CRF, five statistical distribution models are investigated. The STA-CRF methods are tested on 2 scenes around Prydz Bay and Adélie Depression, where contain a variety of ice types during melt season. Experimental results indicate that the proposed method can resolve sea ice edge well in Marginal Ice Zone (MIZ) and show a robust distinction of ice and water.

Boosting bonsai trees for handwritten/printed text discrimination

NASA Astrophysics Data System (ADS)

Ricquebourg, Yann; Raymond, Christian; Poirriez, Baptiste; Lemaitre, Aurélie; Coüasnon, Bertrand

2013-12-01

Boosting over decision-stumps proved its efficiency in Natural Language Processing essentially with symbolic features, and its good properties (fast, few and not critical parameters, not sensitive to over-fitting) could be of great interest in the numeric world of pixel images. In this article we investigated the use of boosting over small decision trees, in image classification processing, for the discrimination of handwritten/printed text. Then, we conducted experiments to compare it to usual SVM-based classification revealing convincing results with very close performance, but with faster predictions and behaving far less as a black-box. Those promising results tend to make use of this classifier in more complex recognition tasks like multiclass problems.

Resolving the percentage of component terrains within single resolution elements

NASA Technical Reports Server (NTRS)

Marsh, S. E.; Switzer, P.; Kowalik, W. S.; Lyon, R. J. P.

1980-01-01

An approximate maximum likelihood technique employing a widely available discriminant analysis program is discussed that has been developed for resolving the percentage of component terrains within single resolution elements. The method uses all four channels of Landsat data simultaneously and does not require prior knowledge of the percentage of components in mixed pixels. It was tested in five cases that were chosen to represent mixtures of outcrop, soil and vegetation which would typically be encountered in geologic studies with Landsat data. For all five cases, the method proved to be superior to single band weighted average and linear regression techniques and permitted an estimate of the total area occupied by component terrains to within plus or minus 6% of the true area covered. Its major drawback is a consistent overestimation of the pixel component percent of the darker materials (vegetation) and an underestimation of the pixel component percent of the brighter materials (sand).

Getting small: new 10μm pixel pitch cooled infrared products

NASA Astrophysics Data System (ADS)

Reibel, Y.; Pere-Laperne, N.; Augey, T.; Rubaldo, L.; Decaens, G.; Bourqui, M.-L.; Manissadjian, A.; Billon-Lanfrey, D.; Bisotto, S.; Gravrand, O.; Destefanis, G.; Druart, G.; Guerineau, N.

2014-06-01

Recent advances in miniaturization of IR imaging technology have led to a burgeoning market for mini thermalimaging sensors. Seen in this context our development on smaller pixel pitch has opened the door to very compact products. When this competitive advantage is mixed with smaller coolers, thanks to HOT technology, we achieve valuable reductions in size, weight and power of the overall package. In the same time, we are moving towards a global offer based on digital interfaces that provides our customers lower power consumption and simplification on the IR system design process while freeing up more space. Additionally, we are also investigating new wafer level camera solution taking advantage of the progress in micro-optics. This paper discusses recent developments on hot and small pixel pitch technologies as well as efforts made on compact packaging solution developed by SOFRADIR in collaboration with CEA-LETI and ONERA.

Spectral Unmixing With Multiple Dictionaries

NASA Astrophysics Data System (ADS)

Cohen, Jeremy E.; Gillis, Nicolas

2018-02-01

Spectral unmixing aims at recovering the spectral signatures of materials, called endmembers, mixed in a hyperspectral or multispectral image, along with their abundances. A typical assumption is that the image contains one pure pixel per endmember, in which case spectral unmixing reduces to identifying these pixels. Many fully automated methods have been proposed in recent years, but little work has been done to allow users to select areas where pure pixels are present manually or using a segmentation algorithm. Additionally, in a non-blind approach, several spectral libraries may be available rather than a single one, with a fixed number (or an upper or lower bound) of endmembers to chose from each. In this paper, we propose a multiple-dictionary constrained low-rank matrix approximation model that address these two problems. We propose an algorithm to compute this model, dubbed M2PALS, and its performance is discussed on both synthetic and real hyperspectral images.

Chaos based video encryption using maps and Ikeda time delay system

NASA Astrophysics Data System (ADS)

Valli, D.; Ganesan, K.

