Local dynamic range compensation for scanning electron microscope imaging system by sub-blocking multiple peak HE with convolution.

PubMed

Sim, K S; Teh, V; Tey, Y C; Kho, T K

2016-11-01

This paper introduces new development technique to improve the Scanning Electron Microscope (SEM) image quality and we name it as sub-blocking multiple peak histogram equalization (SUB-B-MPHE) with convolution operator. By using this new proposed technique, it shows that the new modified MPHE performs better than original MPHE. In addition, the sub-blocking method consists of convolution operator which can help to remove the blocking effect for SEM images after applying this new developed technique. Hence, by using the convolution operator, it effectively removes the blocking effect by properly distributing the suitable pixel value for the whole image. Overall, the SUB-B-MPHE with convolution outperforms the rest of methods. SCANNING 38:492-501, 2016. © 2015 Wiley Periodicals, Inc. © Wiley Periodicals, Inc.

Convolutional Neural Networks for 1-D Many-Channel Data

DTIC Science & Technology

Deep convolutional neural networks (CNNs) represent the state of the art in image recognition. The same properties that led to their success in that... crack detection ( 8,000 data points, 72 channels). Though the models predictive ability is limited to fitting the trend , its partial success suggests that...originally written to classify digits in the MNIST database (28 28 pixels, 1 channel), for use on 1-D acoustic data taken from experiments focused on

Convolutional networks for vehicle track segmentation

NASA Astrophysics Data System (ADS)

Quach, Tu-Thach

2017-10-01

Existing methods to detect vehicle tracks in coherent change detection images, a product of combining two synthetic aperture radar images taken at different times of the same scene, rely on simple and fast models to label track pixels. These models, however, are unable to capture natural track features, such as continuity and parallelism. More powerful but computationally expensive models can be used in offline settings. We present an approach that uses dilated convolutional networks consisting of a series of 3×3 convolutions to segment vehicle tracks. The design of our networks considers the fact that remote sensing applications tend to operate in low power and have limited training data. As a result, we aim for small and efficient networks that can be trained end-to-end to learn natural track features entirely from limited training data. We demonstrate that our six-layer network, trained on just 90 images, is computationally efficient and improves the F-score on a standard dataset to 0.992, up from 0.959 obtained by the current state-of-the-art method.

Convolutional networks for vehicle track segmentation

DOE PAGES

Quach, Tu-Thach

2017-08-19

Existing methods to detect vehicle tracks in coherent change detection images, a product of combining two synthetic aperture radar images taken at different times of the same scene, rely on simple, fast models to label track pixels. These models, however, are unable to capture natural track features such as continuity and parallelism. More powerful, but computationally expensive models can be used in offline settings. We present an approach that uses dilated convolutional networks consisting of a series of 3-by-3 convolutions to segment vehicle tracks. The design of our networks considers the fact that remote sensing applications tend to operate inmore » low power and have limited training data. As a result, we aim for small, efficient networks that can be trained end-to-end to learn natural track features entirely from limited training data. We demonstrate that our 6-layer network, trained on just 90 images, is computationally efficient and improves the F-score on a standard dataset to 0.992, up from 0.959 obtained by the current state-of-the-art method.« less

Convolutional networks for vehicle track segmentation

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

Quach, Tu-Thach

Existing methods to detect vehicle tracks in coherent change detection images, a product of combining two synthetic aperture radar images taken at different times of the same scene, rely on simple, fast models to label track pixels. These models, however, are unable to capture natural track features such as continuity and parallelism. More powerful, but computationally expensive models can be used in offline settings. We present an approach that uses dilated convolutional networks consisting of a series of 3-by-3 convolutions to segment vehicle tracks. The design of our networks considers the fact that remote sensing applications tend to operate inmore » low power and have limited training data. As a result, we aim for small, efficient networks that can be trained end-to-end to learn natural track features entirely from limited training data. We demonstrate that our 6-layer network, trained on just 90 images, is computationally efficient and improves the F-score on a standard dataset to 0.992, up from 0.959 obtained by the current state-of-the-art method.« less

A fully convolutional networks (FCN) based image segmentation algorithm in binocular imaging system

NASA Astrophysics Data System (ADS)

Long, Zourong; Wei, Biao; Feng, Peng; Yu, Pengwei; Liu, Yuanyuan

2018-01-01

This paper proposes an image segmentation algorithm with fully convolutional networks (FCN) in binocular imaging system under various circumstance. Image segmentation is perfectly solved by semantic segmentation. FCN classifies the pixels, so as to achieve the level of image semantic segmentation. Different from the classical convolutional neural networks (CNN), FCN uses convolution layers instead of the fully connected layers. So it can accept image of arbitrary size. In this paper, we combine the convolutional neural network and scale invariant feature matching to solve the problem of visual positioning under different scenarios. All high-resolution images are captured with our calibrated binocular imaging system and several groups of test data are collected to verify this method. The experimental results show that the binocular images are effectively segmented without over-segmentation. With these segmented images, feature matching via SURF method is implemented to obtain regional information for further image processing. The final positioning procedure shows that the results are acceptable in the range of 1.4 1.6 m, the distance error is less than 10mm.

A semiconductor radiation imaging pixel detector for space radiation dosimetry.

PubMed

Kroupa, Martin; Bahadori, Amir; Campbell-Ricketts, Thomas; Empl, Anton; Hoang, Son Minh; Idarraga-Munoz, John; Rios, Ryan; Semones, Edward; Stoffle, Nicholas; Tlustos, Lukas; Turecek, Daniel; Pinsky, Lawrence

2015-07-01

Progress in the development of high-performance semiconductor radiation imaging pixel detectors based on technologies developed for use in high-energy physics applications has enabled the development of a completely new generation of compact low-power active dosimeters and area monitors for use in space radiation environments. Such detectors can provide real-time information concerning radiation exposure, along with detailed analysis of the individual particles incident on the active medium. Recent results from the deployment of detectors based on the Timepix from the CERN-based Medipix2 Collaboration on the International Space Station (ISS) are reviewed, along with a glimpse of developments to come. Preliminary results from Orion MPCV Exploration Flight Test 1 are also presented. Copyright © 2015 The Committee on Space Research (COSPAR). All rights reserved.

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.

A semiconductor radiation imaging pixel detector for space radiation dosimetry

NASA Astrophysics Data System (ADS)

Kroupa, Martin; Bahadori, Amir; Campbell-Ricketts, Thomas; Empl, Anton; Hoang, Son Minh; Idarraga-Munoz, John; Rios, Ryan; Semones, Edward; Stoffle, Nicholas; Tlustos, Lukas; Turecek, Daniel; Pinsky, Lawrence

2015-07-01

Progress in the development of high-performance semiconductor radiation imaging pixel detectors based on technologies developed for use in high-energy physics applications has enabled the development of a completely new generation of compact low-power active dosimeters and area monitors for use in space radiation environments. Such detectors can provide real-time information concerning radiation exposure, along with detailed analysis of the individual particles incident on the active medium. Recent results from the deployment of detectors based on the Timepix from the CERN-based Medipix2 Collaboration on the International Space Station (ISS) are reviewed, along with a glimpse of developments to come. Preliminary results from Orion MPCV Exploration Flight Test 1 are also presented.

Feasibility study of multi-pixel retrieval of optical thickness and droplet effective radius of inhomogeneous clouds using deep learning

NASA Astrophysics Data System (ADS)

Okamura, Rintaro; Iwabuchi, Hironobu; Schmidt, K. Sebastian

2017-12-01

Three-dimensional (3-D) radiative-transfer effects are a major source of retrieval errors in satellite-based optical remote sensing of clouds. The challenge is that 3-D effects manifest themselves across multiple satellite pixels, which traditional single-pixel approaches cannot capture. In this study, we present two multi-pixel retrieval approaches based on deep learning, a technique that is becoming increasingly successful for complex problems in engineering and other areas. Specifically, we use deep neural networks (DNNs) to obtain multi-pixel estimates of cloud optical thickness and column-mean cloud droplet effective radius from multispectral, multi-pixel radiances. The first DNN method corrects traditional bispectral retrievals based on the plane-parallel homogeneous cloud assumption using the reflectances at the same two wavelengths. The other DNN method uses so-called convolutional layers and retrieves cloud properties directly from the reflectances at four wavelengths. The DNN methods are trained and tested on cloud fields from large-eddy simulations used as input to a 3-D radiative-transfer model to simulate upward radiances. The second DNN-based retrieval, sidestepping the bispectral retrieval step through convolutional layers, is shown to be more accurate. It reduces 3-D radiative-transfer effects that would otherwise affect the radiance values and estimates cloud properties robustly even for optically thick clouds.

Hamiltonian Cycle Enumeration via Fermion-Zeon Convolution

NASA Astrophysics Data System (ADS)

Staples, G. Stacey

2017-12-01

Beginning with a simple graph having finite vertex set V, operators are induced on fermion and zeon algebras by the action of the graph's adjacency matrix and combinatorial Laplacian on the vector space spanned by the graph's vertices. When the graph is simple (undirected with no loops or multiple edges), the matrices are symmetric and the induced operators are self-adjoint. The goal of the current paper is to recover a number of known graph-theoretic results from quantum observables constructed as linear operators on fermion and zeon Fock spaces. By considering an "indeterminate" fermion/zeon Fock space, a fermion-zeon convolution operator is defined whose trace recovers the number of Hamiltonian cycles in the graph. This convolution operator is a quantum observable whose expectation reveals the number of Hamiltonian cycles in the graph.

Supervised local error estimation for nonlinear image registration using convolutional neural networks

NASA Astrophysics Data System (ADS)

Eppenhof, Koen A. J.; Pluim, Josien P. W.

2017-02-01

Error estimation in medical image registration is valuable when validating, comparing, or combining registration methods. To validate a nonlinear image registration method, ideally the registration error should be known for the entire image domain. We propose a supervised method for the estimation of a registration error map for nonlinear image registration. The method is based on a convolutional neural network that estimates the norm of the residual deformation from patches around each pixel in two registered images. This norm is interpreted as the registration error, and is defined for every pixel in the image domain. The network is trained using a set of artificially deformed images. Each training example is a pair of images: the original image, and a random deformation of that image. No manually labeled ground truth error is required. At test time, only the two registered images are required as input. We train and validate the network on registrations in a set of 2D digital subtraction angiography sequences, such that errors up to eight pixels can be estimated. We show that for this range of errors the convolutional network is able to learn the registration error in pairs of 2D registered images at subpixel precision. Finally, we present a proof of principle for the extension to 3D registration problems in chest CTs, showing that the method has the potential to estimate errors in 3D registration problems.

Cerebral vessels segmentation for light-sheet microscopy image using convolutional neural networks

NASA Astrophysics Data System (ADS)

Hu, Chaoen; Hui, Hui; Wang, Shuo; Dong, Di; Liu, Xia; Yang, Xin; Tian, Jie

2017-03-01

Cerebral vessel segmentation is an important step in image analysis for brain function and brain disease studies. To extract all the cerebrovascular patterns, including arteries and capillaries, some filter-based methods are used to segment vessels. However, the design of accurate and robust vessel segmentation algorithms is still challenging, due to the variety and complexity of images, especially in cerebral blood vessel segmentation. In this work, we addressed a problem of automatic and robust segmentation of cerebral micro-vessels structures in cerebrovascular images acquired by light-sheet microscope for mouse. To segment micro-vessels in large-scale image data, we proposed a convolutional neural networks (CNNs) architecture trained by 1.58 million pixels with manual label. Three convolutional layers and one fully connected layer were used in the CNNs model. We extracted a patch of size 32x32 pixels in each acquired brain vessel image as training data set to feed into CNNs for classification. This network was trained to output the probability that the center pixel of input patch belongs to vessel structures. To build the CNNs architecture, a series of mouse brain vascular images acquired from a commercial light sheet fluorescence microscopy (LSFM) system were used for training the model. The experimental results demonstrated that our approach is a promising method for effectively segmenting micro-vessels structures in cerebrovascular images with vessel-dense, nonuniform gray-level and long-scale contrast regions.

A kilo-pixel imaging system for future space based far-infrared observatories using microwave kinetic inductance detectors

NASA Astrophysics Data System (ADS)

Baselmans, J. J. A.; Bueno, J.; Yates, S. J. C.; Yurduseven, O.; Llombart, N.; Karatsu, K.; Baryshev, A. M.; Ferrari, L.; Endo, A.; Thoen, D. J.; de Visser, P. J.; Janssen, R. M. J.; Murugesan, V.; Driessen, E. F. C.; Coiffard, G.; Martin-Pintado, J.; Hargrave, P.; Griffin, M.

2017-05-01

Aims: Future astrophysics and cosmic microwave background space missions operating in the far-infrared to millimetre part of the spectrum will require very large arrays of ultra-sensitive detectors in combination with high multiplexing factors and efficient low-noise and low-power readout systems. We have developed a demonstrator system suitable for such applications. Methods: The system combines a 961 pixel imaging array based upon Microwave Kinetic Inductance Detectors (MKIDs) with a readout system capable of reading out all pixels simultaneously with only one readout cable pair and a single cryogenic amplifier. We evaluate, in a representative environment, the system performance in terms of sensitivity, dynamic range, optical efficiency, cosmic ray rejection, pixel-pixel crosstalk and overall yield at an observation centre frequency of 850 GHz and 20% fractional bandwidth. Results: The overall system has an excellent sensitivity, with an average detector sensitivity < NEPdet> =3×10-19 WHz measured using a thermal calibration source. At a loading power per pixel of 50 fW we demonstrate white, photon noise limited detector noise down to 300 mHz. The dynamic range would allow the detection of 1 Jy bright sources within the field of view without tuning the readout of the detectors. The expected dead time due to cosmic ray interactions, when operated in an L2 or a similar far-Earth orbit, is found to be <4%. Additionally, the achieved pixel yield is 83% and the crosstalk between the pixels is <-30 dB. Conclusions: This demonstrates that MKID technology can provide multiplexing ratios on the order of a 1000 with state-of-the-art single pixel performance, and that the technology is now mature enough to be considered for future space based observatories and experiments.

Pixel-by-Pixel SED Fitting of Intermediate Redshift Galaxies

NASA Astrophysics Data System (ADS)

Cohen, Seth H.; Kim, Hwihyun; Petty, Sara M.; Farrah, Duncan

2015-01-01

We select intermediate redshift galaxies from the Hubble Space Telescope CANDELS and GOODS surveys to study their stellar populations on sub-kilo-parsec scales by fitting SED models on a pixel-by-pixel basis. Galaxies are chosen to have measured spectroscopic redshifts (z<1.5), to be bright (H_AB<21 mag), to be relatively face-on (b/a > 0.6), and have a minimum of ten individual resolution elements across the face of the galaxy, as defined by the broadest PSF (F160W-band) in the data. The sample contains ~200 galaxies with BViz(Y)JH band HST photometry. The main goal of the study is to better understand the effects of population blending when using a pixel-by-pixel SED fitting (pSED) approach. We outline our pSED fitting method which gives maps of stellar mass, age, star-formation rate, etc. Several examples of individual pSED-fit maps are presented in detail, as well as some preliminary results on the full sample. The pSED method is necessarily biased by the brightest population in a given pixel outshining the rest of the stars, and, therefore, we intend to study this apparent population blending in a set of artificially redshifted images of nearby galaxies, for which we have star-by-star measurements of their stellar populations. This local sample will be used to better interpret the measurements for the higher redshift galaxies.Based on observations made with the NASA/ESA Hubble Space Telescope, obtained from the Data Archive at the Space Telescope Science Institute, which is operated by the Association of Universities for Research in Astronomy, Inc., under NASA contract NAS 5-26555. This archival research is associated with program #13241.

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.

Squeeze-SegNet: a new fast deep convolutional neural network for semantic segmentation

NASA Astrophysics Data System (ADS)

Nanfack, Geraldin; Elhassouny, Azeddine; Oulad Haj Thami, Rachid

2018-04-01

The recent researches in Deep Convolutional Neural Network have focused their attention on improving accuracy that provide significant advances. However, if they were limited to classification tasks, nowadays with contributions from Scientific Communities who are embarking in this field, they have become very useful in higher level tasks such as object detection and pixel-wise semantic segmentation. Thus, brilliant ideas in the field of semantic segmentation with deep learning have completed the state of the art of accuracy, however this architectures become very difficult to apply in embedded systems as is the case for autonomous driving. We present a new Deep fully Convolutional Neural Network for pixel-wise semantic segmentation which we call Squeeze-SegNet. The architecture is based on Encoder-Decoder style. We use a SqueezeNet-like encoder and a decoder formed by our proposed squeeze-decoder module and upsample layer using downsample indices like in SegNet and we add a deconvolution layer to provide final multi-channel feature map. On datasets like Camvid or City-states, our net gets SegNet-level accuracy with less than 10 times fewer parameters than SegNet.

Rock images classification by using deep convolution neural network

NASA Astrophysics Data System (ADS)

Cheng, Guojian; Guo, Wenhui

2017-08-01

Granularity analysis is one of the most essential issues in authenticate under microscope. To improve the efficiency and accuracy of traditional manual work, an convolutional neural network based method is proposed for granularity analysis from thin section image, which chooses and extracts features from image samples while build classifier to recognize granularity of input image samples. 4800 samples from Ordos basin are used for experiments under colour spaces of HSV, YCbCr and RGB respectively. On the test dataset, the correct rate in RGB colour space is 98.5%, and it is believable in HSV and YCbCr colour space. The results show that the convolution neural network can classify the rock images with high reliability.

Fully convolutional networks with double-label for esophageal cancer image segmentation by self-transfer learning

NASA Astrophysics Data System (ADS)

Xue, Di-Xiu; Zhang, Rong; Zhao, Yuan-Yuan; Xu, Jian-Ming; Wang, Ya-Lei

2017-07-01

Cancer recognition is the prerequisite to determine appropriate treatment. This paper focuses on the semantic segmentation task of microvascular morphological types on narrowband images to aid clinical examination of esophageal cancer. The most challenge for semantic segmentation is incomplete-labeling. Our key insight is to build fully convolutional networks (FCNs) with double-label to make pixel-wise predictions. The roi-label indicating ROIs (region of interest) is introduced as extra constraint to guild feature learning. Trained end-to-end, the FCN model with two target jointly optimizes both segmentation of sem-label (semantic label) and segmentation of roi-label within the framework of self-transfer learning based on multi-task learning theory. The learning representation ability of shared convolutional networks for sem-label is improved with support of roi-label via achieving a better understanding of information outside the ROIs. Our best FCN model gives satisfactory segmentation result with mean IU up to 77.8% (pixel accuracy > 90%). The results show that the proposed approach is able to assist clinical diagnosis to a certain extent.

The analysis of convolutional codes via the extended Smith algorithm

NASA Technical Reports Server (NTRS)

Mceliece, R. J.; Onyszchuk, I.

1993-01-01

Convolutional codes have been the central part of most error-control systems in deep-space communication for many years. Almost all such applications, however, have used the restricted class of (n,1), also known as 'rate 1/n,' convolutional codes. The more general class of (n,k) convolutional codes contains many potentially useful codes, but their algebraic theory is difficult and has proved to be a stumbling block in the evolution of convolutional coding systems. In this article, the situation is improved by describing a set of practical algorithms for computing certain basic things about a convolutional code (among them the degree, the Forney indices, a minimal generator matrix, and a parity-check matrix), which are usually needed before a system using the code can be built. The approach is based on the classic Forney theory for convolutional codes, together with the extended Smith algorithm for polynomial matrices, which is introduced in this article.

Image statistics decoding for convolutional codes

NASA Technical Reports Server (NTRS)

Pitt, G. H., III; Swanson, L.; Yuen, J. H.

1987-01-01

It is a fact that adjacent pixels in a Voyager image are very similar in grey level. This fact can be used in conjunction with the Maximum-Likelihood Convolutional Decoder (MCD) to decrease the error rate when decoding a picture from Voyager. Implementing this idea would require no changes in the Voyager spacecraft and could be used as a backup to the current system without too much expenditure, so the feasibility of it and the possible gains for Voyager were investigated. Simulations have shown that the gain could be as much as 2 dB at certain error rates, and experiments with real data inspired new ideas on ways to get the most information possible out of the received symbol stream.

Image inpainting and super-resolution using non-local recursive deep convolutional network with skip connections

NASA Astrophysics Data System (ADS)

Liu, Miaofeng

2017-07-01

In recent years, deep convolutional neural networks come into use in image inpainting and super-resolution in many fields. Distinct to most of the former methods requiring to know beforehand the local information for corrupted pixels, we propose a 20-depth fully convolutional network to learn an end-to-end mapping a dataset of damaged/ground truth subimage pairs realizing non-local blind inpainting and super-resolution. As there often exist image with huge corruptions or inpainting on a low-resolution image that the existing approaches unable to perform well, we also share parameters in local area of layers to achieve spatial recursion and enlarge the receptive field. To avoid the difficulty of training this deep neural network, skip-connections between symmetric convolutional layers are designed. Experimental results shows that the proposed method outperforms state-of-the-art methods for diverse corrupting and low-resolution conditions, it works excellently when realizing super-resolution and image inpainting simultaneously

Sub-pixel localisation of passive micro-coil fiducial markers in interventional MRI.

PubMed

Rea, Marc; McRobbie, Donald; Elhawary, Haytham; Tse, Zion T H; Lamperth, Michael; Young, Ian

2009-04-01

Electromechanical devices enable increased accuracy in surgical procedures, and the recent development of MRI-compatible mechatronics permits the use of MRI for real-time image guidance. Integrated imaging of resonant micro-coil fiducials provides an accurate method of tracking devices in a scanner with increased flexibility compared to gradient tracking. Here we report on the ability of ten different image-processing algorithms to track micro-coil fiducials with sub-pixel accuracy. Five algorithms: maximum pixel, barycentric weighting, linear interpolation, quadratic fitting and Gaussian fitting were applied both directly to the pixel intensity matrix and to the cross-correlation matrix obtained by 2D convolution with a reference image. Using images of a 3 mm fiducial marker and a pixel size of 1.1 mm, intensity linear interpolation, which calculates the position of the fiducial centre by interpolating the pixel data to find the fiducial edges, was found to give the best performance for minimal computing power; a maximum error of 0.22 mm was observed in fiducial localisation for displacements up to 40 mm. The inherent standard deviation of fiducial localisation was 0.04 mm. This work enables greater accuracy to be achieved in passive fiducial tracking.

Convolutional neural network regression for short-axis left ventricle segmentation in cardiac cine MR sequences.

PubMed

Tan, Li Kuo; Liew, Yih Miin; Lim, Einly; McLaughlin, Robert A

2017-07-01

Automated left ventricular (LV) segmentation is crucial for efficient quantification of cardiac function and morphology to aid subsequent management of cardiac pathologies. In this paper, we parameterize the complete (all short axis slices and phases) LV segmentation task in terms of the radial distances between the LV centerpoint and the endo- and epicardial contours in polar space. We then utilize convolutional neural network regression to infer these parameters. Utilizing parameter regression, as opposed to conventional pixel classification, allows the network to inherently reflect domain-specific physical constraints. We have benchmarked our approach primarily against the publicly-available left ventricle segmentation challenge (LVSC) dataset, which consists of 100 training and 100 validation cardiac MRI cases representing a heterogeneous mix of cardiac pathologies and imaging parameters across multiple centers. Our approach attained a .77 Jaccard index, which is the highest published overall result in comparison to other automated algorithms. To test general applicability, we also evaluated against the Kaggle Second Annual Data Science Bowl, where the evaluation metric was the indirect clinical measures of LV volume rather than direct myocardial contours. Our approach attained a Continuous Ranked Probability Score (CRPS) of .0124, which would have ranked tenth in the original challenge. With this we demonstrate the effectiveness of convolutional neural network regression paired with domain-specific features in clinical segmentation. Copyright © 2017 Elsevier B.V. All rights reserved.

Pixel detectors for x-ray imaging spectroscopy in space

NASA Astrophysics Data System (ADS)

Treis, J.; Andritschke, R.; Hartmann, R.; Herrmann, S.; Holl, P.; Lauf, T.; Lechner, P.; Lutz, G.; Meidinger, N.; Porro, M.; Richter, R. H.; Schopper, F.; Soltau, H.; Strüder, L.

2009-03-01

Pixelated semiconductor detectors for X-ray imaging spectroscopy are foreseen as key components of the payload of various future space missions exploring the x-ray sky. Located on the platform of the new Spectrum-Roentgen-Gamma satellite, the eROSITA (extended Roentgen Survey with an Imaging Telescope Array) instrument will perform an imaging all-sky survey up to an X-ray energy of 10 keV with unprecedented spectral and angular resolution. The instrument will consist of seven parallel oriented mirror modules each having its own pnCCD camera in the focus. The satellite born X-ray observatory SIMBOL-X will be the first mission to use formation-flying techniques to implement an X-ray telescope with an unprecedented focal length of around 20 m. The detector instrumentation consists of separate high- and low energy detectors, a monolithic 128 × 128 DEPFET macropixel array and a pixellated CdZTe detector respectively, making energy band between 0.5 to 80 keV accessible. A similar concept is proposed for the next generation X-ray observatory IXO. Finally, the MIXS (Mercury Imaging X-ray Spectrometer) instrument on the European Mercury exploration mission BepiColombo will use DEPFET macropixel arrays together with a small X-ray telescope to perform a spatially resolved planetary XRF analysis of Mercury's crust. Here, the mission concepts and their scientific targets are briefly discussed, and the resulting requirements on the detector devices together with the implementation strategies are shown.

Pixelated coatings and advanced IR coatings

NASA Astrophysics Data System (ADS)

Pradal, Fabien; Portier, Benjamin; Oussalah, Meihdi; Leplan, Hervé

2017-09-01

Reosc developed pixelated infrared coatings on detector. Reosc manufactured thick pixelated multilayer stacks on IR-focal plane arrays for bi-spectral imaging systems, demonstrating high filter performance, low crosstalk, and no deterioration of the device sensitivities. More recently, a 5-pixel filter matrix was designed and fabricated. Recent developments in pixelated coatings, shows that high performance infrared filters can be coated directly on detector for multispectral imaging. Next generation space instrument can benefit from this technology to reduce their weight and consumptions.