2017-12-01

Chaos based cryptosystems are an efficient method to deal with improved speed and highly secured multimedia encryption because of its elegant features, such as randomness, mixing, ergodicity, sensitivity to initial conditions and control parameters. In this paper, two chaos based cryptosystems are proposed: one is the higher-dimensional 12D chaotic map and the other is based on the Ikeda delay differential equation (DDE) suitable for designing a real-time secure symmetric video encryption scheme. These encryption schemes employ a substitution box (S-box) to diffuse the relationship between pixels of plain video and cipher video along with the diffusion of current input pixel with the previous cipher pixel, called cipher block chaining (CBC). The proposed method enhances the robustness against statistical, differential and chosen/known plain text attacks. Detailed analysis is carried out in this paper to demonstrate the security and uniqueness of the proposed scheme.

Forest Stand Segmentation Using Airborne LIDAR Data and Very High Resolution Multispectral Imagery

NASA Astrophysics Data System (ADS)

Dechesne, Clément; Mallet, Clément; Le Bris, Arnaud; Gouet, Valérie; Hervieu, Alexandre

2016-06-01

Forest stands are the basic units for forest inventory and mapping. Stands are large forested areas (e.g., ≥ 2 ha) of homogeneous tree species composition. The accurate delineation of forest stands is usually performed by visual analysis of human operators on very high resolution (VHR) optical images. This work is highly time consuming and should be automated for scalability purposes. In this paper, a method based on the fusion of airborne laser scanning data (or lidar) and very high resolution multispectral imagery for automatic forest stand delineation and forest land-cover database update is proposed. The multispectral images give access to the tree species whereas 3D lidar point clouds provide geometric information on the trees. Therefore, multi-modal features are computed, both at pixel and object levels. The objects are individual trees extracted from lidar data. A supervised classification is performed at the object level on the computed features in order to coarsely discriminate the existing tree species in the area of interest. The analysis at tree level is particularly relevant since it significantly improves the tree species classification. A probability map is generated through the tree species classification and inserted with the pixel-based features map in an energetical framework. The proposed energy is then minimized using a standard graph-cut method (namely QPBO with α-expansion) in order to produce a segmentation map with a controlled level of details. Comparison with an existing forest land cover database shows that our method provides satisfactory results both in terms of stand labelling and delineation (matching ranges between 94% and 99%).

Crop classification using temporal stacks of multispectral satellite imagery

NASA Astrophysics Data System (ADS)

Moody, Daniela I.; Brumby, Steven P.; Chartrand, Rick; Keisler, Ryan; Longbotham, Nathan; Mertes, Carly; Skillman, Samuel W.; Warren, Michael S.

2017-05-01

The increase in performance, availability, and coverage of multispectral satellite sensor constellations has led to a drastic increase in data volume and data rate. Multi-decadal remote sensing datasets at the petabyte scale are now available in commercial clouds, with new satellite constellations generating petabytes/year of daily high-resolution global coverage imagery. The data analysis capability, however, has lagged behind storage and compute developments, and has traditionally focused on individual scene processing. We present results from an ongoing effort to develop satellite imagery analysis tools that aggregate temporal, spatial, and spectral information and can scale with the high-rate and dimensionality of imagery being collected. We investigate and compare the performance of pixel-level crop identification using tree-based classifiers and its dependence on both temporal and spectral features. Classification performance is assessed using as ground-truth Cropland Data Layer (CDL) crop masks generated by the US Department of Agriculture (USDA). The CDL maps contain 30m spatial resolution, pixel-level labels for around 200 categories of land cover, but are however only available post-growing season. The analysis focuses on McCook county in South Dakota and shows crop classification using a temporal stack of Landsat 8 (L8) imagery over the growing season, from April through October. Specifically, we consider the temporal L8 stack depth, as well as different normalized band difference indices, and evaluate their contribution to crop identification. We also show an extension of our algorithm to map corn and soy crops in the state of Mato Grosso, Brazil.

Graph-Based Semi-Supervised Hyperspectral Image Classification Using Spatial Information

NASA Astrophysics Data System (ADS)

Jamshidpour, N.; Homayouni, S.; Safari, A.

2017-09-01

Hyperspectral image classification has been one of the most popular research areas in the remote sensing community in the past decades. However, there are still some problems that need specific attentions. For example, the lack of enough labeled samples and the high dimensionality problem are two most important issues which degrade the performance of supervised classification dramatically. The main idea of semi-supervised learning is to overcome these issues by the contribution of unlabeled samples, which are available in an enormous amount. In this paper, we propose a graph-based semi-supervised classification method, which uses both spectral and spatial information for hyperspectral image classification. More specifically, two graphs were designed and constructed in order to exploit the relationship among pixels in spectral and spatial spaces respectively. Then, the Laplacians of both graphs were merged to form a weighted joint graph. The experiments were carried out on two different benchmark hyperspectral data sets. The proposed method performed significantly better than the well-known supervised classification methods, such as SVM. The assessments consisted of both accuracy and homogeneity analyses of the produced classification maps. The proposed spectral-spatial SSL method considerably increased the classification accuracy when the labeled training data set is too scarce.When there were only five labeled samples for each class, the performance improved 5.92% and 10.76% compared to spatial graph-based SSL, for AVIRIS Indian Pine and Pavia University data sets respectively.