Fovea detection in optical coherence tomography using convolutional neural networks

NASA Astrophysics Data System (ADS)

Liefers, Bart; Venhuizen, Freerk G.; Theelen, Thomas; Hoyng, Carel; van Ginneken, Bram; Sánchez, Clara I.

2017-02-01

The fovea is an important clinical landmark that is used as a reference for assessing various quantitative measures, such as central retinal thickness or drusen count. In this paper we propose a novel method for automatic detection of the foveal center in Optical Coherence Tomography (OCT) scans. Although the clinician will generally aim to center the OCT scan on the fovea, post-acquisition image processing will give a more accurate estimate of the true location of the foveal center. A Convolutional Neural Network (CNN) was trained on a set of 781 OCT scans that classifies each pixel in the OCT B-scan with a probability of belonging to the fovea. Dilated convolutions were used to obtain a large receptive field, while maintaining pixel-level accuracy. In order to train the network more effectively, negative patches were sampled selectively after each epoch. After CNN classification of the entire OCT volume, the predicted foveal center was chosen as the voxel with maximum output probability, after applying an optimized three-dimensional Gaussian blurring. We evaluate the performance of our method on a data set of 99 OCT scans presenting different stages of Age-related Macular Degeneration (AMD). The fovea was correctly detected in 96:9% of the cases, with a mean distance error of 73 μm(+/-112 μm). This result was comparable to the performance of a second human observer who obtained a mean distance error of 69 μm (+/-94 μm). Experiments showed that the proposed method is accurate and robust even in retinas heavily affected by pathology.

Adaptive pseudo-color enhancement method of weld radiographic images based on HSI color space and self-transformation of pixels.

PubMed

Jiang, Hongquan; Zhao, Yalin; Gao, Jianmin; Gao, Zhiyong

2017-06-01

The radiographic testing (RT) image of a steam turbine manufacturing enterprise has the characteristics of low gray level, low contrast, and blurriness, which lead to a substandard image quality. Moreover, it is not conducive for human eyes to detect and evaluate defects. This study proposes an adaptive pseudo-color enhancement method for weld radiographic images based on the hue, saturation, and intensity (HSI) color space and the self-transformation of pixels to solve these problems. First, the pixel's self-transformation is performed to the pixel value of the original RT image. The function value after the pixel's self-transformation is assigned to the HSI components in the HSI color space. Thereafter, the average intensity of the enhanced image is adaptively adjusted to 0.5 according to the intensity of the original image. Moreover, the hue range and interval can be adjusted according to personal habits. Finally, the HSI components after the adaptive adjustment can be transformed to display in the red, green, and blue color space. Numerous weld radiographic images from a steam turbine manufacturing enterprise are used to validate the proposed method. The experimental results show that the proposed pseudo-color enhancement method can improve image definition and make the target and background areas distinct in weld radiographic images. The enhanced images will be more conducive for defect recognition. Moreover, the image enhanced using the proposed method conforms to the human eye visual properties, and the effectiveness of defect recognition and evaluation can be ensured.

Adaptive pseudo-color enhancement method of weld radiographic images based on HSI color space and self-transformation of pixels

NASA Astrophysics Data System (ADS)

Jiang, Hongquan; Zhao, Yalin; Gao, Jianmin; Gao, Zhiyong

2017-06-01

The radiographic testing (RT) image of a steam turbine manufacturing enterprise has the characteristics of low gray level, low contrast, and blurriness, which lead to a substandard image quality. Moreover, it is not conducive for human eyes to detect and evaluate defects. This study proposes an adaptive pseudo-color enhancement method for weld radiographic images based on the hue, saturation, and intensity (HSI) color space and the self-transformation of pixels to solve these problems. First, the pixel's self-transformation is performed to the pixel value of the original RT image. The function value after the pixel's self-transformation is assigned to the HSI components in the HSI color space. Thereafter, the average intensity of the enhanced image is adaptively adjusted to 0.5 according to the intensity of the original image. Moreover, the hue range and interval can be adjusted according to personal habits. Finally, the HSI components after the adaptive adjustment can be transformed to display in the red, green, and blue color space. Numerous weld radiographic images from a steam turbine manufacturing enterprise are used to validate the proposed method. The experimental results show that the proposed pseudo-color enhancement method can improve image definition and make the target and background areas distinct in weld radiographic images. The enhanced images will be more conducive for defect recognition. Moreover, the image enhanced using the proposed method conforms to the human eye visual properties, and the effectiveness of defect recognition and evaluation can be ensured.

Deep convolutional neural network for prostate MR segmentation

NASA Astrophysics Data System (ADS)

Tian, Zhiqiang; Liu, Lizhi; Fei, Baowei

2017-03-01

Automatic segmentation of the prostate in magnetic resonance imaging (MRI) has many applications in prostate cancer diagnosis and therapy. We propose a deep fully convolutional neural network (CNN) to segment the prostate automatically. Our deep CNN model is trained end-to-end in a single learning stage based on prostate MR images and the corresponding ground truths, and learns to make inference for pixel-wise segmentation. Experiments were performed on our in-house data set, which contains prostate MR images of 20 patients. The proposed CNN model obtained a mean Dice similarity coefficient of 85.3%+/-3.2% as compared to the manual segmentation. Experimental results show that our deep CNN model could yield satisfactory segmentation of the prostate.

Entanglement-assisted quantum convolutional coding

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

Wilde, Mark M.; Brun, Todd A.

2010-04-15

We show how to protect a stream of quantum information from decoherence induced by a noisy quantum communication channel. We exploit preshared entanglement and a convolutional coding structure to develop a theory of entanglement-assisted quantum convolutional coding. Our construction produces a Calderbank-Shor-Steane (CSS) entanglement-assisted quantum convolutional code from two arbitrary classical binary convolutional codes. The rate and error-correcting properties of the classical convolutional codes directly determine the corresponding properties of the resulting entanglement-assisted quantum convolutional code. We explain how to encode our CSS entanglement-assisted quantum convolutional codes starting from a stream of information qubits, ancilla qubits, and shared entangled bits.

DCMDN: Deep Convolutional Mixture Density Network

NASA Astrophysics Data System (ADS)

D'Isanto, Antonio; Polsterer, Kai Lars

2017-09-01

Deep Convolutional Mixture Density Network (DCMDN) estimates probabilistic photometric redshift directly from multi-band imaging data by combining a version of a deep convolutional network with a mixture density network. The estimates are expressed as Gaussian mixture models representing the probability density functions (PDFs) in the redshift space. In addition to the traditional scores, the continuous ranked probability score (CRPS) and the probability integral transform (PIT) are applied as performance criteria. DCMDN is able to predict redshift PDFs independently from the type of source, e.g. galaxies, quasars or stars and renders pre-classification of objects and feature extraction unnecessary; the method is extremely general and allows the solving of any kind of probabilistic regression problems based on imaging data, such as estimating metallicity or star formation rate in galaxies.

Spectral interpolation - Zero fill or convolution. [image processing

NASA Technical Reports Server (NTRS)

Forman, M. L.

1977-01-01

Zero fill, or augmentation by zeros, is a method used in conjunction with fast Fourier transforms to obtain spectral spacing at intervals closer than obtainable from the original input data set. In the present paper, an interpolation technique (interpolation by repetitive convolution) is proposed which yields values accurate enough for plotting purposes and which lie within the limits of calibration accuracies. The technique is shown to operate faster than zero fill, since fewer operations are required. The major advantages of interpolation by repetitive convolution are that efficient use of memory is possible (thus avoiding the difficulties encountered in decimation in time FFTs) and that is is easy to implement.

Fully convolutional network with cluster for semantic segmentation

NASA Astrophysics Data System (ADS)

Ma, Xiao; Chen, Zhongbi; Zhang, Jianlin

2018-04-01

At present, image semantic segmentation technology has been an active research topic for scientists in the field of computer vision and artificial intelligence. Especially, the extensive research of deep neural network in image recognition greatly promotes the development of semantic segmentation. This paper puts forward a method based on fully convolutional network, by cluster algorithm k-means. The cluster algorithm using the image's low-level features and initializing the cluster centers by the super-pixel segmentation is proposed to correct the set of points with low reliability, which are mistakenly classified in great probability, by the set of points with high reliability in each clustering regions. This method refines the segmentation of the target contour and improves the accuracy of the image segmentation.

A robust sub-pixel edge detection method of infrared image based on tremor-based retinal receptive field model

NASA Astrophysics Data System (ADS)

Gao, Kun; Yang, Hu; Chen, Xiaomei; Ni, Guoqiang

2008-03-01

Because of complex thermal objects in an infrared image, the prevalent image edge detection operators are often suitable for a certain scene and extract too wide edges sometimes. From a biological point of view, the image edge detection operators work reliably when assuming a convolution-based receptive field architecture. A DoG (Difference-of- Gaussians) model filter based on ON-center retinal ganglion cell receptive field architecture with artificial eye tremors introduced is proposed for the image contour detection. Aiming at the blurred edges of an infrared image, the subsequent orthogonal polynomial interpolation and sub-pixel level edge detection in rough edge pixel neighborhood is adopted to locate the foregoing rough edges in sub-pixel level. Numerical simulations show that this method can locate the target edge accurately and robustly.

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.

Scene text detection via extremal region based double threshold convolutional network classification

PubMed Central

Zhu, Wei; Lou, Jing; Chen, Longtao; Xia, Qingyuan

2017-01-01

In this paper, we present a robust text detection approach in natural images which is based on region proposal mechanism. A powerful low-level detector named saliency enhanced-MSER extended from the widely-used MSER is proposed by incorporating saliency detection methods, which ensures a high recall rate. Given a natural image, character candidates are extracted from three channels in a perception-based illumination invariant color space by saliency-enhanced MSER algorithm. A discriminative convolutional neural network (CNN) is jointly trained with multi-level information including pixel-level and character-level information as character candidate classifier. Each image patch is classified as strong text, weak text and non-text by double threshold filtering instead of conventional one-step classification, leveraging confident scores obtained via CNN. To further prune non-text regions, we develop a recursive neighborhood search algorithm to track credible texts from weak text set. Finally, characters are grouped into text lines using heuristic features such as spatial location, size, color, and stroke width. We compare our approach with several state-of-the-art methods, and experiments show that our method achieves competitive performance on public datasets ICDAR 2011 and ICDAR 2013. PMID:28820891

Fiber pixelated image database

NASA Astrophysics Data System (ADS)

Shinde, Anant; Perinchery, Sandeep Menon; Matham, Murukeshan Vadakke

2016-08-01

Imaging of physically inaccessible parts of the body such as the colon at micron-level resolution is highly important in diagnostic medical imaging. Though flexible endoscopes based on the imaging fiber bundle are used for such diagnostic procedures, their inherent honeycomb-like structure creates fiber pixelation effects. This impedes the observer from perceiving the information from an image captured and hinders the direct use of image processing and machine intelligence techniques on the recorded signal. Significant efforts have been made by researchers in the recent past in the development and implementation of pixelation removal techniques. However, researchers have often used their own set of images without making source data available which subdued their usage and adaptability universally. A database of pixelated images is the current requirement to meet the growing diagnostic needs in the healthcare arena. An innovative fiber pixelated image database is presented, which consists of pixelated images that are synthetically generated and experimentally acquired. Sample space encompasses test patterns of different scales, sizes, and shapes. It is envisaged that this proposed database will alleviate the current limitations associated with relevant research and development and would be of great help for researchers working on comb structure removal algorithms.

The progress of sub-pixel imaging methods

NASA Astrophysics Data System (ADS)

Wang, Hu; Wen, Desheng

2014-02-01

This paper reviews the Sub-pixel imaging technology principles, characteristics, the current development status at home and abroad and the latest research developments. As Sub-pixel imaging technology has achieved the advantages of high resolution of optical remote sensor, flexible working ways and being miniaturized with no moving parts. The imaging system is suitable for the application of space remote sensor. Its application prospect is very extensive. It is quite possible to be the research development direction of future space optical remote sensing technology.

Accurate Segmentation of Cervical Cytoplasm and Nuclei Based on Multiscale Convolutional Network and Graph Partitioning.

PubMed

Song, Youyi; Zhang, Ling; Chen, Siping; Ni, Dong; Lei, Baiying; Wang, Tianfu

2015-10-01

In this paper, a multiscale convolutional network (MSCN) and graph-partitioning-based method is proposed for accurate segmentation of cervical cytoplasm and nuclei. Specifically, deep learning via the MSCN is explored to extract scale invariant features, and then, segment regions centered at each pixel. The coarse segmentation is refined by an automated graph partitioning method based on the pretrained feature. The texture, shape, and contextual information of the target objects are learned to localize the appearance of distinctive boundary, which is also explored to generate markers to split the touching nuclei. For further refinement of the segmentation, a coarse-to-fine nucleus segmentation framework is developed. The computational complexity of the segmentation is reduced by using superpixel instead of raw pixels. Extensive experimental results demonstrate that the proposed cervical nucleus cell segmentation delivers promising results and outperforms existing methods.

DSN telemetry system performance with convolutionally code data

NASA Technical Reports Server (NTRS)

Mulhall, B. D. L.; Benjauthrit, B.; Greenhall, C. A.; Kuma, D. M.; Lam, J. K.; Wong, J. S.; Urech, J.; Vit, L. D.

1975-01-01

The results obtained to date and the plans for future experiments for the DSN telemetry system were presented. The performance of the DSN telemetry system in decoding convolutionally coded data by both sequential and maximum likelihood techniques is being determined by testing at various deep space stations. The evaluation of performance models is also an objective of this activity.

Carotenoid pixels characterization under color space tests and RGB formulas for mesocarp of mango's fruits cultivars

NASA Astrophysics Data System (ADS)

Hammad, Ahmed Yahya; Kassim, Farid Saad Eid Saad

2010-01-01

This study experimented the pulp (mesocarp) of fourteen cultivars were healthy ripe of Mango fruits (Mangifera indica L.) selected after picking from Mango Spp. namely Taimour [Ta], Dabsha [Da], Aromanis [Ar], Zebda [Ze], Fagri Kelan [Fa], Alphonse [Al], Bulbek heart [Bu], Hindi- Sinnara [Hi], Compania [Co], Langra [La], Mestikawi [Me], Ewais [Ew], Montakhab El Kanater [Mo] and Mabroka [Ma] . Under seven color space tests included (RGB: Red, Green and Blue), (CMY: Cyan, Magenta and Yellow), (CMY: Cyan, Magenta and Yellow), (HSL: Hue, Saturation and Lightness), (CMYK%: Cyan%, Magenta%, Yellow% and Black%), (HSV: Hue, Saturation and Value), (HºSB%: Hueº, Saturation% and Brightness%) and (Lab). (CMY: Cyan, Magenta and Yellow), (HSL: Hue, Saturation and Lightness), (CMYK%: Cyan%, Magenta%, Yellow% and Black%), (HSV: Hue, Saturation and Value), (HºSB%: Hueº, Saturation% and Brightness%) and (Lab). Addition, nine formula of color space tests included (sRGB 0÷1, CMY, CMYK, XYZ, CIE-L*ab, CIE-L*CH, CIE-L*uv, Yxy and Hunter-Lab) and (RGB 0÷FF/hex triplet) and Carotenoid Pixels Scale. Utilizing digital color photographs as tool for obtainment the natural color information for each cultivar then the result expounded with chemical pigment estimations. Our location study in the visual yellow to orange color degrees from the visible color of electromagnetic spectrum in wavelength between (~570 to 620) nm and frequency between (~480 to 530) THz. The results found carotene very strong influence in band Red while chlorophyll (a & b) was very lower subsequently, the values in band Green was depressed. Meanwhile, the general ratios percentage for carotenoid pixels in bands Red, Green and Blue were 50%, 39% and 11% as orderliness opposite the ratios percentage for carotene, chlorophyll a and chlorophyll b which were 63%, 22% and 16% approximately. According to that the pigments influence in all color space tests and RGB formulas. Band Yellow% in color test (CMYK%) as signature

The trellis complexity of convolutional codes

NASA Technical Reports Server (NTRS)

Mceliece, R. J.; Lin, W.

1995-01-01

It has long been known that convolutional codes have a natural, regular trellis structure that facilitates the implementation of Viterbi's algorithm. It has gradually become apparent that linear block codes also have a natural, though not in general a regular, 'minimal' trellis structure, which allows them to be decoded with a Viterbi-like algorithm. In both cases, the complexity of the Viterbi decoding algorithm can be accurately estimated by the number of trellis edges per encoded bit. It would, therefore, appear that we are in a good position to make a fair comparison of the Viterbi decoding complexity of block and convolutional codes. Unfortunately, however, this comparison is somewhat muddled by the fact that some convolutional codes, the punctured convolutional codes, are known to have trellis representations that are significantly less complex than the conventional trellis. In other words, the conventional trellis representation for a convolutional code may not be the minimal trellis representation. Thus, ironically, at present we seem to know more about the minimal trellis representation for block than for convolutional codes. In this article, we provide a remedy, by developing a theory of minimal trellises for convolutional codes. (A similar theory has recently been given by Sidorenko and Zyablov). This allows us to make a direct performance-complexity comparison for block and convolutional codes. A by-product of our work is an algorithm for choosing, from among all generator matrices for a given convolutional code, what we call a trellis-minimal generator matrix, from which the minimal trellis for the code can be directly constructed. Another by-product is that, in the new theory, punctured convolutional codes no longer appear as a special class, but simply as high-rate convolutional codes whose trellis complexity is unexpectedly small.

Convolutional neural network features based change detection in satellite images

NASA Astrophysics Data System (ADS)

Mohammed El Amin, Arabi; Liu, Qingjie; Wang, Yunhong

2016-07-01

With the popular use of high resolution remote sensing (HRRS) satellite images, a huge research efforts have been placed on change detection (CD) problem. An effective feature selection method can significantly boost the final result. While hand-designed features have proven difficulties to design features that effectively capture high and mid-level representations, the recent developments in machine learning (Deep Learning) omit this problem by learning hierarchical representation in an unsupervised manner directly from data without human intervention. In this letter, we propose approaching the change detection problem from a feature learning perspective. A novel deep Convolutional Neural Networks (CNN) features based HR satellite images change detection method is proposed. The main guideline is to produce a change detection map directly from two images using a pretrained CNN. This method can omit the limited performance of hand-crafted features. Firstly, CNN features are extracted through different convolutional layers. Then, a concatenation step is evaluated after an normalization step, resulting in a unique higher dimensional feature map. Finally, a change map was computed using pixel-wise Euclidean distance. Our method has been validated on real bitemporal HRRS satellite images according to qualitative and quantitative analyses. The results obtained confirm the interest of the proposed method.

Convolutional coding techniques for data protection

NASA Technical Reports Server (NTRS)

Massey, J. L.

1975-01-01

Results of research on the use of convolutional codes in data communications are presented. Convolutional coding fundamentals are discussed along with modulation and coding interaction. Concatenated coding systems and data compression with convolutional codes are described.

Pixelation Effects in Weak Lensing

NASA Technical Reports Server (NTRS)

High, F. William; Rhodes, Jason; Massey, Richard; Ellis, Richard

2007-01-01

Weak gravitational lensing can be used to investigate both dark matter and dark energy but requires accurate measurements of the shapes of faint, distant galaxies. Such measurements are hindered by the finite resolution and pixel scale of digital cameras. We investigate the optimum choice of pixel scale for a space-based mission, using the engineering model and survey strategy of the proposed Supernova Acceleration Probe as a baseline. We do this by simulating realistic astronomical images containing a known input shear signal and then attempting to recover the signal using the Rhodes, Refregier, and Groth algorithm. We find that the quality of shear measurement is always improved by smaller pixels. However, in practice, telescopes are usually limited to a finite number of pixels and operational life span, so the total area of a survey increases with pixel size. We therefore fix the survey lifetime and the number of pixels in the focal plane while varying the pixel scale, thereby effectively varying the survey size. In a pure trade-off for image resolution versus survey area, we find that measurements of the matter power spectrum would have minimum statistical error with a pixel scale of 0.09' for a 0.14' FWHM point-spread function (PSF). The pixel scale could be increased to 0.16' if images dithered by exactly half-pixel offsets were always available. Some of our results do depend on our adopted shape measurement method and should be regarded as an upper limit: future pipelines may require smaller pixels to overcome systematic floors not yet accessible, and, in certain circumstances, measuring the shape of the PSF might be more difficult than those of galaxies. However, the relative trends in our analysis are robust, especially those of the surface density of resolved galaxies. Our approach thus provides a snapshot of potential in available technology, and a practical counterpart to analytic studies of pixelation, which necessarily assume an idealized shape

Urban Image Classification: Per-Pixel Classifiers, Sub-Pixel Analysis, Object-Based Image Analysis, and Geospatial Methods. 10; Chapter

NASA Technical Reports Server (NTRS)

Myint, Soe W.; Mesev, Victor; Quattrochi, Dale; Wentz, Elizabeth A.

2013-01-01

Remote sensing methods used to generate base maps to analyze the urban environment rely predominantly on digital sensor data from space-borne platforms. This is due in part from new sources of high spatial resolution data covering the globe, a variety of multispectral and multitemporal sources, sophisticated statistical and geospatial methods, and compatibility with GIS data sources and methods. The goal of this chapter is to review the four groups of classification methods for digital sensor data from space-borne platforms; per-pixel, sub-pixel, object-based (spatial-based), and geospatial methods. Per-pixel methods are widely used methods that classify pixels into distinct categories based solely on the spectral and ancillary information within that pixel. They are used for simple calculations of environmental indices (e.g., NDVI) to sophisticated expert systems to assign urban land covers. Researchers recognize however, that even with the smallest pixel size the spectral information within a pixel is really a combination of multiple urban surfaces. Sub-pixel classification methods therefore aim to statistically quantify the mixture of surfaces to improve overall classification accuracy. While within pixel variations exist, there is also significant evidence that groups of nearby pixels have similar spectral information and therefore belong to the same classification category. Object-oriented methods have emerged that group pixels prior to classification based on spectral similarity and spatial proximity. Classification accuracy using object-based methods show significant success and promise for numerous urban 3 applications. Like the object-oriented methods that recognize the importance of spatial proximity, geospatial methods for urban mapping also utilize neighboring pixels in the classification process. The primary difference though is that geostatistical methods (e.g., spatial autocorrelation methods) are utilized during both the pre- and post

Producing data-based sensitivity kernels from convolution and correlation in exploration geophysics.

NASA Astrophysics Data System (ADS)

Chmiel, M. J.; Roux, P.; Herrmann, P.; Rondeleux, B.

2016-12-01

Many studies have shown that seismic interferometry can be used to estimate surface wave arrivals by correlation of seismic signals recorded at a pair of locations. In the case of ambient noise sources, the convergence towards the surface wave Green's functions is obtained with the criterion of equipartitioned energy. However, seismic acquisition with active, controlled sources gives more possibilities when it comes to interferometry. The use of controlled sources makes it possible to recover the surface wave Green's function between two points using either correlation or convolution. We investigate the convolutional and correlational approaches using land active-seismic data from exploration geophysics. The data were recorded on 10,710 vertical receivers using 51,808 sources (seismic vibrator trucks). The sources spacing is the same in both X and Y directions (30 m) which is known as a "carpet shooting". The receivers are placed in parallel lines with a spacing 150 m in the X direction and 30 m in the Y direction. Invoking spatial reciprocity between sources and receivers, correlation and convolution functions can thus be constructed between either pairs of receivers or pairs of sources. Benefiting from the dense acquisition, we extract sensitivity kernels from correlation and convolution measurements of the seismic data. These sensitivity kernels are subsequently used to produce phase-velocity dispersion curves between two points and to separate the higher mode from the fundamental mode for surface waves. Potential application to surface wave cancellation is also envisaged.

Convolution of Two Series

ERIC Educational Resources Information Center

Umar, A.; Yusau, B.; Ghandi, B. M.

2007-01-01

In this note, we introduce and discuss convolutions of two series. The idea is simple and can be introduced to higher secondary school classes, and has the potential of providing a good background for the well known convolution of function.

WE-G-204-03: Photon-Counting Hexagonal Pixel Array CdTe Detector: Optimal Resampling to Square Pixels

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

Shrestha, S; Vedantham, S; Karellas, A

Purpose: Detectors with hexagonal pixels require resampling to square pixels for distortion-free display of acquired images. In this work, the presampling modulation transfer function (MTF) of a hexagonal pixel array photon-counting CdTe detector for region-of-interest fluoroscopy was measured and the optimal square pixel size for resampling was determined. Methods: A 0.65mm thick CdTe Schottky sensor capable of concurrently acquiring up to 3 energy-windowed images was operated in a single energy-window mode to include ≥10 KeV photons. The detector had hexagonal pixels with apothem of 30 microns resulting in pixel spacing of 60 and 51.96 microns along the two orthogonal directions.more » Images of a tungsten edge test device acquired under IEC RQA5 conditions were double Hough transformed to identify the edge and numerically differentiated. The presampling MTF was determined from the finely sampled line spread function that accounted for the hexagonal sampling. The optimal square pixel size was determined in two ways; the square pixel size for which the aperture function evaluated at the Nyquist frequencies along the two orthogonal directions matched that from the hexagonal pixel aperture functions, and the square pixel size for which the mean absolute difference between the square and hexagonal aperture functions was minimized over all frequencies up to the Nyquist limit. Results: Evaluation of the aperture functions over the entire frequency range resulted in square pixel size of 53 microns with less than 2% difference from the hexagonal pixel. Evaluation of the aperture functions at Nyquist frequencies alone resulted in 54 microns square pixels. For the photon-counting CdTe detector and after resampling to 53 microns square pixels using quadratic interpolation, the presampling MTF at Nyquist frequency of 9.434 cycles/mm along the two directions were 0.501 and 0.507. Conclusion: Hexagonal pixel array photon-counting CdTe detector after resampling to square

High stroke pixel for a deformable mirror

DOEpatents

Miles, Robin R.; Papavasiliou, Alexandros P.

2005-09-20

A mirror pixel that can be fabricated using standard MEMS methods for a deformable mirror. The pixel is electrostatically actuated and is capable of the high deflections needed for spaced-based mirror applications. In one embodiment, the mirror comprises three layers, a top or mirror layer, a middle layer which consists of flexures, and a comb drive layer, with the flexures of the middle layer attached to the mirror layer and to the comb drive layer. The comb drives are attached to a frame via spring flexures. A number of these mirror pixels can be used to construct a large mirror assembly. The actuator for the mirror pixel may be configured as a crenellated beam with one end fixedly secured, or configured as a scissor jack. The mirror pixels may be used in various applications requiring high stroke adaptive optics.

Biased Dropout and Crossmap Dropout: Learning towards effective Dropout regularization in convolutional neural network.

PubMed

Poernomo, Alvin; Kang, Dae-Ki

2018-08-01

Training a deep neural network with a large number of parameters often leads to overfitting problem. Recently, Dropout has been introduced as a simple, yet effective regularization approach to combat overfitting in such models. Although Dropout has shown remarkable results on many deep neural network cases, its actual effect on CNN has not been thoroughly explored. Moreover, training a Dropout model will significantly increase the training time as it takes longer time to converge than a non-Dropout model with the same architecture. To deal with these issues, we address Biased Dropout and Crossmap Dropout, two novel approaches of Dropout extension based on the behavior of hidden units in CNN model. Biased Dropout divides the hidden units in a certain layer into two groups based on their magnitude and applies different Dropout rate to each group appropriately. Hidden units with higher activation value, which give more contributions to the network final performance, will be retained by a lower Dropout rate, while units with lower activation value will be exposed to a higher Dropout rate to compensate the previous part. The second approach is Crossmap Dropout, which is an extension of the regular Dropout in convolution layer. Each feature map in a convolution layer has a strong correlation between each other, particularly in every identical pixel location in each feature map. Crossmap Dropout tries to maintain this important correlation yet at the same time break the correlation between each adjacent pixel with respect to all feature maps by applying the same Dropout mask to all feature maps, so that all pixels or units in equivalent positions in each feature map will be either dropped or active during training. Our experiment with various benchmark datasets shows that our approaches provide better generalization than the regular Dropout. Moreover, our Biased Dropout takes faster time to converge during training phase, suggesting that assigning noise appropriately in

Tweaked residual convolutional network for face alignment

NASA Astrophysics Data System (ADS)

Du, Wenchao; Li, Ke; Zhao, Qijun; Zhang, Yi; Chen, Hu

2017-08-01

We propose a novel Tweaked Residual Convolutional Network approach for face alignment with two-level convolutional networks architecture. Specifically, the first-level Tweaked Convolutional Network (TCN) module predicts the landmark quickly but accurately enough as a preliminary, by taking low-resolution version of the detected face holistically as the input. The following Residual Convolutional Networks (RCN) module progressively refines the landmark by taking as input the local patch extracted around the predicted landmark, particularly, which allows the Convolutional Neural Network (CNN) to extract local shape-indexed features to fine tune landmark position. Extensive evaluations show that the proposed Tweaked Residual Convolutional Network approach outperforms existing methods.

Small Pixel Hybrid CMOS X-ray Detectors

NASA Astrophysics Data System (ADS)

Hull, Samuel; Bray, Evan; Burrows, David N.; Chattopadhyay, Tanmoy; Falcone, Abraham; Kern, Matthew; McQuaide, Maria; Wages, Mitchell

2018-01-01

Concepts for future space-based X-ray observatories call for a large effective area and high angular resolution instrument to enable precision X-ray astronomy at high redshift and low luminosity. Hybrid CMOS detectors are well suited for such high throughput instruments, and the Penn State X-ray detector lab, in collaboration with Teledyne Imaging Sensors, has recently developed new small pixel hybrid CMOS X-ray detectors. These prototype 128x128 pixel devices have 12.5 micron pixel pitch, 200 micron fully depleted depth, and include crosstalk eliminating CTIA amplifiers and in-pixel correlated double sampling (CDS) capability. We report on characteristics of these new detectors, including the best read noise ever measured for an X-ray hybrid CMOS detector, 5.67 e- (RMS).

Comparison of rate one-half, equivalent constraint length 24, binary convolutional codes for use with sequential decoding on the deep-space channel

NASA Technical Reports Server (NTRS)

Massey, J. L.

1976-01-01

Virtually all previously-suggested rate 1/2 binary convolutional codes with KE = 24 are compared. Their distance properties are given; and their performance, both in computation and in error probability, with sequential decoding on the deep-space channel is determined by simulation. Recommendations are made both for the choice of a specific KE = 24 code as well as for codes to be included in future coding standards for the deep-space channel. A new result given in this report is a method for determining the statistical significance of error probability data when the error probability is so small that it is not feasible to perform enough decoding simulations to obtain more than a very small number of decoding errors.

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.

Active pixel sensors: the sensor of choice for future space applications?

NASA Astrophysics Data System (ADS)

Leijtens, Johan; Theuwissen, Albert; Rao, Padmakumar R.; Wang, Xinyang; Xie, Ning

2007-10-01

It is generally known that active pixel sensors (APS) have a number of advantages over CCD detectors if it comes to cost for mass production, power consumption and ease of integration. Nevertheless, most space applications still use CCD detectors because they tend to give better performance and have a successful heritage. To this respect a change may be at hand with the advent of deep sub-micron processed APS imagers (< 0.25-micron feature size). Measurements performed on test structures at the University of Delft have shown that the imagers are very radiation tolerant even if made in a standard process without the use of special design rules. Furthermore it was shown that the 1/f noise associated with deep sub-micron imagers is reduced as compared to previous generations APS imagers due to the improved quality of the gate oxides. Considering that end of life performance will have to be guaranteed, limited budget for adding shielding metal will be available for most applications and lower power operations is always seen as a positive characteristic in space applications, deep sub-micron APS imagers seem to have a number of advantages over CCD's that will probably cause them to replace CCD's in those applications where radiation tolerance and low power operation are important

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.

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.

PSNet: prostate segmentation on MRI based on a convolutional neural network.

PubMed

Tian, Zhiqiang; Liu, Lizhi; Zhang, Zhenfeng; Fei, Baowei

2018-04-01

Automatic segmentation of the prostate on magnetic resonance images (MRI) has many applications in prostate cancer diagnosis and therapy. We proposed a deep fully convolutional neural network (CNN) to segment the prostate automatically. Our deep CNN model is trained end-to-end in a single learning stage, which uses prostate MRI and the corresponding ground truths as inputs. The learned CNN model can be used to make an inference for pixel-wise segmentation. Experiments were performed on three data sets, which contain prostate MRI of 140 patients. The proposed CNN model of prostate segmentation (PSNet) obtained a mean Dice similarity coefficient of [Formula: see text] as compared to the manually labeled ground truth. Experimental results show that the proposed model could yield satisfactory segmentation of the prostate on MRI.

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.

Sensitivity Kernels for the Cross-Convolution Measure: Eliminate the Source in Waveform Tomography

NASA Astrophysics Data System (ADS)

Menke, W. H.

2017-12-01

We use the adjoint method to derive sensitivity kernels for the cross-convolution measure, a goodness-of-fit criterion that is applicable to seismic data containing closely-spaced multiple arrivals, such as reverberating compressional waves and split shear waves. In addition to a general formulation, specific expressions for sensitivity with respect to density, Lamé parameter and shear modulus are derived for a isotropic elastic solid. As is typical of adjoint methods, the kernels depend upon an adjoint field, the source of which, in this case, is the reference displacement field, pre-multiplied by a matrix of cross-correlations of components of the observed field. We use a numerical simulation to evaluate the resolving power of a topographic inversion that employs the cross-convolution measure. The estimated resolving kernel shows is point-like, indicating that the cross-convolution measure will perform well in waveform tomography settings.

Coded aperture detector: an image sensor with sub 20-nm pixel resolution.

PubMed

Miyakawa, Ryan; Mayer, Rafael; Wojdyla, Antoine; Vannier, Nicolas; Lesser, Ian; Aron-Dine, Shifrah; Naulleau, Patrick

2014-08-11

We describe the coded aperture detector, a novel image sensor based on uniformly redundant arrays (URAs) with customizable pixel size, resolution, and operating photon energy regime. In this sensor, a coded aperture is scanned laterally at the image plane of an optical system, and the transmitted intensity is measured by a photodiode. The image intensity is then digitally reconstructed using a simple convolution. We present results from a proof-of-principle optical prototype, demonstrating high-fidelity image sensing comparable to a CCD. A 20-nm half-pitch URA fabricated by the Center for X-ray Optics (CXRO) nano-fabrication laboratory is presented that is suitable for high-resolution image sensing at EUV and soft X-ray wavelengths.

DSN telemetry system performance with convolutionally coded data using operational maximum-likelihood convolutional decoders

NASA Technical Reports Server (NTRS)

Benjauthrit, B.; Mulhall, B.; Madsen, B. D.; Alberda, M. E.

1976-01-01

The DSN telemetry system performance with convolutionally coded data using the operational maximum-likelihood convolutional decoder (MCD) being implemented in the Network is described. Data rates from 80 bps to 115.2 kbps and both S- and X-band receivers are reported. The results of both one- and two-way radio losses are included.

PIXELS: Using field-based learning to investigate students' concepts of pixels and sense of scale

NASA Astrophysics Data System (ADS)

Pope, A.; Tinigin, L.; Petcovic, H. L.; Ormand, C. J.; LaDue, N.

2015-12-01

Empirical work over the past decade supports the notion that a high level of spatial thinking skill is critical to success in the geosciences. Spatial thinking incorporates a host of sub-skills such as mentally rotating an object, imagining the inside of a 3D object based on outside patterns, unfolding a landscape, and disembedding critical patterns from background noise. In this study, we focus on sense of scale, which refers to how an individual quantified space, and is thought to develop through kinesthetic experiences. Remote sensing data are increasingly being used for wide-reaching and high impact research. A sense of scale is critical to many areas of the geosciences, including understanding and interpreting remotely sensed imagery. In this exploratory study, students (N=17) attending the Juneau Icefield Research Program participated in a 3-hour exercise designed to study how a field-based activity might impact their sense of scale and their conceptions of pixels in remotely sensed imagery. Prior to the activity, students had an introductory remote sensing lecture and completed the Sense of Scale inventory. Students walked and/or skied the perimeter of several pixel types, including a 1 m square (representing a WorldView sensor's pixel), a 30 m square (a Landsat pixel) and a 500 m square (a MODIS pixel). The group took reflectance measurements using a field radiometer as they physically traced out the pixel. The exercise was repeated in two different areas, one with homogenous reflectance, and another with heterogeneous reflectance. After the exercise, students again completed the Sense of Scale instrument and a demographic survey. This presentation will share the effects and efficacy of the field-based intervention to teach remote sensing concepts and to investigate potential relationships between students' concepts of pixels and sense of scale.

A multi-pixel InSAR time series analysis method: Simultaneous estimation of atmospheric noise, orbital errors and deformation

NASA Astrophysics Data System (ADS)

Jolivet, R.; Simons, M.

2016-12-01

InSAR time series analysis allows reconstruction of ground deformation with meter-scale spatial resolution and high temporal sampling. For instance, the ESA Sentinel-1 Constellation is capable of providing 6-day temporal sampling, thereby opening a new window on the spatio-temporal behavior of tectonic processes. However, due to computational limitations, most time series methods rely on a pixel-by-pixel approach. This limitation is a concern because (1) accounting for orbital errors requires referencing all interferograms to a common set of pixels before reconstruction of the time series and (2) spatially correlated atmospheric noise due to tropospheric turbulence is ignored. Decomposing interferograms into statistically independent wavelets will mitigate issues of correlated noise, but prior estimation of orbital uncertainties will still be required. Here, we explore a method that considers all pixels simultaneously when solving for the spatio-temporal evolution of interferometric phase Our method is based on a massively parallel implementation of a conjugate direction solver. We consider an interferogram as the sum of the phase difference between 2 SAR acquisitions and the corresponding orbital errors. In addition, we fit the temporal evolution with a physically parameterized function while accounting for spatially correlated noise in the data covariance. We assume noise is isotropic for any given InSAR pair with a covariance described by an exponential function that decays with increasing separation distance between pixels. We regularize our solution in space using a similar exponential function as model covariance. Given the problem size, we avoid matrix multiplications of the full covariances by computing convolutions in the Fourier domain. We first solve the unregularized least squares problem using the LSQR algorithm to approach the final solution, then run our conjugate direction solver to account for data and model covariances. We present synthetic tests

Imaging properties of pixellated scintillators with deep pixels

NASA Astrophysics Data System (ADS)

Barber, H. Bradford; Fastje, David; Lemieux, Daniel; Grim, Gary P.; Furenlid, Lars R.; Miller, Brian W.; Parkhurst, Philip; Nagarkar, Vivek V.

2014-09-01

We have investigated the light-transport properties of scintillator arrays with long, thin pixels (deep pixels) for use in high-energy gamma-ray imaging. We compared 10x10 pixel arrays of YSO:Ce, LYSO:Ce and BGO (1mm x 1mm x 20 mm pixels) made by Proteus, Inc. with similar 10x10 arrays of LSO:Ce and BGO (1mm x 1mm x 15mm pixels) loaned to us by Saint-Gobain. The imaging and spectroscopic behaviors of these scintillator arrays are strongly affected by the choice of a reflector used as an inter-pixel spacer (3M ESR in the case of the Proteus arrays and white, diffuse-reflector for the Saint-Gobain arrays). We have constructed a 3700-pixel LYSO:Ce Prototype NIF Gamma-Ray Imager for use in diagnosing target compression in inertial confinement fusion. This system was tested at the OMEGA Laser and exhibited significant optical, inter-pixel cross-talk that was traced to the use of a single-layer of ESR film as an inter-pixel spacer. We show how the optical cross-talk can be mapped, and discuss correction procedures. We demonstrate a 10x10 YSO:Ce array as part of an iQID (formerly BazookaSPECT) imager and discuss issues related to the internal activity of 176Lu in LSO:Ce and LYSO:Ce detectors.

Imaging properties of pixellated scintillators with deep pixels

PubMed Central

Barber, H. Bradford; Fastje, David; Lemieux, Daniel; Grim, Gary P.; Furenlid, Lars R.; Miller, Brian W.; Parkhurst, Philip; Nagarkar, Vivek V.

2015-01-01

We have investigated the light-transport properties of scintillator arrays with long, thin pixels (deep pixels) for use in high-energy gamma-ray imaging. We compared 10×10 pixel arrays of YSO:Ce, LYSO:Ce and BGO (1mm × 1mm × 20 mm pixels) made by Proteus, Inc. with similar 10×10 arrays of LSO:Ce and BGO (1mm × 1mm × 15mm pixels) loaned to us by Saint-Gobain. The imaging and spectroscopic behaviors of these scintillator arrays are strongly affected by the choice of a reflector used as an inter-pixel spacer (3M ESR in the case of the Proteus arrays and white, diffuse-reflector for the Saint-Gobain arrays). We have constructed a 3700-pixel LYSO:Ce Prototype NIF Gamma-Ray Imager for use in diagnosing target compression in inertial confinement fusion. This system was tested at the OMEGA Laser and exhibited significant optical, inter-pixel cross-talk that was traced to the use of a single-layer of ESR film as an inter-pixel spacer. We show how the optical cross-talk can be mapped, and discuss correction procedures. We demonstrate a 10×10 YSO:Ce array as part of an iQID (formerly BazookaSPECT) imager and discuss issues related to the internal activity of 176Lu in LSO:Ce and LYSO:Ce detectors. PMID:26236070

Achieving unequal error protection with convolutional codes

NASA Technical Reports Server (NTRS)

Mills, D. G.; Costello, D. J., Jr.; Palazzo, R., Jr.

1994-01-01

This paper examines the unequal error protection capabilities of convolutional codes. Both time-invariant and periodically time-varying convolutional encoders are examined. The effective free distance vector is defined and is shown to be useful in determining the unequal error protection (UEP) capabilities of convolutional codes. A modified transfer function is used to determine an upper bound on the bit error probabilities for individual input bit positions in a convolutional encoder. The bound is heavily dependent on the individual effective free distance of the input bit position. A bound relating two individual effective free distances is presented. The bound is a useful tool in determining the maximum possible disparity in individual effective free distances of encoders of specified rate and memory distribution. The unequal error protection capabilities of convolutional encoders of several rates and memory distributions are determined and discussed.

Charge amplitude distribution of the Gossip gaseous pixel detector

NASA Astrophysics Data System (ADS)

Blanco Carballo, V. M.; Chefdeville, M.; Colas, P.; Giomataris, Y.; van der Graaf, H.; Gromov, V.; Hartjes, F.; Kluit, R.; Koffeman, E.; Salm, C.; Schmitz, J.; Smits, S. M.; Timmermans, J.; Visschers, J. L.

2007-12-01

The Gossip gaseous pixel detector is being developed for the detection of charged particles in extreme high radiation environments as foreseen close to the interaction point of the proposed super LHC. The detecting medium is a thin layer of gas. Because of the low density of this medium, only a few primary electron/ion pairs are created by the traversing particle. To get a detectable signal, the electrons drift towards a perforated metal foil (Micromegas) whereafter they are multiplied in a gas avalanche to provide a detectable signal. The gas avalanche occurs in the high field between the Micromegas and the pixel readout chip (ROC). Compared to a silicon pixel detector, Gossip features a low material budget and a low cooling power. An experiment using X-rays has indicated a possible high radiation tolerance exceeding 10 16 hadrons/cm 2. The amplified charge signal has a broad amplitude distribution due to the limited statistics of the primary ionization and the statistical variation of the gas amplification. Therefore, some degree of inefficiency is inevitable. This study presents experimental results on the charge amplitude distribution for CO 2/DME (dimethyl-ether) and Ar/iC 4H 10 mixtures. The measured curves were fitted with the outcome of a theoretical model. In the model, the physical Landau distribution is approximated by a Poisson distribution that is convoluted with the variation of the gas gain and the electronic noise. The value for the fraction of pedestal events is used for a direct calculation of the cluster density. For some gases, the measured cluster density is considerably lower than given in literature.

Optical transmission testing based on asynchronous sampling techniques: images analysis containing chromatic dispersion using convolutional neural network

NASA Astrophysics Data System (ADS)

Mrozek, T.; Perlicki, K.; Tajmajer, T.; Wasilewski, P.

2017-08-01

The article presents an image analysis method, obtained from an asynchronous delay tap sampling (ADTS) technique, which is used for simultaneous monitoring of various impairments occurring in the physical layer of the optical network. The ADTS method enables the visualization of the optical signal in the form of characteristics (so called phase portraits) that change their shape under the influence of impairments such as chromatic dispersion, polarization mode dispersion and ASE noise. Using this method, a simulation model was built with OptSim 4.0. After the simulation study, data were obtained in the form of images that were further analyzed using the convolutional neural network algorithm. The main goal of the study was to train a convolutional neural network to recognize the selected impairment (distortion); then to test its accuracy and estimate the impairment for the selected set of test images. The input data consisted of processed binary images in the form of two-dimensional matrices, with the position of the pixel. This article focuses only on the analysis of images containing chromatic dispersion.

Efficient convolutional sparse coding

DOEpatents

Wohlberg, Brendt

2017-06-20

Computationally efficient algorithms may be applied for fast dictionary learning solving the convolutional sparse coding problem in the Fourier domain. More specifically, efficient convolutional sparse coding may be derived within an alternating direction method of multipliers (ADMM) framework that utilizes fast Fourier transforms (FFT) to solve the main linear system in the frequency domain. Such algorithms may enable a significant reduction in computational cost over conventional approaches by implementing a linear solver for the most critical and computationally expensive component of the conventional iterative algorithm. The theoretical computational cost of the algorithm may be reduced from O(M.sup.3N) to O(MN log N), where N is the dimensionality of the data and M is the number of elements in the dictionary. This significant improvement in efficiency may greatly increase the range of problems that can practically be addressed via convolutional sparse representations.

Multithreaded implicitly dealiased convolutions

NASA Astrophysics Data System (ADS)

Roberts, Malcolm; Bowman, John C.

2018-03-01

Implicit dealiasing is a method for computing in-place linear convolutions via fast Fourier transforms that decouples work memory from input data. It offers easier memory management and, for long one-dimensional input sequences, greater efficiency than conventional zero-padding. Furthermore, for convolutions of multidimensional data, the segregation of data and work buffers can be exploited to reduce memory usage and execution time significantly. This is accomplished by processing and discarding data as it is generated, allowing work memory to be reused, for greater data locality and performance. A multithreaded implementation of implicit dealiasing that accepts an arbitrary number of input and output vectors and a general multiplication operator is presented, along with an improved one-dimensional Hermitian convolution that avoids the loop dependency inherent in previous work. An alternate data format that can accommodate a Nyquist mode and enhance cache efficiency is also proposed.

A novel radiation hard pixel design for space applications

NASA Astrophysics Data System (ADS)

Aurora, A. M.; Marochkin, V. V.; Tuuva, T.

2017-11-01

We have developed a novel radiation hard photon detector concept based on Modified Internal Gate Field Effect Transistor (MIGFET) wherein a buried Modified Internal Gate (MIG) is implanted underneath a channel of a FET. In between the MIG and the channel of the FET there is depleted semiconductor material forming a potential barrier between charges in the channel and similar type signal charges located in the MIG. The signal charges in the MIG have a measurable effect on the conductance of the channel. In this paper a radiation hard double MIGFET pixel is investigated comprising two MIGFETs. By transferring the signal charges between the two MIGs Non-Destructive Correlated Double Sampling Readout (NDCDSR) is enabled. The radiation hardness of the proposed double MIGFET structure stems from the fact that interface related issues can be considerably mitigated. The reason for this is, first of all, that interface generated dark noise can be completely avoided and secondly, that interface generated 1/f noise can be considerably reduced due to a deep buried channel readout configuration. Electrical parameters of the double MIGFET pixel have been evaluated by 3D TCAD simulation study. Simulation results show the absence of interface generated dark noise, significantly reduced interface generated 1/f noise, well performing NDCDSR operation, and blooming protection due to an inherent vertical anti-blooming structure. In addition, the backside illuminated thick fully depleted pixel design results in low crosstalk due to lack of diffusion and good quantum efficiency from visible to Near Infra-Red (NIR) light. These facts result in excellent Signal-to-Noise Ratio (SNR) and very low crosstalk enabling thus excellent image quality. The simulation demonstrates the charge to current conversion gain for source current read-out to be 1.4 nA/e.

Scientific CMOS Pixels

NASA Astrophysics Data System (ADS)

Janesick, James; Gunawan, Ferry; Dosluoglu, Taner; Tower, John; McCaffrey, Niel

2002-08-01

High performance CMOS pixels are introduced; and their development is discussed. 3T (3-transistor) photodiode, 5T pinned diode, 6T photogate and 6T photogate back illuminated CMOS pixels are examined in detail, and the latter three are considered as scientific pixels. The advantages and disadvantagesof these options for scientific CMOS pixels are examined.Pixel characterization, which is used to gain a better understanding of CMOS pixels themselves, is also discussed.

Scientific CMOS Pixels

NASA Astrophysics Data System (ADS)

Janesick, J.; Gunawan, F.; Dosluoglu, T.; Tower, J.; McCaffrey, N.

High performance CMOS pixels are introduced and their development is discussed. 3T (3-transistor) photodiode, 5T pinned diode, 6T photogate and 6T photogate back illuminated CMOS pixels are examined in detail, and the latter three are considered as scientific pixels. The advantages and disadvantages of these options for scientific CMOS pixels are examined. Pixel characterization, which is used to gain a better understanding of CMOS pixels themselves, is also discussed.

An Efficient Implementation of Deep Convolutional Neural Networks for MRI Segmentation.

PubMed

Hoseini, Farnaz; Shahbahrami, Asadollah; Bayat, Peyman

2018-02-27

Image segmentation is one of the most common steps in digital image processing, classifying a digital image into different segments. The main goal of this paper is to segment brain tumors in magnetic resonance images (MRI) using deep learning. Tumors having different shapes, sizes, brightness and textures can appear anywhere in the brain. These complexities are the reasons to choose a high-capacity Deep Convolutional Neural Network (DCNN) containing more than one layer. The proposed DCNN contains two parts: architecture and learning algorithms. The architecture and the learning algorithms are used to design a network model and to optimize parameters for the network training phase, respectively. The architecture contains five convolutional layers, all using 3 × 3 kernels, and one fully connected layer. Due to the advantage of using small kernels with fold, it allows making the effect of larger kernels with smaller number of parameters and fewer computations. Using the Dice Similarity Coefficient metric, we report accuracy results on the BRATS 2016, brain tumor segmentation challenge dataset, for the complete, core, and enhancing regions as 0.90, 0.85, and 0.84 respectively. The learning algorithm includes the task-level parallelism. All the pixels of an MR image are classified using a patch-based approach for segmentation. We attain a good performance and the experimental results show that the proposed DCNN increases the segmentation accuracy compared to previous techniques.