A neural net based architecture for the segmentation of mixed gray-level and binary pictures

NASA Technical Reports Server (NTRS)

Tabatabai, Ali; Troudet, Terry P.

1991-01-01

A neural-net-based architecture is proposed to perform segmentation in real time for mixed gray-level and binary pictures. In this approach, the composite picture is divided into 16 x 16 pixel blocks, which are identified as character blocks or image blocks on the basis of a dichotomy measure computed by an adaptive 16 x 16 neural net. For compression purposes, each image block is further divided into 4 x 4 subblocks; a one-bit nonparametric quantizer is used to encode 16 x 16 character and 4 x 4 image blocks; and the binary map and quantizer levels are obtained through a neural net segmentor over each block. The efficiency of the neural segmentation in terms of computational speed, data compression, and quality of the compressed picture is demonstrated. The effect of weight quantization is also discussed. VLSI implementations of such adaptive neural nets in CMOS technology are described and simulated in real time for a maximum block size of 256 pixels.

A sun-crown-sensor model and adapted C-correction logic for topographic correction of high resolution forest imagery

NASA Astrophysics Data System (ADS)

Fan, Yuanchao; Koukal, Tatjana; Weisberg, Peter J.

2014-10-01

Canopy shadowing mediated by topography is an important source of radiometric distortion on remote sensing images of rugged terrain. Topographic correction based on the sun-canopy-sensor (SCS) model significantly improved over those based on the sun-terrain-sensor (STS) model for surfaces with high forest canopy cover, because the SCS model considers and preserves the geotropic nature of trees. The SCS model accounts for sub-pixel canopy shadowing effects and normalizes the sunlit canopy area within a pixel. However, it does not account for mutual shadowing between neighboring pixels. Pixel-to-pixel shadowing is especially apparent for fine resolution satellite images in which individual tree crowns are resolved. This paper proposes a new topographic correction model: the sun-crown-sensor (SCnS) model based on high-resolution satellite imagery (IKONOS) and high-precision LiDAR digital elevation model. An improvement on the C-correction logic with a radiance partitioning method to address the effects of diffuse irradiance is also introduced (SCnS + C). In addition, we incorporate a weighting variable, based on pixel shadow fraction, on the direct and diffuse radiance portions to enhance the retrieval of at-sensor radiance and reflectance of highly shadowed tree pixels and form another variety of SCnS model (SCnS + W). Model evaluation with IKONOS test data showed that the new SCnS model outperformed the STS and SCS models in quantifying the correlation between terrain-regulated illumination factor and at-sensor radiance. Our adapted C-correction logic based on the sun-crown-sensor geometry and radiance partitioning better represented the general additive effects of diffuse radiation than C parameters derived from the STS or SCS models. The weighting factor Wt also significantly enhanced correction results by reducing within-class standard deviation and balancing the mean pixel radiance between sunlit and shaded slopes. We analyzed these improvements with model comparison on the red and near infrared bands. The advantages of SCnS + C and SCnS + W on both bands are expected to facilitate forest classification and change detection applications.

Rethinking Skin Lesion Segmentation in a Convolutional Classifier.

PubMed

Burdick, Jack; Marques, Oge; Weinthal, Janet; Furht, Borko

2017-10-18

Melanoma is a fatal form of skin cancer when left undiagnosed. Computer-aided diagnosis systems powered by convolutional neural networks (CNNs) can improve diagnostic accuracy and save lives. CNNs have been successfully used in both skin lesion segmentation and classification. For reasons heretofore unclear, previous works have found image segmentation to be, conflictingly, both detrimental and beneficial to skin lesion classification. We investigate the effect of expanding the segmentation border to include pixels surrounding the target lesion. Ostensibly, segmenting a target skin lesion will remove inessential information, non-lesion skin, and artifacts to aid in classification. Our results indicate that segmentation border enlargement produces, to a certain degree, better results across all metrics of interest when using a convolutional based classifier built using the transfer learning paradigm. Consequently, preprocessing methods which produce borders larger than the actual lesion can potentially improve classifier performance, more than both perfect segmentation, using dermatologist created ground truth masks, and no segmentation altogether.