Fully convolutional neural network for removing background in noisy images of uranium bearing particles

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

Tarolli, Jay G.; Naes, Benjamin E.; Butler, Lamar

A fully convolutional neural network (FCN) was developed to supersede automatic or manual thresholding algorithms used for tabulating SIMS particle search data. The FCN was designed to perform a binary classification of pixels in each image belonging to a particle or not, thereby effectively removing background signal without manually or automatically determining an intensity threshold. Using 8,000 images from 28 different particle screening analyses, the FCN was trained to accurately predict pixels belonging to a particle with near 99% accuracy. Background eliminated images were then segmented using a watershed technique in order to determine isotopic ratios of particles. A comparisonmore » of the isotopic distributions of an independent data set segmented using the neural network, compared to a commercially available automated particle measurement (APM) program developed by CAMECA, highlighted the necessity for effective background removal to ensure that resulting particle identification is not only accurate, but preserves valuable signal that could be lost due to improper segmentation. The FCN approach improves the robustness of current state-of-the-art particle searching algorithms by reducing user input biases, resulting in an improved absolute signal per particle and decreased uncertainty of the determined isotope ratios.« less

Fully Convolutional Networks for Ground Classification from LIDAR Point Clouds

NASA Astrophysics Data System (ADS)

Rizaldy, A.; Persello, C.; Gevaert, C. M.; Oude Elberink, S. J.

2018-05-01

Deep Learning has been massively used for image classification in recent years. The use of deep learning for ground classification from LIDAR point clouds has also been recently studied. However, point clouds need to be converted into an image in order to use Convolutional Neural Networks (CNNs). In state-of-the-art techniques, this conversion is slow because each point is converted into a separate image. This approach leads to highly redundant computation during conversion and classification. The goal of this study is to design a more efficient data conversion and ground classification. This goal is achieved by first converting the whole point cloud into a single image. The classification is then performed by a Fully Convolutional Network (FCN), a modified version of CNN designed for pixel-wise image classification. The proposed method is significantly faster than state-of-the-art techniques. On the ISPRS Filter Test dataset, it is 78 times faster for conversion and 16 times faster for classification. Our experimental analysis on the same dataset shows that the proposed method results in 5.22 % of total error, 4.10 % of type I error, and 15.07 % of type II error. Compared to the previous CNN-based technique and LAStools software, the proposed method reduces the total error and type I error (while type II error is slightly higher). The method was also tested on a very high point density LIDAR point clouds resulting in 4.02 % of total error, 2.15 % of type I error and 6.14 % of type II error.

DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs.

PubMed

Chen, Liang-Chieh; Papandreou, George; Kokkinos, Iasonas; Murphy, Kevin; Yuille, Alan L

2018-04-01

In this work we address the task of semantic image segmentation with Deep Learning and make three main contributions that are experimentally shown to have substantial practical merit. First, we highlight convolution with upsampled filters, or 'atrous convolution', as a powerful tool in dense prediction tasks. Atrous convolution allows us to explicitly control the resolution at which feature responses are computed within Deep Convolutional Neural Networks. It also allows us to effectively enlarge the field of view of filters to incorporate larger context without increasing the number of parameters or the amount of computation. Second, we propose atrous spatial pyramid pooling (ASPP) to robustly segment objects at multiple scales. ASPP probes an incoming convolutional feature layer with filters at multiple sampling rates and effective fields-of-views, thus capturing objects as well as image context at multiple scales. Third, we improve the localization of object boundaries by combining methods from DCNNs and probabilistic graphical models. The commonly deployed combination of max-pooling and downsampling in DCNNs achieves invariance but has a toll on localization accuracy. We overcome this by combining the responses at the final DCNN layer with a fully connected Conditional Random Field (CRF), which is shown both qualitatively and quantitatively to improve localization performance. Our proposed "DeepLab" system sets the new state-of-art at the PASCAL VOC-2012 semantic image segmentation task, reaching 79.7 percent mIOU in the test set, and advances the results on three other datasets: PASCAL-Context, PASCAL-Person-Part, and Cityscapes. All of our code is made publicly available online.

Enhanced online convolutional neural networks for object tracking

NASA Astrophysics Data System (ADS)

Zhang, Dengzhuo; Gao, Yun; Zhou, Hao; Li, Tianwen

2018-04-01

In recent several years, object tracking based on convolution neural network has gained more and more attention. The initialization and update of convolution filters can directly affect the precision of object tracking effective. In this paper, a novel object tracking via an enhanced online convolution neural network without offline training is proposed, which initializes the convolution filters by a k-means++ algorithm and updates the filters by an error back-propagation. The comparative experiments of 7 trackers on 15 challenging sequences showed that our tracker can perform better than other trackers in terms of AUC and precision.

HST/WFC3 Characteristics: gain, post-flash stability, UVIS low-sensitivity pixels, persistence, IR flats and bad pixel table

NASA Astrophysics Data System (ADS)

Gunning, Heather C.; Baggett, Sylvia; Gosmeyer, Catherine M.; Long, Knox S.; Ryan, Russell E.; MacKenty, John W.; Durbin, Meredith

2015-08-01

The Wide Field Camera 3 (WFC3) is a fourth-generation imaging instrument on the Hubble Space Telescope (HST). Installed in May 2009, WFC3 is comprised of two observational channels covering wavelengths from UV/Visible (UVIS) to infrared (IR); both have been performing well on-orbit. We discuss the gain stability of both WFC3 channel detectors from ground testing through present day. For UVIS, we detail a low-sensitivity pixel population that evolves during the time between anneals, but is largely reset by the annealing procedure. We characterize the post-flash LED lamp stability, used and recommended to mitigate CTE effects for observations with less than 12e-/pixel backgrounds. We present mitigation options for IR persistence during and after observations. Finally, we give an overview on the construction of the IR flats and provide updates on the bad pixel table.

PIV-DCNN: cascaded deep convolutional neural networks for particle image velocimetry

NASA Astrophysics Data System (ADS)

Lee, Yong; Yang, Hua; Yin, Zhouping

2017-12-01

Velocity estimation (extracting the displacement vector information) from the particle image pairs is of critical importance for particle image velocimetry. This problem is mostly transformed into finding the sub-pixel peak in a correlation map. To address the original displacement extraction problem, we propose a different evaluation scheme (PIV-DCNN) with four-level regression deep convolutional neural networks. At each level, the networks are trained to predict a vector from two input image patches. The low-level network is skilled at large displacement estimation and the high- level networks are devoted to improving the accuracy. Outlier replacement and symmetric window offset operation glue the well- functioning networks in a cascaded manner. Through comparison with the standard PIV methods (one-pass cross-correlation method, three-pass window deformation), the practicability of the proposed PIV-DCNN is verified by the application to a diversity of synthetic and experimental PIV images.

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.

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

Coset Codes Viewed as Terminated Convolutional Codes

NASA Technical Reports Server (NTRS)

Fossorier, Marc P. C.; Lin, Shu

1996-01-01

In this paper, coset codes are considered as terminated convolutional codes. Based on this approach, three new general results are presented. First, it is shown that the iterative squaring construction can equivalently be defined from a convolutional code whose trellis terminates. This convolutional code determines a simple encoder for the coset code considered, and the state and branch labelings of the associated trellis diagram become straightforward. Also, from the generator matrix of the code in its convolutional code form, much information about the trade-off between the state connectivity and complexity at each section, and the parallel structure of the trellis, is directly available. Based on this generator matrix, it is shown that the parallel branches in the trellis diagram of the convolutional code represent the same coset code C(sub 1), of smaller dimension and shorter length. Utilizing this fact, a two-stage optimum trellis decoding method is devised. The first stage decodes C(sub 1), while the second stage decodes the associated convolutional code, using the branch metrics delivered by stage 1. Finally, a bidirectional decoding of each received block starting at both ends is presented. If about the same number of computations is required, this approach remains very attractive from a practical point of view as it roughly doubles the decoding speed. This fact is particularly interesting whenever the second half of the trellis is the mirror image of the first half, since the same decoder can be implemented for both parts.

Molecular graph convolutions: moving beyond fingerprints

NASA Astrophysics Data System (ADS)

Kearnes, Steven; McCloskey, Kevin; Berndl, Marc; Pande, Vijay; Riley, Patrick

2016-08-01

Molecular "fingerprints" encoding structural information are the workhorse of cheminformatics and machine learning in drug discovery applications. However, fingerprint representations necessarily emphasize particular aspects of the molecular structure while ignoring others, rather than allowing the model to make data-driven decisions. We describe molecular graph convolutions, a machine learning architecture for learning from undirected graphs, specifically small molecules. Graph convolutions use a simple encoding of the molecular graph—atoms, bonds, distances, etc.—which allows the model to take greater advantage of information in the graph structure. Although graph convolutions do not outperform all fingerprint-based methods, they (along with other graph-based methods) represent a new paradigm in ligand-based virtual screening with exciting opportunities for future improvement.

Molecular graph convolutions: moving beyond fingerprints.

PubMed

Kearnes, Steven; McCloskey, Kevin; Berndl, Marc; Pande, Vijay; Riley, Patrick

2016-08-01

Molecular "fingerprints" encoding structural information are the workhorse of cheminformatics and machine learning in drug discovery applications. However, fingerprint representations necessarily emphasize particular aspects of the molecular structure while ignoring others, rather than allowing the model to make data-driven decisions. We describe molecular graph convolutions, a machine learning architecture for learning from undirected graphs, specifically small molecules. Graph convolutions use a simple encoding of the molecular graph-atoms, bonds, distances, etc.-which allows the model to take greater advantage of information in the graph structure. Although graph convolutions do not outperform all fingerprint-based methods, they (along with other graph-based methods) represent a new paradigm in ligand-based virtual screening with exciting opportunities for future improvement.

An image-space parallel convolution filtering algorithm based on shadow map

NASA Astrophysics Data System (ADS)

Li, Hua; Yang, Huamin; Zhao, Jianping

2017-07-01

Shadow mapping is commonly used in real-time rendering. In this paper, we presented an accurate and efficient method of soft shadows generation from planar area lights. First this method generated a depth map from light's view, and analyzed the depth-discontinuities areas as well as shadow boundaries. Then these areas were described as binary values in the texture map called binary light-visibility map, and a parallel convolution filtering algorithm based on GPU was enforced to smooth out the boundaries with a box filter. Experiments show that our algorithm is an effective shadow map based method that produces perceptually accurate soft shadows in real time with more details of shadow boundaries compared with the previous works.

The general theory of convolutional codes

NASA Technical Reports Server (NTRS)

Mceliece, R. J.; Stanley, R. P.

1993-01-01

This article presents a self-contained introduction to the algebraic theory of convolutional codes. This introduction is partly a tutorial, but at the same time contains a number of new results which will prove useful for designers of advanced telecommunication systems. Among the new concepts introduced here are the Hilbert series for a convolutional code and the class of compact codes.

Molecular graph convolutions: moving beyond fingerprints

PubMed Central

Kearnes, Steven; McCloskey, Kevin; Berndl, Marc; Pande, Vijay; Riley, Patrick

2016-01-01

Molecular “fingerprints” encoding structural information are the workhorse of cheminformatics and machine learning in drug discovery applications. However, fingerprint representations necessarily emphasize particular aspects of the molecular structure while ignoring others, rather than allowing the model to make data-driven decisions. We describe molecular graph convolutions, a machine learning architecture for learning from undirected graphs, specifically small molecules. Graph convolutions use a simple encoding of the molecular graph—atoms, bonds, distances, etc.—which allows the model to take greater advantage of information in the graph structure. Although graph convolutions do not outperform all fingerprint-based methods, they (along with other graph-based methods) represent a new paradigm in ligand-based virtual screening with exciting opportunities for future improvement. PMID:27558503

Characterization of Pixelated Cadmium-Zinc-Telluride Detectors for Astrophysical Applications

NASA Technical Reports Server (NTRS)

Gaskin, Jessica; Sharma, Dharma; Ramsey, Brian; Seller, Paul

2003-01-01

Comparisons of charge sharing and charge loss measurements between two pixelated Cadmium-Zinc-Telluride (CdZnTe) detectors are discussed. These properties along with the detector geometry help to define the limiting energy resolution and spatial resolution of the detector in question. The first detector consists of a 1-mm-thick piece of CdZnTe sputtered with a 4x4 array of pixels with pixel pitch of 750 microns (inter-pixel gap is 100 microns). Signal readout is via discrete ultra-low-noise preamplifiers, one for each of the 16 pixels. The second detector consists of a 2-mm-thick piece of CdZnTe sputtered with a 16x16 array of pixels with a pixel pitch of 300 microns (inter-pixel gap is 50 microns). This crystal is bonded to a custom-built readout chip (ASIC) providing all front-end electronics to each of the 256 independent pixels. These detectors act as precursors to that which will be used at the focal plane of the High Energy Replicated Optics (HERO) telescope currently being developed at Marshall Space Flight Center. With a telescope focal length of 6 meters, the detector needs to have a spatial resolution of around 200 microns in order to take full advantage of the HERO angular resolution. We discuss to what degree charge sharing will degrade energy resolution but will improve our spatial resolution through position interpolation.

Simplified Syndrome Decoding of (n, 1) Convolutional Codes

NASA Technical Reports Server (NTRS)

Reed, I. S.; Truong, T. K.

1983-01-01

A new syndrome decoding algorithm for the (n, 1) convolutional codes (CC) that is different and simpler than the previous syndrome decoding algorithm of Schalkwijk and Vinck is presented. The new algorithm uses the general solution of the polynomial linear Diophantine equation for the error polynomial vector E(D). This set of Diophantine solutions is a coset of the CC space. A recursive or Viterbi-like algorithm is developed to find the minimum weight error vector cirumflex E(D) in this error coset. An example illustrating the new decoding algorithm is given for the binary nonsymmetric (2,1)CC.

Some partial-unit-memory convolutional codes

NASA Technical Reports Server (NTRS)

Abdel-Ghaffar, K.; Mceliece, R. J.; Solomon, G.

1991-01-01

The results of a study on a class of error correcting codes called partial unit memory (PUM) codes are presented. This class of codes, though not entirely new, has until now remained relatively unexplored. The possibility of using the well developed theory of block codes to construct a large family of promising PUM codes is shown. The performance of several specific PUM codes are compared with that of the Voyager standard (2, 1, 6) convolutional code. It was found that these codes can outperform the Voyager code with little or no increase in decoder complexity. This suggests that there may very well be PUM codes that can be used for deep space telemetry that offer both increased performance and decreased implementational complexity over current coding systems.

Convolutional neural network for road extraction

NASA Astrophysics Data System (ADS)

Li, Junping; Ding, Yazhou; Feng, Fajie; Xiong, Baoyu; Cui, Weihong

2017-11-01

In this paper, the convolution neural network with large block input and small block output was used to extract road. To reflect the complex road characteristics in the study area, a deep convolution neural network VGG19 was conducted for road extraction. Based on the analysis of the characteristics of different sizes of input block, output block and the extraction effect, the votes of deep convolutional neural networks was used as the final road prediction. The study image was from GF-2 panchromatic and multi-spectral fusion in Yinchuan. The precision of road extraction was 91%. The experiments showed that model averaging can improve the accuracy to some extent. At the same time, this paper gave some advice about the choice of input block size and output block size.

Improved convolutional coding

NASA Technical Reports Server (NTRS)

Doland, G. D.

1970-01-01

Convolutional coding, used to upgrade digital data transmission under adverse signal conditions, has been improved by a method which ensures data transitions, permitting bit synchronizer operation at lower signal levels. Method also increases decoding ability by removing ambiguous condition.

Psoriasis skin biopsy image segmentation using Deep Convolutional Neural Network.

PubMed

Pal, Anabik; Garain, Utpal; Chandra, Aditi; Chatterjee, Raghunath; Senapati, Swapan

2018-06-01

Development of machine assisted tools for automatic analysis of psoriasis skin biopsy image plays an important role in clinical assistance. Development of automatic approach for accurate segmentation of psoriasis skin biopsy image is the initial prerequisite for developing such system. However, the complex cellular structure, presence of imaging artifacts, uneven staining variation make the task challenging. This paper presents a pioneering attempt for automatic segmentation of psoriasis skin biopsy images. Several deep neural architectures are tried for segmenting psoriasis skin biopsy images. Deep models are used for classifying the super-pixels generated by Simple Linear Iterative Clustering (SLIC) and the segmentation performance of these architectures is compared with the traditional hand-crafted feature based classifiers built on popularly used classifiers like K-Nearest Neighbor (KNN), Support Vector Machine (SVM) and Random Forest (RF). A U-shaped Fully Convolutional Neural Network (FCN) is also used in an end to end learning fashion where input is the original color image and the output is the segmentation class map for the skin layers. An annotated real psoriasis skin biopsy image data set of ninety (90) images is developed and used for this research. The segmentation performance is evaluated with two metrics namely, Jaccard's Coefficient (JC) and the Ratio of Correct Pixel Classification (RCPC) accuracy. The experimental results show that the CNN based approaches outperform the traditional hand-crafted feature based classification approaches. The present research shows that practical system can be developed for machine assisted analysis of psoriasis disease. Copyright © 2018 Elsevier B.V. All rights reserved.

Image quality of mixed convolution kernel in thoracic computed tomography.

PubMed

Neubauer, Jakob; Spira, Eva Maria; Strube, Juliane; Langer, Mathias; Voss, Christian; Kotter, Elmar

2016-11-01

The mixed convolution kernel alters his properties geographically according to the depicted organ structure, especially for the lung. Therefore, we compared the image quality of the mixed convolution kernel to standard soft and hard kernel reconstructions for different organ structures in thoracic computed tomography (CT) images.Our Ethics Committee approved this prospective study. In total, 31 patients who underwent contrast-enhanced thoracic CT studies were included after informed consent. Axial reconstructions were performed with hard, soft, and mixed convolution kernel. Three independent and blinded observers rated the image quality according to the European Guidelines for Quality Criteria of Thoracic CT for 13 organ structures. The observers rated the depiction of the structures in all reconstructions on a 5-point Likert scale. Statistical analysis was performed with the Friedman Test and post hoc analysis with the Wilcoxon rank-sum test.Compared to the soft convolution kernel, the mixed convolution kernel was rated with a higher image quality for lung parenchyma, segmental bronchi, and the border between the pleura and the thoracic wall (P < 0.03). Compared to the hard convolution kernel, the mixed convolution kernel was rated with a higher image quality for aorta, anterior mediastinal structures, paratracheal soft tissue, hilar lymph nodes, esophagus, pleuromediastinal border, large and medium sized pulmonary vessels and abdomen (P < 0.004) but a lower image quality for trachea, segmental bronchi, lung parenchyma, and skeleton (P < 0.001).The mixed convolution kernel cannot fully substitute the standard CT reconstructions. Hard and soft convolution kernel reconstructions still seem to be mandatory for thoracic CT.

Deep multi-scale convolutional neural network for hyperspectral image classification

NASA Astrophysics Data System (ADS)

Zhang, Feng-zhe; Yang, Xia

2018-04-01

In this paper, we proposed a multi-scale convolutional neural network for hyperspectral image classification task. Firstly, compared with conventional convolution, we utilize multi-scale convolutions, which possess larger respective fields, to extract spectral features of hyperspectral image. We design a deep neural network with a multi-scale convolution layer which contains 3 different convolution kernel sizes. Secondly, to avoid overfitting of deep neural network, dropout is utilized, which randomly sleeps neurons, contributing to improve the classification accuracy a bit. In addition, new skills like ReLU in deep learning is utilized in this paper. We conduct experiments on University of Pavia and Salinas datasets, and obtained better classification accuracy compared with other methods.

Design of convolutional tornado code

NASA Astrophysics Data System (ADS)

Zhou, Hui; Yang, Yao; Gao, Hongmin; Tan, Lu

2017-09-01

As a linear block code, the traditional tornado (tTN) code is inefficient in burst-erasure environment and its multi-level structure may lead to high encoding/decoding complexity. This paper presents a convolutional tornado (cTN) code which is able to improve the burst-erasure protection capability by applying the convolution property to the tTN code, and reduce computational complexity by abrogating the multi-level structure. The simulation results show that cTN code can provide a better packet loss protection performance with lower computation complexity than tTN code.

Diamond Pixel Detectors

NASA Astrophysics Data System (ADS)

Adam, W.; Berdermann, E.; Bergonzo, P.; Bertuccio, G.; Bogani, F.; Borchi, E.; Brambilla, A.; Bruzzi, M.; Colledani, C.; Conway, J.; D'Angelo, P.; Dabrowski, W.; Delpierre, P.; Deneuville, A.; Doroshenko, J.; Dulinski, W.; van Eijk, B.; Fallou, A.; Fizzotti, F.; Foster, J.; Foulon, F.; Friedl, M.; Gan, K. K.; Gheeraert, E.; Gobbi, B.; Grim, G. P.; Hallewell, G.; Han, S.; Hartjes, F.; Hrubec, J.; Husson, D.; Kagan, H.; Kania, D.; Kaplon, J.; Kass, R.; Koeth, T.; Krammer, M.; Lander, R.; Logiudice, A.; Lu, R.; mac Lynne, L.; Manfredotti, C.; Meier, D.; Mishina, M.; Moroni, L.; Oh, A.; Pan, L. S.; Pernicka, M.; Perera, L.; Pirollo, S.; Plano, R.; Procario, M.; Riester, J. L.; Roe, S.; Rott, C.; Rousseau, L.; Rudge, A.; Russ, J.; Sala, S.; Sampietro, M.; Schnetzer, S.; Sciortino, S.; Stelzer, H.; Stone, R.; Suter, B.; Tapper, R. J.; Tesarek, R.; Trischuk, W.; Tromson, D.; Vittone, E.; Wedenig, R.; Weilhammer, P.; White, C.; Zeuner, W.; Zoeller, M.

2001-06-01

Diamond based pixel detectors are a promising radiation-hard technology for use at the LHC. We present first results on a CMS diamond pixel sensor. With a threshold setting of 2000 electrons, an average pixel efficiency of 78% was obtained for normally incident minimum ionizing particles.

32 x 16 CMOS smart pixel array for optical interconnects

NASA Astrophysics Data System (ADS)

Kim, Jongwoo; Guilfoyle, Peter S.; Stone, Richard V.; Hessenbruch, John M.; Choquette, Kent D.; Kiamilev, Fouad E.

2000-05-01

Free space optical interconnects can increase throughput capacities and eliminate much of the energy consumption required for `all electronic' systems. High speed optical interconnects can be achieved by integrating optoelectronic devices with conventional electronics. Smart pixel arrays have been developed which use optical interconnects. An individual smart pixel cell is composed of a vertical cavity surface emitting laser (VCSEL), a photodetector, an optical receiver, a laser driver, and digital logic circuitry. Oxide-confined VCSELs are being developed to operate at 850 nm with a threshold current of approximately 1 mA. Multiple quantum well photodetectors are being fabricated from AlGaAs for use with the 850 nm VCSELs. The VCSELs and photodetectors are being integrated with complementary metal oxide semiconductor (CMOS) circuitry using flip-chip bonding. CMOS circuitry is being integrated with a 32 X 16 smart pixel array. The 512 smart pixels are serially linked. Thus, an entire data stream may be clocked through the chip and output electrically by the last pixel. Electrical testing is being performed on the CMOS smart pixel array. Using an on-chip pseudo random number generator, a digital data sequence was cycled through the chip verifying operation of the digital circuitry. Although, the prototype chip was fabricated in 1.2 micrometers technology, simulations have demonstrated that the array can operate at 1 Gb/s per pixel using 0.5 micrometers technology.

Experimental Investigation of Convoluted Contouring for Aircraft Afterbody Drag Reduction

NASA Technical Reports Server (NTRS)

Deere, Karen A.; Hunter, Craig A.

1999-01-01

An experimental investigation was performed in the NASA Langley 16-Foot Transonic Tunnel to determine the aerodynamic effects of external convolutions, placed on the boattail of a nonaxisymmetric nozzle for drag reduction. Boattail angles of 15 and 22 were tested with convolutions placed at a forward location upstream of the boattail curvature, at a mid location along the curvature and at a full location that spanned the entire boattail flap. Each of the baseline nozzle afterbodies (no convolutions) had a parabolic, converging contour with a parabolically decreasing corner radius. Data were obtained at several Mach numbers from static conditions to 1.2 for a range of nozzle pressure ratios and angles of attack. An oil paint flow visualization technique was used to qualitatively assess the effect of the convolutions. Results indicate that afterbody drag reduction by convoluted contouring is convolution location, Mach number, boattail angle, and NPR dependent. The forward convolution location was the most effective contouring geometry for drag reduction on the 22 afterbody, but was only effective for M < 0.95. At M = 0.8, drag was reduced 20 and 36 percent at NPRs of 5.4 and 7, respectively, but drag was increased 10 percent for M = 0.95 at NPR = 7. Convoluted contouring along the 15 boattail angle afterbody was not effective at reducing drag because the flow was minimally separated from the baseline afterbody, unlike the massive separation along the 22 boattail angle baseline afterbody.

Object class segmentation of RGB-D video using recurrent convolutional neural networks.

PubMed

Pavel, Mircea Serban; Schulz, Hannes; Behnke, Sven

2017-04-01

Object class segmentation is a computer vision task which requires labeling each pixel of an image with the class of the object it belongs to. Deep convolutional neural networks (DNN) are able to learn and take advantage of local spatial correlations required for this task. They are, however, restricted by their small, fixed-sized filters, which limits their ability to learn long-range dependencies. Recurrent Neural Networks (RNN), on the other hand, do not suffer from this restriction. Their iterative interpretation allows them to model long-range dependencies by propagating activity. This property is especially useful when labeling video sequences, where both spatial and temporal long-range dependencies occur. In this work, a novel RNN architecture for object class segmentation is presented. We investigate several ways to train such a network. We evaluate our models on the challenging NYU Depth v2 dataset for object class segmentation and obtain competitive results. Copyright © 2017 Elsevier Ltd. All rights reserved.

A maximum likelihood convolutional decoder model vs experimental data comparison

NASA Technical Reports Server (NTRS)

Chen, R. Y.

1979-01-01

This article describes the comparison of a maximum likelihood convolutional decoder (MCD) prediction model and the actual performance of the MCD at the Madrid Deep Space Station. The MCD prediction model is used to develop a subroutine that has been utilized by the Telemetry Analysis Program (TAP) to compute the MCD bit error rate for a given signal-to-noise ratio. The results indicate that that the TAP can predict quite well compared to the experimental measurements. An optimal modulation index also can be found through TAP.

Adaptive single-pixel imaging with aggregated sampling and continuous differential measurements

NASA Astrophysics Data System (ADS)

Huo, Yaoran; He, Hongjie; Chen, Fan; Tai, Heng-Ming

2018-06-01

This paper proposes an adaptive compressive imaging technique with one single-pixel detector and single arm. The aggregated sampling (AS) method enables the reduction of resolutions of the reconstructed images. It aims to reduce the time and space consumption. The target image with a resolution up to 1024 × 1024 can be reconstructed successfully at the 20% sampling rate. The continuous differential measurement (CDM) method combined with a ratio factor of significant coefficient (RFSC) improves the imaging quality. Moreover, RFSC reduces the human intervention in parameter setting. This technique enhances the practicability of single-pixel imaging with the benefits from less time and space consumption, better imaging quality and less human intervention.

A fast complex integer convolution using a hybrid transform

NASA Technical Reports Server (NTRS)

Reed, I. S.; K Truong, T.

1978-01-01

It is shown that the Winograd transform can be combined with a complex integer transform over the Galois field GF(q-squared) to yield a new algorithm for computing the discrete cyclic convolution of complex number points. By this means a fast method for accurately computing the cyclic convolution of a sequence of complex numbers for long convolution lengths can be obtained. This new hybrid algorithm requires fewer multiplications than previous algorithms.

Village Building Identification Based on Ensemble Convolutional Neural Networks

PubMed Central

Guo, Zhiling; Chen, Qi; Xu, Yongwei; Shibasaki, Ryosuke; Shao, Xiaowei

2017-01-01

In this study, we present the Ensemble Convolutional Neural Network (ECNN), an elaborate CNN frame formulated based on ensembling state-of-the-art CNN models, to identify village buildings from open high-resolution remote sensing (HRRS) images. First, to optimize and mine the capability of CNN for village mapping and to ensure compatibility with our classification targets, a few state-of-the-art models were carefully optimized and enhanced based on a series of rigorous analyses and evaluations. Second, rather than directly implementing building identification by using these models, we exploited most of their advantages by ensembling their feature extractor parts into a stronger model called ECNN based on the multiscale feature learning method. Finally, the generated ECNN was applied to a pixel-level classification frame to implement object identification. The proposed method can serve as a viable tool for village building identification with high accuracy and efficiency. The experimental results obtained from the test area in Savannakhet province, Laos, prove that the proposed ECNN model significantly outperforms existing methods, improving overall accuracy from 96.64% to 99.26%, and kappa from 0.57 to 0.86. PMID:29084154

High-throughput adaptive sampling for whole-slide histopathology image analysis (HASHI) via convolutional neural networks: Application to invasive breast cancer detection.

PubMed

Cruz-Roa, Angel; Gilmore, Hannah; Basavanhally, Ajay; Feldman, Michael; Ganesan, Shridar; Shih, Natalie; Tomaszewski, John; Madabhushi, Anant; González, Fabio

2018-01-01

Precise detection of invasive cancer on whole-slide images (WSI) is a critical first step in digital pathology tasks of diagnosis and grading. Convolutional neural network (CNN) is the most popular representation learning method for computer vision tasks, which have been successfully applied in digital pathology, including tumor and mitosis detection. However, CNNs are typically only tenable with relatively small image sizes (200 × 200 pixels). Only recently, Fully convolutional networks (FCN) are able to deal with larger image sizes (500 × 500 pixels) for semantic segmentation. Hence, the direct application of CNNs to WSI is not computationally feasible because for a WSI, a CNN would require billions or trillions of parameters. To alleviate this issue, this paper presents a novel method, High-throughput Adaptive Sampling for whole-slide Histopathology Image analysis (HASHI), which involves: i) a new efficient adaptive sampling method based on probability gradient and quasi-Monte Carlo sampling, and, ii) a powerful representation learning classifier based on CNNs. We applied HASHI to automated detection of invasive breast cancer on WSI. HASHI was trained and validated using three different data cohorts involving near 500 cases and then independently tested on 195 studies from The Cancer Genome Atlas. The results show that (1) the adaptive sampling method is an effective strategy to deal with WSI without compromising prediction accuracy by obtaining comparative results of a dense sampling (∼6 million of samples in 24 hours) with far fewer samples (∼2,000 samples in 1 minute), and (2) on an independent test dataset, HASHI is effective and robust to data from multiple sites, scanners, and platforms, achieving an average Dice coefficient of 76%.

High-throughput adaptive sampling for whole-slide histopathology image analysis (HASHI) via convolutional neural networks: Application to invasive breast cancer detection

PubMed Central

Gilmore, Hannah; Basavanhally, Ajay; Feldman, Michael; Ganesan, Shridar; Shih, Natalie; Tomaszewski, John; Madabhushi, Anant; González, Fabio

2018-01-01

Precise detection of invasive cancer on whole-slide images (WSI) is a critical first step in digital pathology tasks of diagnosis and grading. Convolutional neural network (CNN) is the most popular representation learning method for computer vision tasks, which have been successfully applied in digital pathology, including tumor and mitosis detection. However, CNNs are typically only tenable with relatively small image sizes (200 × 200 pixels). Only recently, Fully convolutional networks (FCN) are able to deal with larger image sizes (500 × 500 pixels) for semantic segmentation. Hence, the direct application of CNNs to WSI is not computationally feasible because for a WSI, a CNN would require billions or trillions of parameters. To alleviate this issue, this paper presents a novel method, High-throughput Adaptive Sampling for whole-slide Histopathology Image analysis (HASHI), which involves: i) a new efficient adaptive sampling method based on probability gradient and quasi-Monte Carlo sampling, and, ii) a powerful representation learning classifier based on CNNs. We applied HASHI to automated detection of invasive breast cancer on WSI. HASHI was trained and validated using three different data cohorts involving near 500 cases and then independently tested on 195 studies from The Cancer Genome Atlas. The results show that (1) the adaptive sampling method is an effective strategy to deal with WSI without compromising prediction accuracy by obtaining comparative results of a dense sampling (∼6 million of samples in 24 hours) with far fewer samples (∼2,000 samples in 1 minute), and (2) on an independent test dataset, HASHI is effective and robust to data from multiple sites, scanners, and platforms, achieving an average Dice coefficient of 76%. PMID:29795581

Evaluation of color encodings for high dynamic range pixels

NASA Astrophysics Data System (ADS)

Boitard, Ronan; Mantiuk, Rafal K.; Pouli, Tania

2015-03-01

Traditional Low Dynamic Range (LDR) color spaces encode a small fraction of the visible color gamut, which does not encompass the range of colors produced on upcoming High Dynamic Range (HDR) displays. Future imaging systems will require encoding much wider color gamut and luminance range. Such wide color gamut can be represented using floating point HDR pixel values but those are inefficient to encode. They also lack perceptual uniformity of the luminance and color distribution, which is provided (in approximation) by most LDR color spaces. Therefore, there is a need to devise an efficient, perceptually uniform and integer valued representation for high dynamic range pixel values. In this paper we evaluate several methods for encoding colour HDR pixel values, in particular for use in image and video compression. Unlike other studies we test both luminance and color difference encoding in a rigorous 4AFC threshold experiments to determine the minimum bit-depth required. Results show that the Perceptual Quantizer (PQ) encoding provides the best perceptual uniformity in the considered luminance range, however the gain in bit-depth is rather modest. More significant difference can be observed between color difference encoding schemes, from which YDuDv encoding seems to be the most efficient.

Framework for Detection and Localization of Extreme Climate Event with Pixel Recursive Super Resolution

NASA Astrophysics Data System (ADS)

Kim, S. K.; Lee, J.; Zhang, C.; Ames, S.; Williams, D. N.

2017-12-01

Deep learning techniques have been successfully applied to solve many problems in climate and geoscience using massive-scaled observed and modeled data. For extreme climate event detections, several models based on deep neural networks have been recently proposed and attend superior performance that overshadows all previous handcrafted expert based method. The issue arising, though, is that accurate localization of events requires high quality of climate data. In this work, we propose framework capable of detecting and localizing extreme climate events in very coarse climate data. Our framework is based on two models using deep neural networks, (1) Convolutional Neural Networks (CNNs) to detect and localize extreme climate events, and (2) Pixel recursive recursive super resolution model to reconstruct high resolution climate data from low resolution climate data. Based on our preliminary work, we have presented two CNNs in our framework for different purposes, detection and localization. Our results using CNNs for extreme climate events detection shows that simple neural nets can capture the pattern of extreme climate events with high accuracy from very coarse reanalysis data. However, localization accuracy is relatively low due to the coarse resolution. To resolve this issue, the pixel recursive super resolution model reconstructs the resolution of input of localization CNNs. We present a best networks using pixel recursive super resolution model that synthesizes details of tropical cyclone in ground truth data while enhancing their resolution. Therefore, this approach not only dramat- ically reduces the human effort, but also suggests possibility to reduce computing cost required for downscaling process to increase resolution of data.

A Deep Convolutional Coupling Network for Change Detection Based on Heterogeneous Optical and Radar Images.

PubMed

Liu, Jia; Gong, Maoguo; Qin, Kai; Zhang, Puzhao

2018-03-01

We propose an unsupervised deep convolutional coupling network for change detection based on two heterogeneous images acquired by optical sensors and radars on different dates. Most existing change detection methods are based on homogeneous images. Due to the complementary properties of optical and radar sensors, there is an increasing interest in change detection based on heterogeneous images. The proposed network is symmetric with each side consisting of one convolutional layer and several coupling layers. The two input images connected with the two sides of the network, respectively, are transformed into a feature space where their feature representations become more consistent. In this feature space, the different map is calculated, which then leads to the ultimate detection map by applying a thresholding algorithm. The network parameters are learned by optimizing a coupling function. The learning process is unsupervised, which is different from most existing change detection methods based on heterogeneous images. Experimental results on both homogenous and heterogeneous images demonstrate the promising performance of the proposed network compared with several existing approaches.

Towards dropout training for convolutional neural networks.

PubMed

Wu, Haibing; Gu, Xiaodong

2015-11-01

Recently, dropout has seen increasing use in deep learning. For deep convolutional neural networks, dropout is known to work well in fully-connected layers. However, its effect in convolutional and pooling layers is still not clear. This paper demonstrates that max-pooling dropout is equivalent to randomly picking activation based on a multinomial distribution at training time. In light of this insight, we advocate employing our proposed probabilistic weighted pooling, instead of commonly used max-pooling, to act as model averaging at test time. Empirical evidence validates the superiority of probabilistic weighted pooling. We also empirically show that the effect of convolutional dropout is not trivial, despite the dramatically reduced possibility of over-fitting due to the convolutional architecture. Elaborately designing dropout training simultaneously in max-pooling and fully-connected layers, we achieve state-of-the-art performance on MNIST, and very competitive results on CIFAR-10 and CIFAR-100, relative to other approaches without data augmentation. Finally, we compare max-pooling dropout and stochastic pooling, both of which introduce stochasticity based on multinomial distributions at pooling stage. Copyright © 2015 Elsevier Ltd. All rights reserved.

Variable pixel size ionospheric tomography

NASA Astrophysics Data System (ADS)

Zheng, Dunyong; Zheng, Hongwei; Wang, Yanjun; Nie, Wenfeng; Li, Chaokui; Ao, Minsi; Hu, Wusheng; Zhou, Wei

2017-06-01

A novel ionospheric tomography technique based on variable pixel size was developed for the tomographic reconstruction of the ionospheric electron density (IED) distribution. In variable pixel size computerized ionospheric tomography (VPSCIT) model, the IED distribution is parameterized by a decomposition of the lower and upper ionosphere with different pixel sizes. Thus, the lower and upper IED distribution may be very differently determined by the available data. The variable pixel size ionospheric tomography and constant pixel size tomography are similar in most other aspects. There are some differences between two kinds of models with constant and variable pixel size respectively, one is that the segments of GPS signal pay should be assigned to the different kinds of pixel in inversion; the other is smoothness constraint factor need to make the appropriate modified where the pixel change in size. For a real dataset, the variable pixel size method distinguishes different electron density distribution zones better than the constant pixel size method. Furthermore, it can be non-chided that when the effort is spent to identify the regions in a model with best data coverage. The variable pixel size method can not only greatly improve the efficiency of inversion, but also produce IED images with high fidelity which are the same as a used uniform pixel size method. In addition, variable pixel size tomography can reduce the underdetermined problem in an ill-posed inverse problem when the data coverage is irregular or less by adjusting quantitative proportion of pixels with different sizes. In comparison with constant pixel size tomography models, the variable pixel size ionospheric tomography technique achieved relatively good results in a numerical simulation. A careful validation of the reliability and superiority of variable pixel size ionospheric tomography was performed. Finally, according to the results of the statistical analysis and quantitative comparison, the

Convolutional encoding of self-dual codes

NASA Technical Reports Server (NTRS)

Solomon, G.

1994-01-01

There exist almost complete convolutional encodings of self-dual codes, i.e., block codes of rate 1/2 with weights w, w = 0 mod 4. The codes are of length 8m with the convolutional portion of length 8m-2 and the nonsystematic information of length 4m-1. The last two bits are parity checks on the two (4m-1) length parity sequences. The final information bit complements one of the extended parity sequences of length 4m. Solomon and van Tilborg have developed algorithms to generate these for the Quadratic Residue (QR) Codes of lengths 48 and beyond. For these codes and reasonable constraint lengths, there are sequential decodings for both hard and soft decisions. There are also possible Viterbi-type decodings that may be simple, as in a convolutional encoding/decoding of the extended Golay Code. In addition, the previously found constraint length K = 9 for the QR (48, 24;12) Code is lowered here to K = 8.

Study of sub-pixel position resolution with time-correlated transient signals in 3D pixelated CdZnTe detectors with varying pixel sizes

NASA Astrophysics Data System (ADS)

Ocampo Giraldo, L.; Bolotnikov, A. E.; Camarda, G. S.; De Geronimo, G.; Fried, J.; Gul, R.; Hodges, D.; Hossain, A.; Ünlü, K.; Vernon, E.; Yang, G.; James, R. B.

2018-03-01

We evaluated the sub-pixel position resolution achievable in large-volume CdZnTe pixelated detectors with conventional pixel patterns and for several different pixel sizes: 2.8 mm, 1.72 mm, 1.4 mm and 0.8 mm. Achieving position resolution below the physical dimensions of pixels (sub-pixel resolution) is a practical path for making high-granularity position-sensitive detectors, <100 μm, using a limited number of pixels dictated by the mechanical constraints and multi-channel readout electronics. High position sensitivity is important for improving the imaging capability of CZT gamma cameras. It also allows for making more accurate corrections of response non-uniformities caused by crystal defects, thus enabling use of standard-grade (unselected) and less expensive CZT crystals for producing large-volume position-sensitive CZT detectors feasible for many practical applications. We analyzed the digitized charge signals from a representative 9 pixels and the cathode, generated using a pulsed-laser light beam focused down to 10 μm (650 nm) to scan over a selected 3 × 3 pixel area. We applied our digital pulse processing technique to the time-correlated signals captured from adjacent pixels to achieve and evaluate the capability for sub-pixel position resolution. As an example, we also demonstrated an application of 3D corrections to improve the energy resolution and positional information of the events for the tested detectors.

Multipurpose active pixel sensor (APS)-based microtracker

NASA Astrophysics Data System (ADS)

Eisenman, Allan R.; Liebe, Carl C.; Zhu, David Q.

1998-12-01

A new, photon-sensitive, imaging array, the active pixel sensor (APS) has emerged as a competitor to the CCD imager for use in star and target trackers. The Jet Propulsion Laboratory (JPL) has undertaken a program to develop a new generation, highly integrated, APS-based, multipurpose tracker: the Programmable Intelligent Microtracker (PIM). The supporting hardware used in the PIM has been carefully selected to enhance the inherent advantages of the APS. Adequate computation power is included to perform star identification, star tracking, attitude determination, space docking, feature tracking, descent imaging for landing control, and target tracking capabilities. Its first version uses a JPL developed 256 X 256-pixel APS and an advanced 32-bit RISC microcontroller. By taking advantage of the unique features of the APS/microcontroller combination, the microtracker will achieve about an order-of-magnitude reduction in mass and power consumption compared to present state-of-the-art star trackers. It will also add the advantage of programmability to enable it to perform a variety of star, other celestial body, and target tracking tasks. The PIM is already proving the usefulness of its design concept for space applications. It is demonstrating the effectiveness of taking such an integrated approach in building a new generation of high performance, general purpose, tracking instruments to be applied to a large variety of future space missions.

ELHnet: a convolutional neural network for classifying cochlear endolymphatic hydrops imaged with optical coherence tomography.

PubMed

Liu, George S; Zhu, Michael H; Kim, Jinkyung; Raphael, Patrick; Applegate, Brian E; Oghalai, John S

2017-10-01

Detection of endolymphatic hydrops is important for diagnosing Meniere's disease, and can be performed non-invasively using optical coherence tomography (OCT) in animal models as well as potentially in the clinic. Here, we developed ELHnet, a convolutional neural network to classify endolymphatic hydrops in a mouse model using learned features from OCT images of mice cochleae. We trained ELHnet on 2159 training and validation images from 17 mice, using only the image pixels and observer-determined labels of endolymphatic hydrops as the inputs. We tested ELHnet on 37 images from 37 mice that were previously not used, and found that the neural network correctly classified 34 of the 37 mice. This demonstrates an improvement in performance from previous work on computer-aided classification of endolymphatic hydrops. To the best of our knowledge, this is the first deep CNN designed for endolymphatic hydrops classification.

PIXEL PUSHER

NASA Technical Reports Server (NTRS)

Stanfill, D. F.

1994-01-01

Pixel Pusher is a Macintosh application used for viewing and performing minor enhancements on imagery. It will read image files in JPL's two primary image formats- VICAR and PDS - as well as the Macintosh PICT format. VICAR (NPO-18076) handles an array of image processing capabilities which may be used for a variety of applications including biomedical image processing, cartography, earth resources, and geological exploration. Pixel Pusher can also import VICAR format color lookup tables for viewing images in pseudocolor (256 colors). This program currently supports only eight bit images but will work on monitors with any number of colors. Arbitrarily large image files may be viewed in a normal Macintosh window. Color and contrast enhancement can be performed with a graphical "stretch" editor (as in contrast stretch). In addition, VICAR images may be saved as Macintosh PICT files for exporting into other Macintosh programs, and individual pixels can be queried to determine their locations and actual data values. Pixel Pusher is written in Symantec's Think C and was developed for use on a Macintosh SE30, LC, or II series computer running System Software 6.0.3 or later and 32 bit QuickDraw. Pixel Pusher will only run on a Macintosh which supports color (whether a color monitor is being used or not). The standard distribution medium for this program is a set of three 3.5 inch Macintosh format diskettes. The program price includes documentation. Pixel Pusher was developed in 1991 and is a copyrighted work with all copyright vested in NASA. Think C is a trademark of Symantec Corporation. Macintosh is a registered trademark of Apple Computer, Inc.

Hot pixel generation in active pixel sensors: dosimetric and micro-dosimetric response

NASA Technical Reports Server (NTRS)

Scheick, Leif; Novak, Frank

2003-01-01

The dosimetric response of an active pixel sensor is analyzed. heavy ions are seen to damage the pixel in much the same way as gamma radiation. The probability of a hot pixel is seen to exhibit behavior that is not typical with other microdose effects.

Selecting Pixels for Kepler Downlink

NASA Technical Reports Server (NTRS)

Bryson, Stephen T.; Jenkins, Jon M.; Klaus, Todd C.; Cote, Miles T.; Quintana, Elisa V.; Hall, Jennifer R.; Ibrahim, Khadeejah; Chandrasekaran, Hema; Caldwell, Douglas A.; Van Cleve, Jeffrey E.;

Study of sub-pixel position resolution with time-correlated transient signals in 3D pixelated CdZnTe detectors with varying pixel sizes

DOE PAGES

Giraldo, L. Ocampo; Bolotnikov, A. E.; Camarda, G. S.; ...

2017-12-18

Here, we evaluated the sub-pixel position resolution achievable in large-volume CdZnTe pixelated detectors with conventional pixel patterns and for several different pixel sizes: 2.8 mm, 1.72 mm, 1.4 mm and 0.8 mm. Achieving position resolution below the physical dimensions of pixels (sub-pixel resolution) is a practical path for making high-granularity position-sensitive detectors, <100 μμm, using a limited number of pixels dictated by the mechanical constraints and multi-channel readout electronics. High position sensitivity is important for improving the imaging capability of CZT gamma cameras. It also allows for making more accurate corrections of response non-uniformities caused by crystal defects, thus enablingmore » use of standard-grade (unselected) and less expensive CZT crystals for producing large-volume position-sensitive CZT detectors feasible for many practical applications. We analyzed the digitized charge signals from a representative 9 pixels and the cathode, generated using a pulsed-laser light beam focused down to 10 m (650 nm) to scan over a selected 3×3 pixel area. We applied our digital pulse processing technique to the time-correlated signals captured from adjacent pixels to achieve and evaluate the capability for sub-pixel position resolution. As an example, we also demonstrated an application of 3D corrections to improve the energy resolution and positional information of the events for the tested detectors.« less

Study of sub-pixel position resolution with time-correlated transient signals in 3D pixelated CdZnTe detectors with varying pixel sizes

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

Giraldo, L. Ocampo; Bolotnikov, A. E.; Camarda, G. S.

Here, we evaluated the sub-pixel position resolution achievable in large-volume CdZnTe pixelated detectors with conventional pixel patterns and for several different pixel sizes: 2.8 mm, 1.72 mm, 1.4 mm and 0.8 mm. Achieving position resolution below the physical dimensions of pixels (sub-pixel resolution) is a practical path for making high-granularity position-sensitive detectors, <100 μμm, using a limited number of pixels dictated by the mechanical constraints and multi-channel readout electronics. High position sensitivity is important for improving the imaging capability of CZT gamma cameras. It also allows for making more accurate corrections of response non-uniformities caused by crystal defects, thus enablingmore » use of standard-grade (unselected) and less expensive CZT crystals for producing large-volume position-sensitive CZT detectors feasible for many practical applications. We analyzed the digitized charge signals from a representative 9 pixels and the cathode, generated using a pulsed-laser light beam focused down to 10 m (650 nm) to scan over a selected 3×3 pixel area. We applied our digital pulse processing technique to the time-correlated signals captured from adjacent pixels to achieve and evaluate the capability for sub-pixel position resolution. As an example, we also demonstrated an application of 3D corrections to improve the energy resolution and positional information of the events for the tested detectors.« less

Deep feature learning for knee cartilage segmentation using a triplanar convolutional neural network.

PubMed

Prasoon, Adhish; Petersen, Kersten; Igel, Christian; Lauze, François; Dam, Erik; Nielsen, Mads

2013-01-01

Segmentation of anatomical structures in medical images is often based on a voxel/pixel classification approach. Deep learning systems, such as convolutional neural networks (CNNs), can infer a hierarchical representation of images that fosters categorization. We propose a novel system for voxel classification integrating three 2D CNNs, which have a one-to-one association with the xy, yz and zx planes of 3D image, respectively. We applied our method to the segmentation of tibial cartilage in low field knee MRI scans and tested it on 114 unseen scans. Although our method uses only 2D features at a single scale, it performs better than a state-of-the-art method using 3D multi-scale features. In the latter approach, the features and the classifier have been carefully adapted to the problem at hand. That we were able to get better results by a deep learning architecture that autonomously learns the features from the images is the main insight of this study.

Comparison between Frame-Constrained Fix-Pixel-Value and Frame-Free Spiking-Dynamic-Pixel ConvNets for Visual Processing

PubMed Central

Farabet, Clément; Paz, Rafael; Pérez-Carrasco, Jose; Zamarreño-Ramos, Carlos; Linares-Barranco, Alejandro; LeCun, Yann; Culurciello, Eugenio; Serrano-Gotarredona, Teresa; Linares-Barranco, Bernabe

2012-01-01

Most scene segmentation and categorization architectures for the extraction of features in images and patches make exhaustive use of 2D convolution operations for template matching, template search, and denoising. Convolutional Neural Networks (ConvNets) are one example of such architectures that can implement general-purpose bio-inspired vision systems. In standard digital computers 2D convolutions are usually expensive in terms of resource consumption and impose severe limitations for efficient real-time applications. Nevertheless, neuro-cortex inspired solutions, like dedicated Frame-Based or Frame-Free Spiking ConvNet Convolution Processors, are advancing real-time visual processing. These two approaches share the neural inspiration, but each of them solves the problem in different ways. Frame-Based ConvNets process frame by frame video information in a very robust and fast way that requires to use and share the available hardware resources (such as: multipliers, adders). Hardware resources are fixed- and time-multiplexed by fetching data in and out. Thus memory bandwidth and size is important for good performance. On the other hand, spike-based convolution processors are a frame-free alternative that is able to perform convolution of a spike-based source of visual information with very low latency, which makes ideal for very high-speed applications. However, hardware resources need to be available all the time and cannot be time-multiplexed. Thus, hardware should be modular, reconfigurable, and expansible. Hardware implementations in both VLSI custom integrated circuits (digital and analog) and FPGA have been already used to demonstrate the performance of these systems. In this paper we present a comparison study of these two neuro-inspired solutions. A brief description of both systems is presented and also discussions about their differences, pros and cons. PMID:22518097

Comparison between Frame-Constrained Fix-Pixel-Value and Frame-Free Spiking-Dynamic-Pixel ConvNets for Visual Processing.