Directional Multi-scale Modeling of High-Resolution Computed Tomography (HRCT) Lung Images for Diffuse Lung Disease Classification

NASA Astrophysics Data System (ADS)

Vo, Kiet T.; Sowmya, Arcot

A directional multi-scale modeling scheme based on wavelet and contourlet transforms is employed to describe HRCT lung image textures for classifying four diffuse lung disease patterns: normal, emphysema, ground glass opacity (GGO) and honey-combing. Generalized Gaussian density parameters are used to represent the detail sub-band features obtained by wavelet and contourlet transforms. In addition, support vector machines (SVMs) with excellent performance in a variety of pattern classification problems are used as classifier. The method is tested on a collection of 89 slices from 38 patients, each slice of size 512x512, 16 bits/pixel in DICOM format. The dataset contains 70,000 ROIs of those slices marked by experienced radiologists. We employ this technique at different wavelet and contourlet transform scales for diffuse lung disease classification. The technique presented here has best overall sensitivity 93.40% and specificity 98.40%.

Alternative Optimizations of X-ray TES Arrays: Soft X-rays, High Count Rates, and Mixed-Pixel Arrays

NASA Technical Reports Server (NTRS)

Kilbourne, C. A.; Bandler, S. R.; Brown, A.-D.; Chervenak, J. A.; Figueroa-Feliciano, E.; Finkbeiner, F. M.; Iyomoto, N.; Kelley, R. L.; Porter, F. S.; Smith, S. J.

2007-01-01

We are developing arrays of superconducting transition-edge sensors (TES) for imaging spectroscopy telescopes such as the XMS on Constellation-X. While our primary focus has been on arrays that meet the XMS requirements (of which, foremost, is an energy resolution of 2.5 eV at 6 keV and a bandpass from approx. 0.3 keV to 12 keV), we have also investigated other optimizations that might be used to extend the XMS capabilities. In one of these optimizations, improved resolution below 1 keV is achieved by reducing the heat capacity. Such pixels can be based on our XMS-style TES's with the separate absorbers omitted. These pixels can added to an array with broadband response either as a separate array or interspersed, depending on other factors that include telescope design and science requirements. In one version of this approach, we have designed and fabricated a composite array of low-energy and broad-band pixels to provide high spectral resolving power over a broader energy bandpass than could be obtained with a single TES design. The array consists of alternating pixels with and without overhanging absorbers. To explore optimizations for higher count rates, we are also optimizing the design and operating temperature of pixels that are coupled to a solid substrate. We will present the performance of these variations and discuss other optimizations that could be used to enhance the XMS or enable other astrophysics experiments.

High-dynamic-range coherent diffractive imaging: ptychography using the mixed-mode pixel array detector.

PubMed

Giewekemeyer, Klaus; Philipp, Hugh T; Wilke, Robin N; Aquila, Andrew; Osterhoff, Markus; Tate, Mark W; Shanks, Katherine S; Zozulya, Alexey V; Salditt, Tim; Gruner, Sol M; Mancuso, Adrian P

2014-09-01

Coherent (X-ray) diffractive imaging (CDI) is an increasingly popular form of X-ray microscopy, mainly due to its potential to produce high-resolution images and the lack of an objective lens between the sample and its corresponding imaging detector. One challenge, however, is that very high dynamic range diffraction data must be collected to produce both quantitative and high-resolution images. In this work, hard X-ray ptychographic coherent diffractive imaging has been performed at the P10 beamline of the PETRA III synchrotron to demonstrate the potential of a very wide dynamic range imaging X-ray detector (the Mixed-Mode Pixel Array Detector, or MM-PAD). The detector is capable of single photon detection, detecting fluxes exceeding 1 × 10(8) 8-keV photons pixel(-1) s(-1), and framing at 1 kHz. A ptychographic reconstruction was performed using a peak focal intensity on the order of 1 × 10(10) photons µm(-2) s(-1) within an area of approximately 325 nm × 603 nm. This was done without need of a beam stop and with a very modest attenuation, while `still' images of the empty beam far-field intensity were recorded without any attenuation. The treatment of the detector frames and CDI methodology for reconstruction of non-sensitive detector regions, partially also extending the active detector area, are described.

Prevalence of Mixed-Methods Sampling Designs in Social Science Research

ERIC Educational Resources Information Center

Collins, Kathleen M. T.