PubMed

Farabet, Clément; Paz, Rafael; Pérez-Carrasco, Jose; Zamarreño-Ramos, Carlos; Linares-Barranco, Alejandro; Lecun, Yann; Culurciello, Eugenio; Serrano-Gotarredona, Teresa; Linares-Barranco, Bernabe

2012-01-01

Most scene segmentation and categorization architectures for the extraction of features in images and patches make exhaustive use of 2D convolution operations for template matching, template search, and denoising. Convolutional Neural Networks (ConvNets) are one example of such architectures that can implement general-purpose bio-inspired vision systems. In standard digital computers 2D convolutions are usually expensive in terms of resource consumption and impose severe limitations for efficient real-time applications. Nevertheless, neuro-cortex inspired solutions, like dedicated Frame-Based or Frame-Free Spiking ConvNet Convolution Processors, are advancing real-time visual processing. These two approaches share the neural inspiration, but each of them solves the problem in different ways. Frame-Based ConvNets process frame by frame video information in a very robust and fast way that requires to use and share the available hardware resources (such as: multipliers, adders). Hardware resources are fixed- and time-multiplexed by fetching data in and out. Thus memory bandwidth and size is important for good performance. On the other hand, spike-based convolution processors are a frame-free alternative that is able to perform convolution of a spike-based source of visual information with very low latency, which makes ideal for very high-speed applications. However, hardware resources need to be available all the time and cannot be time-multiplexed. Thus, hardware should be modular, reconfigurable, and expansible. Hardware implementations in both VLSI custom integrated circuits (digital and analog) and FPGA have been already used to demonstrate the performance of these systems. In this paper we present a comparison study of these two neuro-inspired solutions. A brief description of both systems is presented and also discussions about their differences, pros and cons.

Video-based face recognition via convolutional neural networks

NASA Astrophysics Data System (ADS)

Bao, Tianlong; Ding, Chunhui; Karmoshi, Saleem; Zhu, Ming

2017-06-01

Face recognition has been widely studied recently while video-based face recognition still remains a challenging task because of the low quality and large intra-class variation of video captured face images. In this paper, we focus on two scenarios of video-based face recognition: 1)Still-to-Video(S2V) face recognition, i.e., querying a still face image against a gallery of video sequences; 2)Video-to-Still(V2S) face recognition, in contrast to S2V scenario. A novel method was proposed in this paper to transfer still and video face images to an Euclidean space by a carefully designed convolutional neural network, then Euclidean metrics are used to measure the distance between still and video images. Identities of still and video images that group as pairs are used as supervision. In the training stage, a joint loss function that measures the Euclidean distance between the predicted features of training pairs and expanding vectors of still images is optimized to minimize the intra-class variation while the inter-class variation is guaranteed due to the large margin of still images. Transferred features are finally learned via the designed convolutional neural network. Experiments are performed on COX face dataset. Experimental results show that our method achieves reliable performance compared with other state-of-the-art methods.

Convolution Operation of Optical Information via Quantum Storage

NASA Astrophysics Data System (ADS)

Li, Zhixiang; Liu, Jianji; Fan, Hongming; Zhang, Guoquan

2017-06-01

We proposed a novel method to achieve optical convolution of two input images via quantum storage based on electromagnetically induced transparency (EIT) effect. By placing an EIT media in the confocal Fourier plane of the 4f-imaging system, the optical convolution of the two input images can be achieved in the image plane.

Error-trellis Syndrome Decoding Techniques for Convolutional Codes

NASA Technical Reports Server (NTRS)

Reed, I. S.; Truong, T. K.

1984-01-01

An error-trellis syndrome decoding technique for convolutional codes is developed. This algorithm is then applied to the entire class of systematic convolutional codes and to the high-rate, Wyner-Ash convolutional codes. A special example of the one-error-correcting Wyner-Ash code, a rate 3/4 code, is treated. The error-trellis syndrome decoding method applied to this example shows in detail how much more efficient syndrome decoding is than Viterbi decoding if applied to the same problem. For standard Viterbi decoding, 64 states are required, whereas in the example only 7 states are needed. Also, within the 7 states required for decoding, many fewer transitions are needed between the states.

Error-trellis syndrome decoding techniques for convolutional codes

NASA Technical Reports Server (NTRS)

Reed, I. S.; Truong, T. K.

1985-01-01

An error-trellis syndrome decoding technique for convolutional codes is developed. This algorithm is then applied to the entire class of systematic convolutional codes and to the high-rate, Wyner-Ash convolutional codes. A special example of the one-error-correcting Wyner-Ash code, a rate 3/4 code, is treated. The error-trellis syndrome decoding method applied to this example shows in detail how much more efficient syndrome decordig is than Viterbi decoding if applied to the same problem. For standard Viterbi decoding, 64 states are required, whereas in the example only 7 states are needed. Also, within the 7 states required for decoding, many fewer transitions are needed between the states.

Ultra-low power high-dynamic range color pixel embedding RGB to r-g chromaticity transformation

NASA Astrophysics Data System (ADS)

Lecca, Michela; Gasparini, Leonardo; Gottardi, Massimo

2014-05-01

This work describes a novel color pixel topology that converts the three chromatic components from the standard RGB space into the normalized r-g chromaticity space. This conversion is implemented with high-dynamic range and with no dc power consumption, and the auto-exposure capability of the sensor ensures to capture a high quality chromatic signal, even in presence of very bright illuminants or in the darkness. The pixel is intended to become the basic building block of a CMOS color vision sensor, targeted to ultra-low power applications for mobile devices, such as human machine interfaces, gesture recognition, face detection. The experiments show that significant improvements of the proposed pixel with respect to standard cameras in terms of energy saving and accuracy on data acquisition. An application to skin color-based description is presented.

Video Super-Resolution via Bidirectional Recurrent Convolutional Networks.

PubMed

Huang, Yan; Wang, Wei; Wang, Liang

2018-04-01

Super resolving a low-resolution video, namely video super-resolution (SR), is usually handled by either single-image SR or multi-frame SR. Single-Image SR deals with each video frame independently, and ignores intrinsic temporal dependency of video frames which actually plays a very important role in video SR. Multi-Frame SR generally extracts motion information, e.g., optical flow, to model the temporal dependency, but often shows high computational cost. Considering that recurrent neural networks (RNNs) can model long-term temporal dependency of video sequences well, we propose a fully convolutional RNN named bidirectional recurrent convolutional network for efficient multi-frame SR. Different from vanilla RNNs, 1) the commonly-used full feedforward and recurrent connections are replaced with weight-sharing convolutional connections. So they can greatly reduce the large number of network parameters and well model the temporal dependency in a finer level, i.e., patch-based rather than frame-based, and 2) connections from input layers at previous timesteps to the current hidden layer are added by 3D feedforward convolutions, which aim to capture discriminate spatio-temporal patterns for short-term fast-varying motions in local adjacent frames. Due to the cheap convolutional operations, our model has a low computational complexity and runs orders of magnitude faster than other multi-frame SR methods. With the powerful temporal dependency modeling, our model can super resolve videos with complex motions and achieve well performance.

Detecting atrial fibrillation by deep convolutional neural networks.

PubMed

Xia, Yong; Wulan, Naren; Wang, Kuanquan; Zhang, Henggui

2018-02-01

Atrial fibrillation (AF) is the most common cardiac arrhythmia. The incidence of AF increases with age, causing high risks of stroke and increased morbidity and mortality. Efficient and accurate diagnosis of AF based on the ECG is valuable in clinical settings and remains challenging. In this paper, we proposed a novel method with high reliability and accuracy for AF detection via deep learning. The short-term Fourier transform (STFT) and stationary wavelet transform (SWT) were used to analyze ECG segments to obtain two-dimensional (2-D) matrix input suitable for deep convolutional neural networks. Then, two different deep convolutional neural network models corresponding to STFT output and SWT output were developed. Our new method did not require detection of P or R peaks, nor feature designs for classification, in contrast to existing algorithms. Finally, the performances of the two models were evaluated and compared with those of existing algorithms. Our proposed method demonstrated favorable performances on ECG segments as short as 5 s. The deep convolutional neural network using input generated by STFT, presented a sensitivity of 98.34%, specificity of 98.24% and accuracy of 98.29%. For the deep convolutional neural network using input generated by SWT, a sensitivity of 98.79%, specificity of 97.87% and accuracy of 98.63% was achieved. The proposed method using deep convolutional neural networks shows high sensitivity, specificity and accuracy, and, therefore, is a valuable tool for AF detection. Copyright © 2017 Elsevier Ltd. All rights reserved.

Classification of crystal structure using a convolutional neural network

PubMed Central

Park, Woon Bae; Chung, Jiyong; Sohn, Keemin; Pyo, Myoungho

2017-01-01

A deep machine-learning technique based on a convolutional neural network (CNN) is introduced. It has been used for the classification of powder X-ray diffraction (XRD) patterns in terms of crystal system, extinction group and space group. About 150 000 powder XRD patterns were collected and used as input for the CNN with no handcrafted engineering involved, and thereby an appropriate CNN architecture was obtained that allowed determination of the crystal system, extinction group and space group. In sharp contrast with the traditional use of powder XRD pattern analysis, the CNN never treats powder XRD patterns as a deconvoluted and discrete peak position or as intensity data, but instead the XRD patterns are regarded as nothing but a pattern similar to a picture. The CNN interprets features that humans cannot recognize in a powder XRD pattern. As a result, accuracy levels of 81.14, 83.83 and 94.99% were achieved for the space-group, extinction-group and crystal-system classifications, respectively. The well trained CNN was then used for symmetry identification of unknown novel inorganic compounds. PMID:28875035

Classification of crystal structure using a convolutional neural network.

PubMed

Park, Woon Bae; Chung, Jiyong; Jung, Jaeyoung; Sohn, Keemin; Singh, Satendra Pal; Pyo, Myoungho; Shin, Namsoo; Sohn, Kee-Sun

2017-07-01

A deep machine-learning technique based on a convolutional neural network (CNN) is introduced. It has been used for the classification of powder X-ray diffraction (XRD) patterns in terms of crystal system, extinction group and space group. About 150 000 powder XRD patterns were collected and used as input for the CNN with no handcrafted engineering involved, and thereby an appropriate CNN architecture was obtained that allowed determination of the crystal system, extinction group and space group. In sharp contrast with the traditional use of powder XRD pattern analysis, the CNN never treats powder XRD patterns as a deconvoluted and discrete peak position or as intensity data, but instead the XRD patterns are regarded as nothing but a pattern similar to a picture. The CNN interprets features that humans cannot recognize in a powder XRD pattern. As a result, accuracy levels of 81.14, 83.83 and 94.99% were achieved for the space-group, extinction-group and crystal-system classifications, respectively. The well trained CNN was then used for symmetry identification of unknown novel inorganic compounds.

A novel end-to-end classifier using domain transferred deep convolutional neural networks for biomedical images.

PubMed

Pang, Shuchao; Yu, Zhezhou; Orgun, Mehmet A

2017-03-01

Highly accurate classification of biomedical images is an essential task in the clinical diagnosis of numerous medical diseases identified from those images. Traditional image classification methods combined with hand-crafted image feature descriptors and various classifiers are not able to effectively improve the accuracy rate and meet the high requirements of classification of biomedical images. The same also holds true for artificial neural network models directly trained with limited biomedical images used as training data or directly used as a black box to extract the deep features based on another distant dataset. In this study, we propose a highly reliable and accurate end-to-end classifier for all kinds of biomedical images via deep learning and transfer learning. We first apply domain transferred deep convolutional neural network for building a deep model; and then develop an overall deep learning architecture based on the raw pixels of original biomedical images using supervised training. In our model, we do not need the manual design of the feature space, seek an effective feature vector classifier or segment specific detection object and image patches, which are the main technological difficulties in the adoption of traditional image classification methods. Moreover, we do not need to be concerned with whether there are large training sets of annotated biomedical images, affordable parallel computing resources featuring GPUs or long times to wait for training a perfect deep model, which are the main problems to train deep neural networks for biomedical image classification as observed in recent works. With the utilization of a simple data augmentation method and fast convergence speed, our algorithm can achieve the best accuracy rate and outstanding classification ability for biomedical images. We have evaluated our classifier on several well-known public biomedical datasets and compared it with several state-of-the-art approaches. We propose a robust

Classification of urine sediment based on convolution neural network

NASA Astrophysics Data System (ADS)

Pan, Jingjing; Jiang, Cunbo; Zhu, Tiantian

2018-04-01

By designing a new convolution neural network framework, this paper breaks the constraints of the original convolution neural network framework requiring large training samples and samples of the same size. Move and cropping the input images, generate the same size of the sub-graph. And then, the generated sub-graph uses the method of dropout, increasing the diversity of samples and preventing the fitting generation. Randomly select some proper subset in the sub-graphic set and ensure that the number of elements in the proper subset is same and the proper subset is not the same. The proper subsets are used as input layers for the convolution neural network. Through the convolution layer, the pooling, the full connection layer and output layer, we can obtained the classification loss rate of test set and training set. In the red blood cells, white blood cells, calcium oxalate crystallization classification experiment, the classification accuracy rate of 97% or more.

Characterisation of Vanilla—A novel active pixel sensor for radiation detection

NASA Astrophysics Data System (ADS)

Blue, A.; Bates, R.; Laing, A.; Maneuski, D.; O'Shea, V.; Clark, A.; Prydderch, M.; Turchetta, R.; Arvanitis, C.; Bohndiek, S.

2007-10-01

Novel features of a new monolithic active pixel sensor, Vanilla, with 520×520 pixels ( 25 μm square) has been characterised for the first time. Optimisation of the sensor operation was made through variation of frame rates, integration times and on-chip biases and voltages. Features such as flushed reset operation, ROI capturing and readout modes have been fully tested. Stability measurements were performed to test its suitablility for long-term applications. These results suggest the Vanilla sensor—along with bio-medical and space applications—is suitable for use in particle physics experiments.

ELHnet: a convolutional neural network for classifying cochlear endolymphatic hydrops imaged with optical coherence tomography

PubMed Central

Liu, George S.; Zhu, Michael H.; Kim, Jinkyung; Raphael, Patrick; Applegate, Brian E.; Oghalai, John S.

2017-01-01

Detection of endolymphatic hydrops is important for diagnosing Meniere’s disease, and can be performed non-invasively using optical coherence tomography (OCT) in animal models as well as potentially in the clinic. Here, we developed ELHnet, a convolutional neural network to classify endolymphatic hydrops in a mouse model using learned features from OCT images of mice cochleae. We trained ELHnet on 2159 training and validation images from 17 mice, using only the image pixels and observer-determined labels of endolymphatic hydrops as the inputs. We tested ELHnet on 37 images from 37 mice that were previously not used, and found that the neural network correctly classified 34 of the 37 mice. This demonstrates an improvement in performance from previous work on computer-aided classification of endolymphatic hydrops. To the best of our knowledge, this is the first deep CNN designed for endolymphatic hydrops classification. PMID:29082086

A nonvoxel-based dose convolution/superposition algorithm optimized for scalable GPU architectures.

PubMed

Neylon, J; Sheng, K; Yu, V; Chen, Q; Low, D A; Kupelian, P; Santhanam, A

2014-10-01

Real-time adaptive planning and treatment has been infeasible due in part to its high computational complexity. There have been many recent efforts to utilize graphics processing units (GPUs) to accelerate the computational performance and dose accuracy in radiation therapy. Data structure and memory access patterns are the key GPU factors that determine the computational performance and accuracy. In this paper, the authors present a nonvoxel-based (NVB) approach to maximize computational and memory access efficiency and throughput on the GPU. The proposed algorithm employs a ray-tracing mechanism to restructure the 3D data sets computed from the CT anatomy into a nonvoxel-based framework. In a process that takes only a few milliseconds of computing time, the algorithm restructured the data sets by ray-tracing through precalculated CT volumes to realign the coordinate system along the convolution direction, as defined by zenithal and azimuthal angles. During the ray-tracing step, the data were resampled according to radial sampling and parallel ray-spacing parameters making the algorithm independent of the original CT resolution. The nonvoxel-based algorithm presented in this paper also demonstrated a trade-off in computational performance and dose accuracy for different coordinate system configurations. In order to find the best balance between the computed speedup and the accuracy, the authors employed an exhaustive parameter search on all sampling parameters that defined the coordinate system configuration: zenithal, azimuthal, and radial sampling of the convolution algorithm, as well as the parallel ray spacing during ray tracing. The angular sampling parameters were varied between 4 and 48 discrete angles, while both radial sampling and parallel ray spacing were varied from 0.5 to 10 mm. The gamma distribution analysis method (γ) was used to compare the dose distributions using 2% and 2 mm dose difference and distance-to-agreement criteria, respectively

A nonvoxel-based dose convolution/superposition algorithm optimized for scalable GPU architectures

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

Neylon, J., E-mail: jneylon@mednet.ucla.edu; Sheng, K.; Yu, V.

Purpose: Real-time adaptive planning and treatment has been infeasible due in part to its high computational complexity. There have been many recent efforts to utilize graphics processing units (GPUs) to accelerate the computational performance and dose accuracy in radiation therapy. Data structure and memory access patterns are the key GPU factors that determine the computational performance and accuracy. In this paper, the authors present a nonvoxel-based (NVB) approach to maximize computational and memory access efficiency and throughput on the GPU. Methods: The proposed algorithm employs a ray-tracing mechanism to restructure the 3D data sets computed from the CT anatomy intomore » a nonvoxel-based framework. In a process that takes only a few milliseconds of computing time, the algorithm restructured the data sets by ray-tracing through precalculated CT volumes to realign the coordinate system along the convolution direction, as defined by zenithal and azimuthal angles. During the ray-tracing step, the data were resampled according to radial sampling and parallel ray-spacing parameters making the algorithm independent of the original CT resolution. The nonvoxel-based algorithm presented in this paper also demonstrated a trade-off in computational performance and dose accuracy for different coordinate system configurations. In order to find the best balance between the computed speedup and the accuracy, the authors employed an exhaustive parameter search on all sampling parameters that defined the coordinate system configuration: zenithal, azimuthal, and radial sampling of the convolution algorithm, as well as the parallel ray spacing during ray tracing. The angular sampling parameters were varied between 4 and 48 discrete angles, while both radial sampling and parallel ray spacing were varied from 0.5 to 10 mm. The gamma distribution analysis method (γ) was used to compare the dose distributions using 2% and 2 mm dose difference and distance

Charge Sharing and Charge Loss in a Cadmium-Zinc-Telluride Fine-Pixel Detector Array

NASA Technical Reports Server (NTRS)

Gaskin, J. A.; Sharma, D. P.; Ramsey, B. D.; Six, N. Frank (Technical Monitor)

2002-01-01

Because of its high atomic number, room temperature operation, low noise, and high spatial resolution a Cadmium-Zinc-Telluride (CZT) multi-pixel detector is ideal for hard x-ray astrophysical observation. As part of on-going research at MSFC (Marshall Space Flight Center) to develop multi-pixel CdZnTe detectors for this purpose, we have measured charge sharing and charge loss for a 4x4 (750micron pitch), lmm thick pixel array and modeled these results using a Monte-Carlo simulation. This model was then used to predict the amount of charge sharing for a much finer pixel array (with a 300micron pitch). Future work will enable us to compare the simulated results for the finer array to measured values.

Error control techniques for satellite and space communications

NASA Technical Reports Server (NTRS)

Costello, Daniel J., Jr.

1992-01-01

Worked performed during the reporting period is summarized. Construction of robustly good trellis codes for use with sequential decoding was developed. The robustly good trellis codes provide a much better trade off between free distance and distance profile. The unequal error protection capabilities of convolutional codes was studied. The problem of finding good large constraint length, low rate convolutional codes for deep space applications is investigated. A formula for computing the free distance of 1/n convolutional codes was discovered. Double memory (DM) codes, codes with two memory units per unit bit position, were studied; a search for optimal DM codes is being conducted. An algorithm for constructing convolutional codes from a given quasi-cyclic code was developed. Papers based on the above work are included in the appendix.

Landcover Classification Using Deep Fully Convolutional Neural Networks

NASA Astrophysics Data System (ADS)

Wang, J.; Li, X.; Zhou, S.; Tang, J.

2017-12-01

Land cover classification has always been an essential application in remote sensing. Certain image features are needed for land cover classification whether it is based on pixel or object-based methods. Different from other machine learning methods, deep learning model not only extracts useful information from multiple bands/attributes, but also learns spatial characteristics. In recent years, deep learning methods have been developed rapidly and widely applied in image recognition, semantic understanding, and other application domains. However, there are limited studies applying deep learning methods in land cover classification. In this research, we used fully convolutional networks (FCN) as the deep learning model to classify land covers. The National Land Cover Database (NLCD) within the state of Kansas was used as training dataset and Landsat images were classified using the trained FCN model. We also applied an image segmentation method to improve the original results from the FCN model. In addition, the pros and cons between deep learning and several machine learning methods were compared and explored. Our research indicates: (1) FCN is an effective classification model with an overall accuracy of 75%; (2) image segmentation improves the classification results with better match of spatial patterns; (3) FCN has an excellent ability of learning which can attains higher accuracy and better spatial patterns compared with several machine learning methods.

A Geometric Construction of Cyclic Cocycles on Twisted Convolution Algebras

NASA Astrophysics Data System (ADS)

Angel, Eitan

2010-09-01

In this thesis we give a construction of cyclic cocycles on convolution algebras twisted by gerbes over discrete translation groupoids. In his seminal book, Connes constructs a map from the equivariant cohomology of a manifold carrying the action of a discrete group into the periodic cyclic cohomology of the associated convolution algebra. Furthermore, for proper étale groupoids, J.-L. Tu and P. Xu provide a map between the periodic cyclic cohomology of a gerbe twisted convolution algebra and twisted cohomology groups. Our focus will be the convolution algebra with a product defined by a gerbe over a discrete translation groupoid. When the action is not proper, we cannot construct an invariant connection on the gerbe; therefore to study this algebra, we instead develop simplicial notions related to ideas of J. Dupont to construct a simplicial form representing the Dixmier-Douady class of the gerbe. Then by using a JLO formula we define a morphism from a simplicial complex twisted by this simplicial Dixmier-Douady form to the mixed bicomplex of certain matrix algebras. Finally, we define a morphism from this complex to the mixed bicomplex computing the periodic cyclic cohomology of the twisted convolution algebras.

Convolution of large 3D images on GPU and its decomposition

NASA Astrophysics Data System (ADS)

Karas, Pavel; Svoboda, David

2011-12-01

In this article, we propose a method for computing convolution of large 3D images. The convolution is performed in a frequency domain using a convolution theorem. The algorithm is accelerated on a graphic card by means of the CUDA parallel computing model. Convolution is decomposed in a frequency domain using the decimation in frequency algorithm. We pay attention to keeping our approach efficient in terms of both time and memory consumption and also in terms of memory transfers between CPU and GPU which have a significant inuence on overall computational time. We also study the implementation on multiple GPUs and compare the results between the multi-GPU and multi-CPU implementations.

Efficient space-time sampling with pixel-wise coded exposure for high-speed imaging.

PubMed

Liu, Dengyu; Gu, Jinwei; Hitomi, Yasunobu; Gupta, Mohit; Mitsunaga, Tomoo; Nayar, Shree K

2014-02-01

Cameras face a fundamental trade-off between spatial and temporal resolution. Digital still cameras can capture images with high spatial resolution, but most high-speed video cameras have relatively low spatial resolution. It is hard to overcome this trade-off without incurring a significant increase in hardware costs. In this paper, we propose techniques for sampling, representing, and reconstructing the space-time volume to overcome this trade-off. Our approach has two important distinctions compared to previous works: 1) We achieve sparse representation of videos by learning an overcomplete dictionary on video patches, and 2) we adhere to practical hardware constraints on sampling schemes imposed by architectures of current image sensors, which means that our sampling function can be implemented on CMOS image sensors with modified control units in the future. We evaluate components of our approach, sampling function and sparse representation, by comparing them to several existing approaches. We also implement a prototype imaging system with pixel-wise coded exposure control using a liquid crystal on silicon device. System characteristics such as field of view and modulation transfer function are evaluated for our imaging system. Both simulations and experiments on a wide range of scenes show that our method can effectively reconstruct a video from a single coded image while maintaining high spatial resolution.

Performance of MIMO-OFDM using convolution codes with QAM modulation

NASA Astrophysics Data System (ADS)

Astawa, I. Gede Puja; Moegiharto, Yoedy; Zainudin, Ahmad; Salim, Imam Dui Agus; Anggraeni, Nur Annisa

2014-04-01

Performance of Orthogonal Frequency Division Multiplexing (OFDM) system can be improved by adding channel coding (error correction code) to detect and correct errors that occur during data transmission. One can use the convolution code. This paper present performance of OFDM using Space Time Block Codes (STBC) diversity technique use QAM modulation with code rate ½. The evaluation is done by analyzing the value of Bit Error Rate (BER) vs Energy per Bit to Noise Power Spectral Density Ratio (Eb/No). This scheme is conducted 256 subcarrier which transmits Rayleigh multipath fading channel in OFDM system. To achieve a BER of 10-3 is required 10dB SNR in SISO-OFDM scheme. For 2×2 MIMO-OFDM scheme requires 10 dB to achieve a BER of 10-3. For 4×4 MIMO-OFDM scheme requires 5 dB while adding convolution in a 4x4 MIMO-OFDM can improve performance up to 0 dB to achieve the same BER. This proves the existence of saving power by 3 dB of 4×4 MIMO-OFDM system without coding, power saving 7 dB of 2×2 MIMO-OFDM and significant power savings from SISO-OFDM system.

Frame prediction using recurrent convolutional encoder with residual learning

NASA Astrophysics Data System (ADS)

Yue, Boxuan; Liang, Jun

2018-05-01

The prediction for the frame of a video is difficult but in urgent need in auto-driving. Conventional methods can only predict some abstract trends of the region of interest. The boom of deep learning makes the prediction for frames possible. In this paper, we propose a novel recurrent convolutional encoder and DE convolutional decoder structure to predict frames. We introduce the residual learning in the convolution encoder structure to solve the gradient issues. The residual learning can transform the gradient back propagation to an identity mapping. It can reserve the whole gradient information and overcome the gradient issues in Recurrent Neural Networks (RNN) and Convolutional Neural Networks (CNN). Besides, compared with the branches in CNNs and the gated structures in RNNs, the residual learning can save the training time significantly. In the experiments, we use UCF101 dataset to train our networks, the predictions are compared with some state-of-the-art methods. The results show that our networks can predict frames fast and efficiently. Furthermore, our networks are used for the driving video to verify the practicability.

Segmentation of corneal endothelium images using a U-Net-based convolutional neural network.

PubMed

Fabijańska, Anna

2018-04-18

Diagnostic information regarding the health status of the corneal endothelium may be obtained by analyzing the size and the shape of the endothelial cells in specular microscopy images. Prior to the analysis, the endothelial cells need to be extracted from the image. Up to today, this has been performed manually or semi-automatically. Several approaches to automatic segmentation of endothelial cells exist; however, none of them is perfect. Therefore this paper proposes to perform cell segmentation using a U-Net-based convolutional neural network. Particularly, the network is trained to discriminate pixels located at the borders between cells. The edge probability map outputted by the network is next binarized and skeletonized in order to obtain one-pixel wide edges. The proposed solution was tested on a dataset consisting of 30 corneal endothelial images presenting cells of different sizes, achieving an AUROC level of 0.92. The resulting DICE is on average equal to 0.86, which is a good result, regarding the thickness of the compared edges. The corresponding mean absolute percentage error of cell number is at the level of 4.5% which confirms the high accuracy of the proposed approach. The resulting cell edges are well aligned to the ground truths and require a limited number of manual corrections. This also results in accurate values of the cell morphometric parameters. The corresponding errors range from 5.2% for endothelial cell density, through 6.2% for cell hexagonality to 11.93% for the coefficient of variation of the cell size. Copyright © 2018 Elsevier B.V. All rights reserved.

Single image super-resolution based on convolutional neural networks

NASA Astrophysics Data System (ADS)

Zou, Lamei; Luo, Ming; Yang, Weidong; Li, Peng; Jin, Liujia

2018-03-01

We present a deep learning method for single image super-resolution (SISR). The proposed approach learns end-to-end mapping between low-resolution (LR) images and high-resolution (HR) images. The mapping is represented as a deep convolutional neural network which inputs the LR image and outputs the HR image. Our network uses 5 convolution layers, which kernels size include 5×5, 3×3 and 1×1. In our proposed network, we use residual-learning and combine different sizes of convolution kernels at the same layer. The experiment results show that our proposed method performs better than the existing methods in reconstructing quality index and human visual effects on benchmarked images.

High-speed imaging using CMOS image sensor with quasi pixel-wise exposure

NASA Astrophysics Data System (ADS)

Sonoda, T.; Nagahara, H.; Endo, K.; Sugiyama, Y.; Taniguchi, R.

2017-02-01

Several recent studies in compressive video sensing have realized scene capture beyond the fundamental trade-off limit between spatial resolution and temporal resolution using random space-time sampling. However, most of these studies showed results for higher frame rate video that were produced by simulation experiments or using an optically simulated random sampling camera, because there are currently no commercially available image sensors with random exposure or sampling capabilities. We fabricated a prototype complementary metal oxide semiconductor (CMOS) image sensor with quasi pixel-wise exposure timing that can realize nonuniform space-time sampling. The prototype sensor can reset exposures independently by columns and fix these amount of exposure by rows for each 8x8 pixel block. This CMOS sensor is not fully controllable via the pixels, and has line-dependent controls, but it offers flexibility when compared with regular CMOS or charge-coupled device sensors with global or rolling shutters. We propose a method to realize pseudo-random sampling for high-speed video acquisition that uses the flexibility of the CMOS sensor. We reconstruct the high-speed video sequence from the images produced by pseudo-random sampling using an over-complete dictionary.

Error-Trellis Construction for Convolutional Codes Using Shifted Error/Syndrome-Subsequences

NASA Astrophysics Data System (ADS)

Tajima, Masato; Okino, Koji; Miyagoshi, Takashi

In this paper, we extend the conventional error-trellis construction for convolutional codes to the case where a given check matrix H(D) has a factor Dl in some column (row). In the first case, there is a possibility that the size of the state space can be reduced using shifted error-subsequences, whereas in the second case, the size of the state space can be reduced using shifted syndrome-subsequences. The construction presented in this paper is based on the adjoint-obvious realization of the corresponding syndrome former HT(D). In the case where all the columns and rows of H(D) are delay free, the proposed construction is reduced to the conventional one of Schalkwijk et al. We also show that the proposed construction can equally realize the state-space reduction shown by Ariel et al. Moreover, we clarify the difference between their construction and that of ours using examples.

Pixel-by-pixel absolute phase retrieval using three phase-shifted fringe patterns without markers

NASA Astrophysics Data System (ADS)

Jiang, Chufan; Li, Beiwen; Zhang, Song

2017-04-01

This paper presents a method that can recover absolute phase pixel by pixel without embedding markers on three phase-shifted fringe patterns, acquiring additional images, or introducing additional hardware component(s). The proposed three-dimensional (3D) absolute shape measurement technique includes the following major steps: (1) segment the measured object into different regions using rough priori knowledge of surface geometry; (2) artificially create phase maps at different z planes using geometric constraints of structured light system; (3) unwrap the phase pixel by pixel for each region by properly referring to the artificially created phase map; and (4) merge unwrapped phases from all regions into a complete absolute phase map for 3D reconstruction. We demonstrate that conventional three-step phase-shifted fringe patterns can be used to create absolute phase map pixel by pixel even for large depth range objects. We have successfully implemented our proposed computational framework to achieve absolute 3D shape measurement at 40 Hz.

How many pixels does it take to make a good 4"×6" print? Pixel count wars revisited

NASA Astrophysics Data System (ADS)

Kriss, Michael A.

2011-01-01

In the early 1980's the future of conventional silver-halide photographic systems was of great concern due to the potential introduction of electronic imaging systems then typified by the Sony Mavica analog electronic camera. The focus was on the quality of film-based systems as expressed in the number of equivalent number pixels and bits-per-pixel, and how many pixels would be required to create an equivalent quality image from a digital camera. It was found that 35-mm frames, for ISO 100 color negative film, contained equivalent pixels of 12 microns for a total of 18 million pixels per frame (6 million pixels per layer) with about 6 bits of information per pixel; the introduction of new emulsion technology, tabular AgX grains, increased the value to 8 bit per pixel. Higher ISO speed films had larger equivalent pixels, fewer pixels per frame, but retained the 8 bits per pixel. Further work found that a high quality 3.5" x 5.25" print could be obtained from a three layer system containing 1300 x 1950 pixels per layer or about 7.6 million pixels in all. In short, it became clear that when a digital camera contained about 6 million pixels (in a single layer using a color filter array and appropriate image processing) that digital systems would challenge and replace conventional film-based system for the consumer market. By 2005 this became the reality. Since 2005 there has been a "pixel war" raging amongst digital camera makers. The question arises about just how many pixels are required and are all pixels equal? This paper will provide a practical look at how many pixels are needed for a good print based on the form factor of the sensor (sensor size) and the effective optical modulation transfer function (optical spread function) of the camera lens. Is it better to have 16 million, 5.7-micron pixels or 6 million 7.8-micron pixels? How does intrinsic (no electronic boost) ISO speed and exposure latitude vary with pixel size? A systematic review of these issues will

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.

Space suit

NASA Technical Reports Server (NTRS)

Shepard, L. F.; Durney, G. P.; Case, M. C.; Kenneway, A. J., III; Wise, R. C.; Rinehart, D.; Bessette, R. J.; Pulling, R. C. (Inventor)

1973-01-01

A pressure suit for high altitude flights, particularly space missions is reported. The suit is designed for astronauts in the Apollo space program and may be worn both inside and outside a space vehicle, as well as on the lunar surface. It comprises an integrated assembly of inner comfort liner, intermediate pressure garment, and outer thermal protective garment with removable helmet, and gloves. The pressure garment comprises an inner convoluted sealing bladder and outer fabric restraint to which are attached a plurality of cable restraint assemblies. It provides versitility in combination with improved sealing and increased mobility for internal pressures suitable for life support in the near vacuum of outer space.

Development of pixellated Ir-TESs

NASA Astrophysics Data System (ADS)

Zen, Nobuyuki; Takahashi, Hiroyuki; Kunieda, Yuichi; Damayanthi, Rathnayaka M. T.; Mori, Fumiakira; Fujita, Kaoru; Nakazawa, Masaharu; Fukuda, Daiji; Ohkubo, Masataka

2006-04-01

We have been developing Ir-based pixellated superconducting transition edge sensors (TESs). In the area of material or astronomical applications, the sensor with few eV energy resolution and over 1000 pixels imaging property is desired. In order to achieve this goal, we have been analyzing signals from pixellated TESs. In the case of a 20 pixel array of Ir-TESs, with 45 μm×45 μm pixel sizes, the incident X-ray signals have been classified into 16 groups. We have applied numerical signal analysis. On the one hand, the energy resolution of our pixellated TES is strongly degraded. However, using pulse shape analysis, we can dramatically improve the resolution. Thus, we consider that the pulse signal analysis will lead this device to be used as a practical photon incident position identifying TES.

Low complexity pixel-based halftone detection

NASA Astrophysics Data System (ADS)

Ok, Jiheon; Han, Seong Wook; Jarno, Mielikainen; Lee, Chulhee

2011-10-01

With the rapid advances of the internet and other multimedia technologies, the digital document market has been growing steadily. Since most digital images use halftone technologies, quality degradation occurs when one tries to scan and reprint them. Therefore, it is necessary to extract the halftone areas to produce high quality printing. In this paper, we propose a low complexity pixel-based halftone detection algorithm. For each pixel, we considered a surrounding block. If the block contained any flat background regions, text, thin lines, or continuous or non-homogeneous regions, the pixel was classified as a non-halftone pixel. After excluding those non-halftone pixels, the remaining pixels were considered to be halftone pixels. Finally, documents were classified as pictures or photo documents by calculating the halftone pixel ratio. The proposed algorithm proved to be memory-efficient and required low computation costs. The proposed algorithm was easily implemented using GPU.

Pixelated gamma detector

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

Dolinsky, Sergei Ivanovich; Yanoff, Brian David; Guida, Renato

2016-12-27

A pixelated gamma detector includes a scintillator column assembly having scintillator crystals and optical transparent elements alternating along a longitudinal axis, a collimator assembly having longitudinal walls separated by collimator septum, the collimator septum spaced apart to form collimator channels, the scintillator column assembly positioned adjacent to the collimator assembly so that the respective ones of the scintillator crystal are positioned adjacent to respective ones of the collimator channels, the respective ones of the optical transparent element are positioned adjacent to respective ones of the collimator septum, and a first photosensor and a second photosensor, the first and the secondmore » photosensor each connected to an opposing end of the scintillator column assembly. A system and a method for inspecting and/or detecting defects in an interior of an object are also disclosed.« less

Detector Sampling of Optical/IR Spectra: How Many Pixels per FWHM?

NASA Astrophysics Data System (ADS)

Robertson, J. Gordon

2017-08-01

Most optical and IR spectra are now acquired using detectors with finite-width pixels in a square array. Each pixel records the received intensity integrated over its own area, and pixels are separated by the array pitch. This paper examines the effects of such pixellation, using computed simulations to illustrate the effects which most concern the astronomer end-user. It is shown that coarse sampling increases the random noise errors in wavelength by typically 10-20 % at 2 pixels per Full Width at Half Maximum, but with wide variation depending on the functional form of the instrumental Line Spread Function (i.e. the instrumental response to a monochromatic input) and on the pixel phase. If line widths are determined, they are even more strongly affected at low sampling frequencies. However, the noise in fitted peak amplitudes is minimally affected by pixellation, with increases less than about 5%. Pixellation has a substantial but complex effect on the ability to see a relative minimum between two closely spaced peaks (or relative maximum between two absorption lines). The consistent scale of resolving power presented by Robertson to overcome the inadequacy of the Full Width at Half Maximum as a resolution measure is here extended to cover pixellated spectra. The systematic bias errors in wavelength introduced by pixellation, independent of signal/noise ratio, are examined. While they may be negligible for smooth well-sampled symmetric Line Spread Functions, they are very sensitive to asymmetry and high spatial frequency sub-structure. The Modulation Transfer Function for sampled data is shown to give a useful indication of the extent of improperly sampled signal in an Line Spread Function. The common maxim that 2 pixels per Full Width at Half Maximum is the Nyquist limit is incorrect and most Line Spread Functions will exhibit some aliasing at this sample frequency. While 2 pixels per Full Width at Half Maximum is nevertheless often an acceptable minimum for

Development and application of deep convolutional neural network in target detection

NASA Astrophysics Data System (ADS)

Jiang, Xiaowei; Wang, Chunping; Fu, Qiang

2018-04-01

With the development of big data and algorithms, deep convolution neural networks with more hidden layers have more powerful feature learning and feature expression ability than traditional machine learning methods, making artificial intelligence surpass human level in many fields. This paper first reviews the development and application of deep convolutional neural networks in the field of object detection in recent years, then briefly summarizes and ponders some existing problems in the current research, and the future development of deep convolutional neural network is prospected.

Optical proximity correction (OPC) in near-field lithography with pixel-based field sectioning time modulation

NASA Astrophysics Data System (ADS)

Oh, Seonghyeon; Han, Dandan; Shim, Hyeon Bo; Hahn, Jae W.

2018-01-01

Subwavelength features have been successfully demonstrated in near-field lithography. In this study, the point spread function (PSF) of a near-field beam spot from a plasmonic ridge nanoaperture is discussed with regard to the complex decaying characteristic of a non-propagating wave and the asymmetry of the field distribution for pattern design. We relaxed the shape complexity of the field distribution with pixel-based optical proximity correction (OPC) for simplifying the pattern image distortion. To enhance the pattern fidelity for a variety of arbitrary patterns, field-sectioning structures are formulated via convolutions with a time-modulation function and a transient PSF along the near-field dominant direction. The sharpness of corners and edges, and line shortening can be improved by modifying the original target pattern shape using the proposed approach by considering both the pattern geometry and directionality of the field decay for OPC in near-field lithography.

Optical proximity correction (OPC) in near-field lithography with pixel-based field sectioning time modulation.

PubMed

Oh, Seonghyeon; Han, Dandan; Shim, Hyeon Bo; Hahn, Jae W

2018-01-26

Subwavelength features have been successfully demonstrated in near-field lithography. In this study, the point spread function (PSF) of a near-field beam spot from a plasmonic ridge nanoaperture is discussed with regard to the complex decaying characteristic of a non-propagating wave and the asymmetry of the field distribution for pattern design. We relaxed the shape complexity of the field distribution with pixel-based optical proximity correction (OPC) for simplifying the pattern image distortion. To enhance the pattern fidelity for a variety of arbitrary patterns, field-sectioning structures are formulated via convolutions with a time-modulation function and a transient PSF along the near-field dominant direction. The sharpness of corners and edges, and line shortening can be improved by modifying the original target pattern shape using the proposed approach by considering both the pattern geometry and directionality of the field decay for OPC in near-field lithography.

Automatic sleep stage classification of single-channel EEG by using complex-valued convolutional neural network.

PubMed

Zhang, Junming; Wu, Yan

2018-03-28

Many systems are developed for automatic sleep stage classification. However, nearly all models are based on handcrafted features. Because of the large feature space, there are so many features that feature selection should be used. Meanwhile, designing handcrafted features is a difficult and time-consuming task because the feature designing needs domain knowledge of experienced experts. Results vary when different sets of features are chosen to identify sleep stages. Additionally, many features that we may be unaware of exist. However, these features may be important for sleep stage classification. Therefore, a new sleep stage classification system, which is based on the complex-valued convolutional neural network (CCNN), is proposed in this study. Unlike the existing sleep stage methods, our method can automatically extract features from raw electroencephalography data and then classify sleep stage based on the learned features. Additionally, we also prove that the decision boundaries for the real and imaginary parts of a complex-valued convolutional neuron intersect orthogonally. The classification performances of handcrafted features are compared with those of learned features via CCNN. Experimental results show that the proposed method is comparable to the existing methods. CCNN obtains a better classification performance and considerably faster convergence speed than convolutional neural network. Experimental results also show that the proposed method is a useful decision-support tool for automatic sleep stage classification.

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.

A Data-Driven Response Virtual Sensor Technique with Partial Vibration Measurements Using Convolutional Neural Network.

PubMed

Sun, Shan-Bin; He, Yuan-Yuan; Zhou, Si-Da; Yue, Zhen-Jiang

2017-12-12

Measurement of dynamic responses plays an important role in structural health monitoring, damage detection and other fields of research. However, in aerospace engineering, the physical sensors are limited in the operational conditions of spacecraft, due to the severe environment in outer space. This paper proposes a virtual sensor model with partial vibration measurements using a convolutional neural network. The transmissibility function is employed as prior knowledge. A four-layer neural network with two convolutional layers, one fully connected layer, and an output layer is proposed as the predicting model. Numerical examples of two different structural dynamic systems demonstrate the performance of the proposed approach. The excellence of the novel technique is further indicated using a simply supported beam experiment comparing to a modal-model-based virtual sensor, which uses modal parameters, such as mode shapes, for estimating the responses of the faulty sensors. The results show that the presented data-driven response virtual sensor technique can predict structural response with high accuracy.

A Data-Driven Response Virtual Sensor Technique with Partial Vibration Measurements Using Convolutional Neural Network

PubMed Central

Sun, Shan-Bin; He, Yuan-Yuan; Zhou, Si-Da; Yue, Zhen-Jiang

2017-01-01

Measurement of dynamic responses plays an important role in structural health monitoring, damage detection and other fields of research. However, in aerospace engineering, the physical sensors are limited in the operational conditions of spacecraft, due to the severe environment in outer space. This paper proposes a virtual sensor model with partial vibration measurements using a convolutional neural network. The transmissibility function is employed as prior knowledge. A four-layer neural network with two convolutional layers, one fully connected layer, and an output layer is proposed as the predicting model. Numerical examples of two different structural dynamic systems demonstrate the performance of the proposed approach. The excellence of the novel technique is further indicated using a simply supported beam experiment comparing to a modal-model-based virtual sensor, which uses modal parameters, such as mode shapes, for estimating the responses of the faulty sensors. The results show that the presented data-driven response virtual sensor technique can predict structural response with high accuracy. PMID:29231868

Convoluted nozzle design for the RL10 derivative 2B engine

NASA Technical Reports Server (NTRS)

1985-01-01

The convoluted nozzle is a conventional refractory metal nozzle extension that is formed with a portion of the nozzle convoluted to show the extendible nozzle within the length of the rocket engine. The convoluted nozzle (CN) was deployed by a system of four gas driven actuators. For spacecraft applications the optimum CN may be self-deployed by internal pressure retained, during deployment, by a jettisonable exit closure. The convoluted nozzle is included in a study of extendible nozzles for the RL10 Engine Derivative 2B for use in an early orbit transfer vehicle (OTV). Four extendible nozzle configurations for the RL10-2B engine were evaluated. Three configurations of the two position nozzle were studied including a hydrogen dump cooled metal nozzle and radiation cooled nozzles of refractory metal and carbon/carbon composite construction respectively.

Study on pixel matching method of the multi-angle observation from airborne AMPR measurements

NASA Astrophysics Data System (ADS)

Hou, Weizhen; Qie, Lili; Li, Zhengqiang; Sun, Xiaobing; Hong, Jin; Chen, Xingfeng; Xu, Hua; Sun, Bin; Wang, Han

2015-10-01

For the along-track scanning mode, the same place along the ground track could be detected by the Advanced Multi-angular Polarized Radiometer (AMPR) with several different scanning angles from -55 to 55 degree, which provides a possible means to get the multi-angular detection for some nearby pixels. However, due to the ground sample spacing and spatial footprint of the detection, the different sizes of footprints cannot guarantee the spatial matching of some partly overlap pixels, which turn into a bottleneck for the effective use of the multi-angular detected information of AMPR to study the aerosol and surface polarized properties. Based on our definition and calculation of t he pixel coincidence rate for the multi-angular detection, an effective multi-angle observation's pixel matching method is presented to solve the spatial matching problem for airborne AMPR. Assuming the shape of AMPR's each pixel is an ellipse, and the major axis and minor axis depends on the flying attitude and each scanning angle. By the definition of coordinate system and origin of coordinate, the latitude and longitude could be transformed into the Euclidian distance, and the pixel coincidence rate of two nearby ellipses could be calculated. Via the traversal of each ground pixel, those pixels with high coincidence rate could be selected and merged, and with the further quality control of observation data, thus the ground pixels dataset with multi-angular detection could be obtained and analyzed, providing the support for the multi-angular and polarized retrieval algorithm research in t he next study.

Charge Loss and Charge Sharing Measurements for Two Different Pixelated Cadmium-Zinc-Telluride Detectors

NASA Technical Reports Server (NTRS)

Gaskin, Jessica; Sharma, Dharma; Ramsey, Brian; Seller, Paul

2003-01-01

As part of ongoing research at Marshall Space Flight Center, Cadmium-Zinc- Telluride (CdZnTe) pixilated detectors are being developed for use at the focal plane of the High Energy Replicated Optics (HERO) telescope. HERO requires a 64x64 pixel array with a spatial resolution of around 200 microns (with a 6m focal length) and high energy resolution (< 2% at 60keV). We are currently testing smaller arrays as a necessary first step towards this goal. In this presentation, we compare charge sharing and charge loss measurements between two devices that differ both electronically and geometrically. The first device consists of a 1-mm-thick piece of CdZnTe that is sputtered with a 4x4 array of pixels with pixel pitch of 750 microns (inter-pixel gap is 100 microns). The signal is read out using discrete ultra-low-noise preamplifiers, one for each of the 16 pixels. The second detector consists of a 2-mm-thick piece of CdZnTe that is sputtered with a 16x16 array of pixels with a pixel pitch of 300 microns (inter-pixel gap is 50 microns). Instead of using discrete preamplifiers, the crystal is bonded to an ASIC that provides all of the front-end electronics to each of the 256 pixels. what degree the bias voltage (i.e. the electric field) and hence the drift and diffusion coefficients affect our measurements. Further, we compare the measured results with simulated results and discuss to

Polarimetric analysis of a CdZnTe spectro-imager under multi-pixel irradiation conditions

NASA Astrophysics Data System (ADS)

Pinto, M.; da Silva, R. M. Curado; Maia, J. M.; Simões, N.; Marques, J.; Pereira, L.; Trindade, A. M. F.; Caroli, E.; Auricchio, N.; Stephen, J. B.; Gonçalves, P.

2016-12-01

So far, polarimetry in high-energy astrophysics has been insufficiently explored due to the complexity of the required detection, electronic and signal processing systems. However, its importance is today largely recognized by the astrophysical community, therefore the next generation of high-energy space instruments will certainly provide polarimetric observations, contemporaneously with spectroscopy and imaging. We have been participating in high-energy observatory proposals submitted to ESA Cosmic Vision calls, such as GRI (Gamma-Ray Imager), DUAL and ASTROGAM, where the main instrument was a spectro-imager with polarimetric capabilities. More recently, the H2020 AHEAD project was launched with the objective to promote more coherent and mature future high-energy space mission proposals. In this context of high-energy proposal development, we have tested a CdZnTe detection plane prototype polarimeter under a partially polarized gamma-ray beam generated from an aluminum target irradiated by a 22Na (511 keV) radioactive source. The polarized beam cross section was 1 cm2, allowing the irradiation of a wide multi-pixelated area where all the pixels operate simultaneously as a scatterer and as an absorber. The methods implemented to analyze such multi-pixel irradiation are similar to those required to analyze a spectro-imager polarimeter operating in space, since celestial source photons should irradiate its full pixilated area. Correction methods to mitigate systematic errors inherent to CdZnTe and to the experimental conditions were also implemented. The polarization level ( 40%) and the polarization angle (precision of ±5° up to ±9°) obtained under multi-pixel irradiation conditions are presented and compared with simulated data.

Super-Resolution for "Jilin-1" Satellite Video Imagery via a Convolutional Network.

PubMed

Xiao, Aoran; Wang, Zhongyuan; Wang, Lei; Ren, Yexian

2018-04-13

Super-resolution for satellite video attaches much significance to earth observation accuracy, and the special imaging and transmission conditions on the video satellite pose great challenges to this task. The existing deep convolutional neural-network-based methods require pre-processing or post-processing to be adapted to a high-resolution size or pixel format, leading to reduced performance and extra complexity. To this end, this paper proposes a five-layer end-to-end network structure without any pre-processing and post-processing, but imposes a reshape or deconvolution layer at the end of the network to retain the distribution of ground objects within the image. Meanwhile, we formulate a joint loss function by combining the output and high-dimensional features of a non-linear mapping network to precisely learn the desirable mapping relationship between low-resolution images and their high-resolution counterparts. Also, we use satellite video data itself as a training set, which favors consistency between training and testing images and promotes the method's practicality. Experimental results on "Jilin-1" satellite video imagery show that this method demonstrates a superior performance in terms of both visual effects and measure metrics over competing methods.

Sub-pixel Area Calculation Methods for Estimating Irrigated Areas.