2006-01-01

The purpose of this mixed-methods study was to document the prevalence of sampling designs utilised in mixed-methods research and to examine the interpretive consistency between interpretations made in mixed-methods studies and the sampling design used. Classification of studies was based on a two-dimensional mixed-methods sampling model. This…

Spectral Target Detection using Schroedinger Eigenmaps

NASA Astrophysics Data System (ADS)

Dorado-Munoz, Leidy P.

Applications of optical remote sensing processes include environmental monitoring, military monitoring, meteorology, mapping, surveillance, etc. Many of these tasks include the detection of specific objects or materials, usually few or small, which are surrounded by other materials that clutter the scene and hide the relevant information. This target detection process has been boosted lately by the use of hyperspectral imagery (HSI) since its high spectral dimension provides more detailed spectral information that is desirable in data exploitation. Typical spectral target detectors rely on statistical or geometric models to characterize the spectral variability of the data. However, in many cases these parametric models do not fit well HSI data that impacts the detection performance. On the other hand, non-linear transformation methods, mainly based on manifold learning algorithms, have shown a potential use in HSI transformation, dimensionality reduction and classification. In target detection, non-linear transformation algorithms are used as preprocessing techniques that transform the data to a more suitable lower dimensional space, where the statistical or geometric detectors are applied. One of these non-linear manifold methods is the Schroedinger Eigenmaps (SE) algorithm that has been introduced as a technique for semi-supervised classification. The core tool of the SE algorithm is the Schroedinger operator that includes a potential term that encodes prior information about the materials present in a scene, and enables the embedding to be steered in some convenient directions in order to cluster similar pixels together. A completely novel target detection methodology based on SE algorithm is proposed for the first time in this thesis. The proposed methodology does not just include the transformation of the data to a lower dimensional space but also includes the definition of a detector that capitalizes on the theory behind SE. The fact that target pixels and those similar pixels are clustered in a predictable region of the low-dimensional representation is used to define a decision rule that allows one to identify target pixels over the rest of pixels in a given image. In addition, a knowledge propagation scheme is used to combine spectral and spatial information as a means to propagate the "potential constraints" to nearby points. The propagation scheme is introduced to reinforce weak connections and improve the separability between most of the target pixels and the background. Experiments using different HSI data sets are carried out in order to test the proposed methodology. The assessment is performed from a quantitative and qualitative point of view, and by comparing the SE-based methodology against two other detection methodologies that use linear/non-linear algorithms as transformations and the well-known Adaptive Coherence/Cosine Estimator (ACE) detector. Overall results show that the SE-based detector outperforms the other two detection methodologies, which indicates the usefulness of the SE transformation in spectral target detection problems.

Capability of Integrated MODIS Imagery and ALOS for Oil Palm, Rubber and Forest Areas Mapping in Tropical Forest Regions

PubMed Central

Razali, Sheriza Mohd; Marin, Arnaldo; Nuruddin, Ahmad Ainuddin; Shafri, Helmi Zulhaidi Mohd; Hamid, Hazandy Abdul

2014-01-01

Various classification methods have been applied for low resolution of the entire Earth's surface from recorded satellite images, but insufficient study has determined which method, for which satellite data, is economically viable for tropical forest land use mapping. This study employed Iterative Self Organizing Data Analysis Techniques (ISODATA) and K-Means classification techniques to classified Moderate Resolution Imaging Spectroradiometer (MODIS) Surface Reflectance satellite image into forests, oil palm groves, rubber plantations, mixed horticulture, mixed oil palm and rubber and mixed forest and rubber. Even though frequent cloud cover has been a challenge for mapping tropical forests, our MODIS land use classification map found that 2008 ISODATA-1 performed well with overall accuracy of 94%, with the highest Producer's Accuracy of Forest with 86%, and were consistent with MODIS Land Cover 2008 (MOD12Q1), respectively. The MODIS land use classification was able to distinguish young oil palm groves from open areas, rubber and mature oil palm plantations, on the Advanced Land Observing Satellite (ALOS) map, whereas rubber was more easily distinguished from an open area than from mixed rubber and forest. This study provides insight on the potential for integrating regional databases and temporal MODIS data, in order to map land use in tropical forest regions. PMID:24811079

Capability of integrated MODIS imagery and ALOS for oil palm, rubber and forest areas mapping in tropical forest regions.

PubMed

Razali, Sheriza Mohd; Marin, Arnaldo; Nuruddin, Ahmad Ainuddin; Shafri, Helmi Zulhaidi Mohd; Hamid, Hazandy Abdul

2014-05-07

Various classification methods have been applied for low resolution of the entire Earth's surface from recorded satellite images, but insufficient study has determined which method, for which satellite data, is economically viable for tropical forest land use mapping. This study employed Iterative Self Organizing Data Analysis Techniques (ISODATA) and K-Means classification techniques to classified Moderate Resolution Imaging Spectroradiometer (MODIS) Surface Reflectance satellite image into forests, oil palm groves, rubber plantations, mixed horticulture, mixed oil palm and rubber and mixed forest and rubber. Even though frequent cloud cover has been a challenge for mapping tropical forests, our MODIS land use classification map found that 2008 ISODATA-1 performed well with overall accuracy of 94%, with the highest Producer's Accuracy of Forest with 86%, and were consistent with MODIS Land Cover 2008 (MOD12Q1), respectively. The MODIS land use classification was able to distinguish young oil palm groves from open areas, rubber and mature oil palm plantations, on the Advanced Land Observing Satellite (ALOS) map, whereas rubber was more easily distinguished from an open area than from mixed rubber and forest. This study provides insight on the potential for integrating regional databases and temporal MODIS data, in order to map land use in tropical forest regions.

Automatic classification of retinal vessels into arteries and veins

NASA Astrophysics Data System (ADS)

Niemeijer, Meindert; van Ginneken, Bram; Abràmoff, Michael D.

2009-02-01

Separating the retinal vascular tree into arteries and veins is important for quantifying vessel changes that preferentially affect either the veins or the arteries. For example the ratio of arterial to venous diameter, the retinal a/v ratio, is well established to be predictive of stroke and other cardiovascular events in adults, as well as the staging of retinopathy of prematurity in premature infants. This work presents a supervised, automatic method that can determine whether a vessel is an artery or a vein based on intensity and derivative information. After thinning of the vessel segmentation, vessel crossing and bifurcation points are removed leaving a set of vessel segments containing centerline pixels. A set of features is extracted from each centerline pixel and using these each is assigned a soft label indicating the likelihood that it is part of a vein. As all centerline pixels in a connected segment should be the same type we average the soft labels and assign this average label to each centerline pixel in the segment. We train and test the algorithm using the data (40 color fundus photographs) from the DRIVE database1 with an enhanced reference standard. In the enhanced reference standard a fellowship trained retinal specialist (MDA) labeled all vessels for which it was possible to visually determine whether it was a vein or an artery. After applying the proposed method to the 20 images of the DRIVE test set we obtained an area under the receiver operator characteristic (ROC) curve of 0.88 for correctly assigning centerline pixels to either the vein or artery classes.

Urban land-cover change detection through sub-pixel imperviousness mapping using remotely sensed data

USGS Publications Warehouse

Yang, Limin; Xian, George Z.; Klaver, Jacqueline M.; Deal, Brian

2003-01-01

We developed a Sub-pixel Imperviousness Change Detection (SICD) approach to detect urban land-cover changes using Landsat and high-resolution imagery. The sub-pixel percent imperviousness was mapped for two dates (09 March 1993 and 11 March 2001) over western Georgia using a regression tree algorithm. The accuracy of the predicted imperviousness was reasonable based on a comparison using independent reference data. The average absolute error between predicted and reference data was 16.4 percent for 1993 and 15.3 percent for 2001. The correlation coefficient (r) was 0.73 for 1993 and 0.78 for 2001, respectively. Areas with a significant increase (greater than 20 percent) in impervious surface from 1993 to 2001 were mostly related to known land-cover/land-use changes that occurred in this area, suggesting that the spatial change of an impervious surface is a useful indicator for identifying spatial extent, intensity, and, potentially, type of urban land-cover/land-use changes. Compared to other pixel-based change-detection methods (band differencing, rationing, change vector, post-classification), information on changes in sub-pixel percent imperviousness allow users to quantify and interpret urban land-cover/land-use changes based on their own definition. Such information is considered complementary to products generated using other change-detection methods. In addition, the procedure for mapping imperviousness is objective and repeatable, hence, can be used for monitoring urban land-cover/land-use change over a large geographic area. Potential applications and limitations of the products developed through this study in urban environmental studies are also discussed.

A Comparative Analysis of Machine Learning with WorldView-2 Pan-Sharpened Imagery for Tea Crop Mapping

PubMed Central

Chuang, Yung-Chung Matt; Shiu, Yi-Shiang

2016-01-01

Tea is an important but vulnerable economic crop in East Asia, highly impacted by climate change. This study attempts to interpret tea land use/land cover (LULC) using very high resolution WorldView-2 imagery of central Taiwan with both pixel and object-based approaches. A total of 80 variables derived from each WorldView-2 band with pan-sharpening, standardization, principal components and gray level co-occurrence matrix (GLCM) texture indices transformation, were set as the input variables. For pixel-based image analysis (PBIA), 34 variables were selected, including seven principal components, 21 GLCM texture indices and six original WorldView-2 bands. Results showed that support vector machine (SVM) had the highest tea crop classification accuracy (OA = 84.70% and KIA = 0.690), followed by random forest (RF), maximum likelihood algorithm (ML), and logistic regression analysis (LR). However, the ML classifier achieved the highest classification accuracy (OA = 96.04% and KIA = 0.887) in object-based image analysis (OBIA) using only six variables. The contribution of this study is to create a new framework for accurately identifying tea crops in a subtropical region with real-time high-resolution WorldView-2 imagery without field survey, which could further aid agriculture land management and a sustainable agricultural product supply. PMID:27128915

A Comparative Analysis of Machine Learning with WorldView-2 Pan-Sharpened Imagery for Tea Crop Mapping.

PubMed

Chuang, Yung-Chung Matt; Shiu, Yi-Shiang

2016-04-26

Tea is an important but vulnerable economic crop in East Asia, highly impacted by climate change. This study attempts to interpret tea land use/land cover (LULC) using very high resolution WorldView-2 imagery of central Taiwan with both pixel and object-based approaches. A total of 80 variables derived from each WorldView-2 band with pan-sharpening, standardization, principal components and gray level co-occurrence matrix (GLCM) texture indices transformation, were set as the input variables. For pixel-based image analysis (PBIA), 34 variables were selected, including seven principal components, 21 GLCM texture indices and six original WorldView-2 bands. Results showed that support vector machine (SVM) had the highest tea crop classification accuracy (OA = 84.70% and KIA = 0.690), followed by random forest (RF), maximum likelihood algorithm (ML), and logistic regression analysis (LR). However, the ML classifier achieved the highest classification accuracy (OA = 96.04% and KIA = 0.887) in object-based image analysis (OBIA) using only six variables. The contribution of this study is to create a new framework for accurately identifying tea crops in a subtropical region with real-time high-resolution WorldView-2 imagery without field survey, which could further aid agriculture land management and a sustainable agricultural product supply.

Segment fusion of ToF-SIMS images.

PubMed

Milillo, Tammy M; Miller, Mary E; Fischione, Remo; Montes, Angelina; Gardella, Joseph A

2016-06-08

The imaging capabilities of time-of-flight secondary ion mass spectrometry (ToF-SIMS) have not been used to their full potential in the analysis of polymer and biological samples. Imaging has been limited by the size of the dataset and the chemical complexity of the sample being imaged. Pixel and segment based image fusion algorithms commonly used in remote sensing, ecology, geography, and geology provide a way to improve spatial resolution and classification of biological images. In this study, a sample of Arabidopsis thaliana was treated with silver nanoparticles and imaged with ToF-SIMS. These images provide insight into the uptake mechanism for the silver nanoparticles into the plant tissue, giving new understanding to the mechanism of uptake of heavy metals in the environment. The Munechika algorithm was programmed in-house and applied to achieve pixel based fusion, which improved the spatial resolution of the image obtained. Multispectral and quadtree segment or region based fusion algorithms were performed using ecognition software, a commercially available remote sensing software suite, and used to classify the images. The Munechika fusion improved the spatial resolution for the images containing silver nanoparticles, while the segment fusion allowed classification and fusion based on the tissue types in the sample, suggesting potential pathways for the uptake of the silver nanoparticles.

Comparative study of different approaches for multivariate image analysis in HPTLC fingerprinting of natural products such as plant resin.

PubMed

Ristivojević, Petar; Trifković, Jelena; Vovk, Irena; Milojković-Opsenica, Dušanka

2017-01-01

Considering the introduction of phytochemical fingerprint analysis, as a method of screening the complex natural products for the presence of most bioactive compounds, use of chemometric classification methods, application of powerful scanning and image capturing and processing devices and algorithms, advancement in development of novel stationary phases as well as various separation modalities, high-performance thin-layer chromatography (HPTLC) fingerprinting is becoming attractive and fruitful field of separation science. Multivariate image analysis is crucial in the light of proper data acquisition. In a current study, different image processing procedures were studied and compared in detail on the example of HPTLC chromatograms of plant resins. In that sense, obtained variables such as gray intensities of pixels along the solvent front, peak area and mean values of peak were used as input data and compared to obtained best classification models. Important steps in image analysis, baseline removal, denoising, target peak alignment and normalization were pointed out. Numerical data set based on mean value of selected bands and intensities of pixels along the solvent front proved to be the most convenient for planar-chromatographic profiling, although required at least the basic knowledge on image processing methodology, and could be proposed for further investigation in HPLTC fingerprinting. Copyright © 2016 Elsevier B.V. All rights reserved.

The fusion of large scale classified side-scan sonar image mosaics.

PubMed

Reed, Scott; Tena, Ruiz Ioseba; Capus, Chris; Petillot, Yvan

2006-07-01

This paper presents a unified framework for the creation of classified maps of the seafloor from sonar imagery. Significant challenges in photometric correction, classification, navigation and registration, and image fusion are addressed. The techniques described are directly applicable to a range of remote sensing problems. Recent advances in side-scan data correction are incorporated to compensate for the sonar beam pattern and motion of the acquisition platform. The corrected images are segmented using pixel-based textural features and standard classifiers. In parallel, the navigation of the sonar device is processed using Kalman filtering techniques. A simultaneous localization and mapping framework is adopted to improve the navigation accuracy and produce georeferenced mosaics of the segmented side-scan data. These are fused within a Markovian framework and two fusion models are presented. The first uses a voting scheme regularized by an isotropic Markov random field and is applicable when the reliability of each information source is unknown. The Markov model is also used to inpaint regions where no final classification decision can be reached using pixel level fusion. The second model formally introduces the reliability of each information source into a probabilistic model. Evaluation of the two models using both synthetic images and real data from a large scale survey shows significant quantitative and qualitative improvement using the fusion approach.

Seasonal cultivated and fallow cropland mapping using MODIS-based automated cropland classification algorithm

USGS Publications Warehouse

Wu, Zhuoting; Thenkabail, Prasad S.; Mueller, Rick; Zakzeski, Audra; Melton, Forrest; Johnson, Lee; Rosevelt, Carolyn; Dwyer, John; Jones, Jeanine; Verdin, James P.

2014-01-01

Increasing drought occurrences and growing populations demand accurate, routine, and consistent cultivated and fallow cropland products to enable water and food security analysis. The overarching goal of this research was to develop and test automated cropland classification algorithm (ACCA) that provide accurate, consistent, and repeatable information on seasonal cultivated as well as seasonal fallow cropland extents and areas based on the Moderate Resolution Imaging Spectroradiometer remote sensing data. Seasonal ACCA development process involves writing series of iterative decision tree codes to separate cultivated and fallow croplands from noncroplands, aiming to accurately mirror reliable reference data sources. A pixel-by-pixel accuracy assessment when compared with the U.S. Department of Agriculture (USDA) cropland data showed, on average, a producer’s accuracy of 93% and a user’s accuracy of 85% across all months. Further, ACCA-derived cropland maps agreed well with the USDA Farm Service Agency crop acreage-reported data for both cultivated and fallow croplands with R-square values over 0.7 and field surveys with an accuracy of ≥95% for cultivated croplands and ≥76% for fallow croplands. Our results demonstrated the ability of ACCA to generate cropland products, such as cultivated and fallow cropland extents and areas, accurately, automatically, and repeatedly throughout the growing season.

Semantic Building FAÇADE Segmentation from Airborne Oblique Images

NASA Astrophysics Data System (ADS)

Lin, Y.; Nex, F.; Yang, M. Y.

2018-05-01

With the introduction of airborne oblique camera systems and the improvement of photogrammetric techniques, high-resolution 2D and 3D data can be acquired in urban areas. This high-resolution data allows us to perform detailed investigations on building roofs and façades which can contribute to LoD3 city modeling. Normally, façade segmentation is achieved from terrestrial views. In this paper, we address the problem from aerial views by using high resolution oblique aerial images as the data source in urban areas. In addition to traditional image features, such as RGB and SIFT, normal vector and planarity are also extracted from dense matching point clouds. Then, these 3D geometrical features are projected back to 2D space to assist façade interpretation. Random forest is trained and applied to label façade pixels. Fully connected conditional random field (CRF), capturing long-range spatial interactions, is used as a post-processing to refine our classification results. Its pairwise potential is defined by a linear combination of Gaussian kernels and the CRF model is efficiently solved by mean field approximation. Experiments show that 3D features can significantly improve classification results. Also, fully connected CRF performs well in correcting noisy pixels.