PubMed

Thenkabailc, Prasad S; Biradar, Chandrashekar M; Noojipady, Praveen; Cai, Xueliang; Dheeravath, Venkateswarlu; Li, Yuanjie; Velpuri, Manohar; Gumma, Muralikrishna; Pandey, Suraj

2007-10-31

The goal of this paper was to develop and demonstrate practical methods forcomputing sub-pixel areas (SPAs) from coarse-resolution satellite sensor data. Themethods were tested and verified using: (a) global irrigated area map (GIAM) at 10-kmresolution based, primarily, on AVHRR data, and (b) irrigated area map for India at 500-mbased, primarily, on MODIS data. The sub-pixel irrigated areas (SPIAs) from coarse-resolution satellite sensor data were estimated by multiplying the full pixel irrigated areas(FPIAs) with irrigated area fractions (IAFs). Three methods were presented for IAFcomputation: (a) Google Earth Estimate (IAF-GEE); (b) High resolution imagery (IAF-HRI); and (c) Sub-pixel de-composition technique (IAF-SPDT). The IAF-GEE involvedthe use of "zoom-in-views" of sub-meter to 4-meter very high resolution imagery (VHRI)from Google Earth and helped determine total area available for irrigation (TAAI) or netirrigated areas that does not consider intensity or seasonality of irrigation. The IAF-HRI isa well known method that uses finer-resolution data to determine SPAs of the coarser-resolution imagery. The IAF-SPDT is a unique and innovative method wherein SPAs aredetermined based on the precise location of every pixel of a class in 2-dimensionalbrightness-greenness-wetness (BGW) feature-space plot of red band versus near-infraredband spectral reflectivity. The SPIAs computed using IAF-SPDT for the GIAM was within2 % of the SPIA computed using well known IAF-HRI. Further the fractions from the 2 methods were significantly correlated. The IAF-HRI and IAF-SPDT help to determine annualized or gross irrigated areas (AIA) that does consider intensity or seasonality (e.g., sum of areas from season 1, season 2, and continuous year-round crops). The national census based irrigated areas for the top 40 irrigated nations (which covers about 90% of global irrigation) was significantly better related (and had lesser uncertainties and errors) when compared to SPIAs than

Symmetric convolution of asymmetric multidimensional sequences using discrete trigonometric transforms.

PubMed

Foltz, T M; Welsh, B M

1999-01-01

This paper uses the fact that the discrete Fourier transform diagonalizes a circulant matrix to provide an alternate derivation of the symmetric convolution-multiplication property for discrete trigonometric transforms. Derived in this manner, the symmetric convolution-multiplication property extends easily to multiple dimensions using the notion of block circulant matrices and generalizes to multidimensional asymmetric sequences. The symmetric convolution of multidimensional asymmetric sequences can then be accomplished by taking the product of the trigonometric transforms of the sequences and then applying an inverse trigonometric transform to the result. An example is given of how this theory can be used for applying a two-dimensional (2-D) finite impulse response (FIR) filter with nonlinear phase which models atmospheric turbulence.

High density pixel array

NASA Technical Reports Server (NTRS)

McFall, James Earl (Inventor); Wiener-Avnear, Eliezer (Inventor)

2004-01-01

A pixel array device is fabricated by a laser micro-milling method under strict process control conditions. The device has an array of pixels bonded together with an adhesive filling the grooves between adjacent pixels. The array is fabricated by moving a substrate relative to a laser beam of predetermined intensity at a controlled, constant velocity along a predetermined path defining a set of grooves between adjacent pixels so that a predetermined laser flux per unit area is applied to the material, and repeating the movement for a plurality of passes of the laser beam until the grooves are ablated to a desired depth. The substrate is of an ultrasonic transducer material in one example for fabrication of a 2D ultrasonic phase array transducer. A substrate of phosphor material is used to fabricate an X-ray focal plane array detector.

Survey of on-road image projection with pixel light systems

NASA Astrophysics Data System (ADS)

Rizvi, Sadiq; Knöchelmann, Marvin; Ley, Peer-Phillip; Lachmayer, Roland

2017-12-01

HID, LED and laser-based high resolution automotive headlamps, as of late known as `pixel light systems', are at the forefront of the developing technologies paving the way for autonomous driving. In addition to light distribution capabilities that outperform Adaptive Front Lighting and Matrix Beam systems, pixel light systems provide the possibility of image projection directly onto the street. The underlying objective is to improve the driving experience, in any given scenario, in terms of safety, comfort and interaction for all road users. The focus of this work is to conduct a short survey on this state-of-the-art image projection functionality. A holistic research regarding the image projection functionality can be divided into three major categories: scenario selection, technological development and evaluation design. Consequently, the work presented in this paper is divided into three short studies. Section 1 provides a brief introduction to pixel light systems and a justification for the approach adopted for this study. Section 2 deals with the selection of scenarios (and driving maneuvers) where image projection can play a critical role. Section 3 discusses high power LED and LED array based prototypes that are currently under development. Section 4 demonstrates results from an experiment conducted to evaluate the illuminance of an image space projected using a pixel light system prototype developed at the Institute of Product Development (IPeG). Findings from this work can help to identify and advance future research work relating to: further development of pixel light systems, scenario planning, examination of optimal light sources, behavioral response studies etc.

High Performance Implementation of 3D Convolutional Neural Networks on a GPU.

PubMed

Lan, Qiang; Wang, Zelong; Wen, Mei; Zhang, Chunyuan; Wang, Yijie

2017-01-01

Convolutional neural networks have proven to be highly successful in applications such as image classification, object tracking, and many other tasks based on 2D inputs. Recently, researchers have started to apply convolutional neural networks to video classification, which constitutes a 3D input and requires far larger amounts of memory and much more computation. FFT based methods can reduce the amount of computation, but this generally comes at the cost of an increased memory requirement. On the other hand, the Winograd Minimal Filtering Algorithm (WMFA) can reduce the number of operations required and thus can speed up the computation, without increasing the required memory. This strategy was shown to be successful for 2D neural networks. We implement the algorithm for 3D convolutional neural networks and apply it to a popular 3D convolutional neural network which is used to classify videos and compare it to cuDNN. For our highly optimized implementation of the algorithm, we observe a twofold speedup for most of the 3D convolution layers of our test network compared to the cuDNN version.

High Performance Implementation of 3D Convolutional Neural Networks on a GPU

PubMed Central

Wang, Zelong; Wen, Mei; Zhang, Chunyuan; Wang, Yijie

2017-01-01

Convolutional neural networks have proven to be highly successful in applications such as image classification, object tracking, and many other tasks based on 2D inputs. Recently, researchers have started to apply convolutional neural networks to video classification, which constitutes a 3D input and requires far larger amounts of memory and much more computation. FFT based methods can reduce the amount of computation, but this generally comes at the cost of an increased memory requirement. On the other hand, the Winograd Minimal Filtering Algorithm (WMFA) can reduce the number of operations required and thus can speed up the computation, without increasing the required memory. This strategy was shown to be successful for 2D neural networks. We implement the algorithm for 3D convolutional neural networks and apply it to a popular 3D convolutional neural network which is used to classify videos and compare it to cuDNN. For our highly optimized implementation of the algorithm, we observe a twofold speedup for most of the 3D convolution layers of our test network compared to the cuDNN version. PMID:29250109

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.

Development of Kilo-Pixel Arrays of Transition-Edge Sensors for X-Ray Spectroscopy

NASA Technical Reports Server (NTRS)

Adams, J. S.; Bandler, S. R.; Busch, S. E.; Chervenak, J. A.; Chiao, M. P.; Eckart, M. E.; Ewin, A. J.; Finkbeiner, F. M.; Kelley, R. L.; Kelly, D. P.;

Baryon Acoustic Oscillations reconstruction with pixels

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

Obuljen, Andrej; Villaescusa-Navarro, Francisco; Castorina, Emanuele

2017-09-01

Gravitational non-linear evolution induces a shift in the position of the baryon acoustic oscillations (BAO) peak together with a damping and broadening of its shape that bias and degrades the accuracy with which the position of the peak can be determined. BAO reconstruction is a technique developed to undo part of the effect of non-linearities. We present and analyse a reconstruction method that consists of displacing pixels instead of galaxies and whose implementation is easier than the standard reconstruction method. We show that this method is equivalent to the standard reconstruction technique in the limit where the number of pixelsmore » becomes very large. This method is particularly useful in surveys where individual galaxies are not resolved, as in 21cm intensity mapping observations. We validate this method by reconstructing mock pixelated maps, that we build from the distribution of matter and halos in real- and redshift-space, from a large set of numerical simulations. We find that this method is able to decrease the uncertainty in the BAO peak position by 30-50% over the typical angular resolution scales of 21 cm intensity mapping experiments.« less

Superpixel-based and boundary-sensitive convolutional neural network for automated liver segmentation

NASA Astrophysics Data System (ADS)

Qin, Wenjian; Wu, Jia; Han, Fei; Yuan, Yixuan; Zhao, Wei; Ibragimov, Bulat; Gu, Jia; Xing, Lei

2018-05-01

Segmentation of liver in abdominal computed tomography (CT) is an important step for radiation therapy planning of hepatocellular carcinoma. Practically, a fully automatic segmentation of liver remains challenging because of low soft tissue contrast between liver and its surrounding organs, and its highly deformable shape. The purpose of this work is to develop a novel superpixel-based and boundary sensitive convolutional neural network (SBBS-CNN) pipeline for automated liver segmentation. The entire CT images were first partitioned into superpixel regions, where nearby pixels with similar CT number were aggregated. Secondly, we converted the conventional binary segmentation into a multinomial classification by labeling the superpixels into three classes: interior liver, liver boundary, and non-liver background. By doing this, the boundary region of the liver was explicitly identified and highlighted for the subsequent classification. Thirdly, we computed an entropy-based saliency map for each CT volume, and leveraged this map to guide the sampling of image patches over the superpixels. In this way, more patches were extracted from informative regions (e.g. the liver boundary with irregular changes) and fewer patches were extracted from homogeneous regions. Finally, deep CNN pipeline was built and trained to predict the probability map of the liver boundary. We tested the proposed algorithm in a cohort of 100 patients. With 10-fold cross validation, the SBBS-CNN achieved mean Dice similarity coefficients of 97.31  ±  0.36% and average symmetric surface distance of 1.77  ±  0.49 mm. Moreover, it showed superior performance in comparison with state-of-art methods, including U-Net, pixel-based CNN, active contour, level-sets and graph-cut algorithms. SBBS-CNN provides an accurate and effective tool for automated liver segmentation. It is also envisioned that the proposed framework is directly applicable in other medical image segmentation scenarios.

Reconfigurable Gabor Filter For Fingerprint Recognition Using FPGA Verilog

NASA Astrophysics Data System (ADS)

Rosshidi, H. T.; Hadi, A. R.

2009-06-01

This paper present the implementations of Gabor filter for fingerprint recognition using Verilog HDL. This work demonstrates the application of Gabor Filter technique to enhance the fingerprint image. The incoming signal in form of image pixel will be filter out or convolute by the Gabor filter to define the ridge and valley regions of fingerprint. This is done with the application of a real time convolve based on Field Programmable Gate Array (FPGA) to perform the convolution operation. The main characteristic of the proposed approach are the usage of memory to store the incoming image pixel and the coefficient of the Gabor filter before the convolution matrix take place. The result was the signal convoluted with the Gabor coefficient.

Single Image Super-Resolution Based on Multi-Scale Competitive Convolutional Neural Network

PubMed Central

Qu, Xiaobo; He, Yifan

2018-01-01

Deep convolutional neural networks (CNNs) are successful in single-image super-resolution. Traditional CNNs are limited to exploit multi-scale contextual information for image reconstruction due to the fixed convolutional kernel in their building modules. To restore various scales of image details, we enhance the multi-scale inference capability of CNNs by introducing competition among multi-scale convolutional filters, and build up a shallow network under limited computational resources. The proposed network has the following two advantages: (1) the multi-scale convolutional kernel provides the multi-context for image super-resolution, and (2) the maximum competitive strategy adaptively chooses the optimal scale of information for image reconstruction. Our experimental results on image super-resolution show that the performance of the proposed network outperforms the state-of-the-art methods. PMID:29509666

Single Image Super-Resolution Based on Multi-Scale Competitive Convolutional Neural Network.

PubMed

Du, Xiaofeng; Qu, Xiaobo; He, Yifan; Guo, Di

2018-03-06

Deep convolutional neural networks (CNNs) are successful in single-image super-resolution. Traditional CNNs are limited to exploit multi-scale contextual information for image reconstruction due to the fixed convolutional kernel in their building modules. To restore various scales of image details, we enhance the multi-scale inference capability of CNNs by introducing competition among multi-scale convolutional filters, and build up a shallow network under limited computational resources. The proposed network has the following two advantages: (1) the multi-scale convolutional kernel provides the multi-context for image super-resolution, and (2) the maximum competitive strategy adaptively chooses the optimal scale of information for image reconstruction. Our experimental results on image super-resolution show that the performance of the proposed network outperforms the state-of-the-art methods.

Spatial clustering of pixels of a multispectral image

DOEpatents

Conger, James Lynn

2014-08-19

A method and system for clustering the pixels of a multispectral image is provided. A clustering system computes a maximum spectral similarity score for each pixel that indicates the similarity between that pixel and the most similar neighboring. To determine the maximum similarity score for a pixel, the clustering system generates a similarity score between that pixel and each of its neighboring pixels and then selects the similarity score that represents the highest similarity as the maximum similarity score. The clustering system may apply a filtering criterion based on the maximum similarity score so that pixels with similarity scores below a minimum threshold are not clustered. The clustering system changes the current pixel values of the pixels in a cluster based on an averaging of the original pixel values of the pixels in the cluster.

On-Orbit Solar Dynamics Observatory (SDO) Star Tracker Warm Pixel Analysis

NASA Technical Reports Server (NTRS)

Felikson, Denis; Ekinci, Matthew; Hashmall, Joseph A.; Vess, Melissa

2011-01-01

This paper describes the process of identification and analysis of warm pixels in two autonomous star trackers on the Solar Dynamics Observatory (SDO) mission. A brief description of the mission orbit and attitude regimes is discussed and pertinent star tracker hardware specifications are given. Warm pixels are defined and the Quality Index parameter is introduced, which can be explained qualitatively as a manifestation of a possible warm pixel event. A description of the algorithm used to identify warm pixel candidates is given. Finally, analysis of dumps of on-orbit star tracker charge coupled devices (CCD) images is presented and an operational plan going forward is discussed. SDO, launched on February 11, 2010, is operated from the NASA Goddard Space Flight Center (GSFC). SDO is in a geosynchronous orbit with a 28.5 inclination. The nominal mission attitude points the spacecraft X-axis at the Sun, with the spacecraft Z-axis roughly aligned with the Solar North Pole. The spacecraft Y-axis completes the triad. In attitude, SDO moves approximately 0.04 per hour, mostly about the spacecraft Z-axis. The SDO star trackers, manufactured by Galileo Avionica, project the images of stars in their 16.4deg x 16.4deg fields-of-view onto CCD detectors consisting of 512 x 512 pixels. The trackers autonomously identify the star patterns and provide an attitude estimate. Each unit is able to track up to 9 stars. Additionally, each tracker calculates a parameter called the Quality Index, which is a measure of the quality of the attitude solution. Each pixel in the CCD measures the intensity of light and a warns pixel is defined as having a measurement consistently and significantly higher than the mean background intensity level. A warns pixel should also have lower intensity than a pixel containing a star image and will not move across the field of view as the attitude changes (as would a dim star image). It should be noted that the maximum error introduced in the star tracker

Iterative deep convolutional encoder-decoder network for medical image segmentation.

PubMed

Jung Uk Kim; Hak Gu Kim; Yong Man Ro

2017-07-01

In this paper, we propose a novel medical image segmentation using iterative deep learning framework. We have combined an iterative learning approach and an encoder-decoder network to improve segmentation results, which enables to precisely localize the regions of interest (ROIs) including complex shapes or detailed textures of medical images in an iterative manner. The proposed iterative deep convolutional encoder-decoder network consists of two main paths: convolutional encoder path and convolutional decoder path with iterative learning. Experimental results show that the proposed iterative deep learning framework is able to yield excellent medical image segmentation performances for various medical images. The effectiveness of the proposed method has been proved by comparing with other state-of-the-art medical image segmentation methods.

Design and characterization of high precision in-pixel discriminators for rolling shutter CMOS pixel sensors with full CMOS capability

NASA Astrophysics Data System (ADS)

Fu, Y.; Hu-Guo, C.; Dorokhov, A.; Pham, H.; Hu, Y.

2013-07-01

In order to exploit the ability to integrate a charge collecting electrode with analog and digital processing circuitry down to the pixel level, a new type of CMOS pixel sensors with full CMOS capability is presented in this paper. The pixel array is read out based on a column-parallel read-out architecture, where each pixel incorporates a diode, a preamplifier with a double sampling circuitry and a discriminator to completely eliminate analog read-out bottlenecks. The sensor featuring a pixel array of 8 rows and 32 columns with a pixel pitch of 80 μm×16 μm was fabricated in a 0.18 μm CMOS process. The behavior of each pixel-level discriminator isolated from the diode and the preamplifier was studied. The experimental results indicate that all in-pixel discriminators which are fully operational can provide significant improvements in the read-out speed and the power consumption of CMOS pixel sensors.

Optimized color decomposition of localized whole slide images and convolutional neural network for intermediate prostate cancer classification

NASA Astrophysics Data System (ADS)

Zhou, Naiyun; Gao, Yi

2017-03-01

This paper presents a fully automatic approach to grade intermediate prostate malignancy with hematoxylin and eosin-stained whole slide images. Deep learning architectures such as convolutional neural networks have been utilized in the domain of histopathology for automated carcinoma detection and classification. However, few work show its power in discriminating intermediate Gleason patterns, due to sporadic distribution of prostate glands on stained surgical section samples. We propose optimized hematoxylin decomposition on localized images, followed by convolutional neural network to classify Gleason patterns 3+4 and 4+3 without handcrafted features or gland segmentation. Crucial glands morphology and structural relationship of nuclei are extracted twice in different color space by the multi-scale strategy to mimic pathologists' visual examination. Our novel classification scheme evaluated on 169 whole slide images yielded a 70.41% accuracy and corresponding area under the receiver operating characteristic curve of 0.7247.

Development of a Cost-Effective Modular Pixelated NaI(Tl) Detector for Clinical SPECT Applications

PubMed Central

Rozler, Mike; Liang, Haoning; Chang, Wei

2013-01-01

A new pixelated detector for high-resolution clinical SPECT applications was designed and tested. The modular detector is based on a scintillator block comprised of 2.75×2.75×10 mm3 NaI(Tl) pixels and decoded by an array of 51 mm diameter single-anode PMTs. Several configurations, utilizing two types of PMTs, were evaluated using a collimated beam source to measure positioning accuracy directly. Good pixel separation was observed, with correct pixel identification ranging from 60 to 72% averaged over the entire area of the modules, depending on the PMT type and configuration. This translates to a significant improvement in positioning accuracy compared to continuous slab detectors of the same thickness, along with effective reduction of “dead” space at the edges. The observed 10% average energy resolution compares well to continuous slab detectors. The combined performance demonstrates the suitability of pixelated detectors decoded with a relatively small number of medium-sized PMTs as a cost-effective approach for high resolution clinical SPECT applications, in particular those involving curved detector geometries. PMID:24146436

Convolutional virtual electric field for image segmentation using active contours.

PubMed

Wang, Yuanquan; Zhu, Ce; Zhang, Jiawan; Jian, Yuden

2014-01-01

Gradient vector flow (GVF) is an effective external force for active contours; however, it suffers from heavy computation load. The virtual electric field (VEF) model, which can be implemented in real time using fast Fourier transform (FFT), has been proposed later as a remedy for the GVF model. In this work, we present an extension of the VEF model, which is referred to as CONvolutional Virtual Electric Field, CONVEF for short. This proposed CONVEF model takes the VEF model as a convolution operation and employs a modified distance in the convolution kernel. The CONVEF model is also closely related to the vector field convolution (VFC) model. Compared with the GVF, VEF and VFC models, the CONVEF model possesses not only some desirable properties of these models, such as enlarged capture range, u-shape concavity convergence, subject contour convergence and initialization insensitivity, but also some other interesting properties such as G-shape concavity convergence, neighboring objects separation, and noise suppression and simultaneously weak edge preserving. Meanwhile, the CONVEF model can also be implemented in real-time by using FFT. Experimental results illustrate these advantages of the CONVEF model on both synthetic and natural images.

Applying Gradient Descent in Convolutional Neural Networks

NASA Astrophysics Data System (ADS)

Cui, Nan

2018-04-01

With the development of the integrated circuit and computer science, people become caring more about solving practical issues via information technologies. Along with that, a new subject called Artificial Intelligent (AI) comes up. One popular research interest of AI is about recognition algorithm. In this paper, one of the most common algorithms, Convolutional Neural Networks (CNNs) will be introduced, for image recognition. Understanding its theory and structure is of great significance for every scholar who is interested in this field. Convolution Neural Network is an artificial neural network which combines the mathematical method of convolution and neural network. The hieratical structure of CNN provides it reliable computer speed and reasonable error rate. The most significant characteristics of CNNs are feature extraction, weight sharing and dimension reduction. Meanwhile, combining with the Back Propagation (BP) mechanism and the Gradient Descent (GD) method, CNNs has the ability to self-study and in-depth learning. Basically, BP provides an opportunity for backwardfeedback for enhancing reliability and GD is used for self-training process. This paper mainly discusses the CNN and the related BP and GD algorithms, including the basic structure and function of CNN, details of each layer, the principles and features of BP and GD, and some examples in practice with a summary in the end.

Filling schemes of silver dots inkjet-printed on pixelated nanostructured surfaces

NASA Astrophysics Data System (ADS)

Alan, Sheida; Jiang, Hao; Shahbazbegian, Haleh; Patel, Jasbir N.; Kaminska, Bozena

2017-03-01

Recently, our group demonstrated an inkjet-based technique to enable high-throughput, versatile and full-colour printing of structural colours on generic pixelated nanostructures, termed as molded ink on nanostructured surfaces. The printed colours are controlled by the area of printed silver on the pixelated red, green and blue polymer nanostructure arrays. This paper investigates the behaviour of jetted silver ink droplets on nanostructured surfaces and the microscale dot patterns implemented during printing process, for achieving accurate and consistent colours in the printed images. The surface wettability and the schemes of filling silver dots inside the subpixels are crucial to the quality of printed images. Several related concepts and definitions are introduced, such as filling ratio, full dots per subpixel (DPSP), number of printable colours, colour leaking and dot merging. In our experiments, we first chemically modified the surface to control the wettability and dot size. From each type of modified surface, various filling schemes were experimented and the printed results were evaluated with comprehensive considerations on the number of printable colours and the negative effects of colour leaking and dot merging. Rational selection of the best filling scheme resulted in a 2-line filling scheme using 20 μm dot spacing and line spacing capable of printing 9261 different colours with 121 pixel per inch display resolution, on low-wettability surface. This study is of vital importance for scaling up the printing technique in industrial applications and provides meaningful insights for inkjet-printing on nanostructures.

How Many Pixels Does It Take to Make a Good 4"×6" Print? Pixel Count Wars Revisited

NASA Astrophysics Data System (ADS)

Kriss, Michael A.

Digital still cameras emerged following the introduction of the Sony Mavica analog prototype camera in 1981. These early cameras produced poor image quality and did not challenge film cameras for overall quality. By 1995 digital still cameras in expensive SLR formats had 6 mega-pixels and produced high quality images (with significant image processing). In 2005 significant improvement in image quality was apparent and lower prices for digital still cameras (DSCs) started a rapid decline in film usage and film camera sells. By 2010 film usage was mostly limited to professionals and the motion picture industry. The rise of DSCs was marked by a “pixel war” where the driving feature of the cameras was the pixel count where even moderate cost, ˜120, DSCs would have 14 mega-pixels. The improvement of CMOS technology pushed this trend of lower prices and higher pixel counts. Only the single lens reflex cameras had large sensors and large pixels. The drive for smaller pixels hurt the quality aspects of the final image (sharpness, noise, speed, and exposure latitude). Only today are camera manufactures starting to reverse their course and producing DSCs with larger sensors and pixels. This paper will explore why larger pixels and sensors are key to the future of DSCs.

Iterative algorithm for reconstructing rotationally asymmetric surface deviation with pixel-level spatial resolution

NASA Astrophysics Data System (ADS)

Quan, Haiyang; Wu, Fan; Hou, Xi

2015-10-01

New method for reconstructing rotationally asymmetric surface deviation with pixel-level spatial resolution is proposed. It is based on basic iterative scheme and accelerates the Gauss-Seidel method by introducing an acceleration parameter. This modified Successive Over-relaxation (SOR) is effective for solving the rotationally asymmetric components with pixel-level spatial resolution, without the usage of a fitting procedure. Compared to the Jacobi and Gauss-Seidel method, the modified SOR method with an optimal relaxation factor converges much faster and saves more computational costs and memory space without reducing accuracy. It has been proved by real experimental results.

Efficient airport detection using region-based fully convolutional neural networks

NASA Astrophysics Data System (ADS)

Xin, Peng; Xu, Yuelei; Zhang, Xulei; Ma, Shiping; Li, Shuai; Lv, Chao

2018-04-01

This paper presents a model for airport detection using region-based fully convolutional neural networks. To achieve fast detection with high accuracy, we shared the conv layers between the region proposal procedure and the airport detection procedure and used graphics processing units (GPUs) to speed up the training and testing time. For lack of labeled data, we transferred the convolutional layers of ZF net pretrained by ImageNet to initialize the shared convolutional layers, then we retrained the model using the alternating optimization training strategy. The proposed model has been tested on an airport dataset consisting of 600 images. Experiments show that the proposed method can distinguish airports in our dataset from similar background scenes almost real-time with high accuracy, which is much better than traditional methods.

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.

1981-01-01

Plotted transects made from south Texas daytime HCMM data show the effect of subvisible cirrus (SCI) clouds in the emissive (IR) band but the effect is unnoticable in the reflective (VIS) band. The depression of satellite indicated temperatures ws greatest in the center of SCi streamers and tapered off at the edges. Pixels of uncontaminated land and water features in the HCMM test area shared identical VIS and IR digital count combinations with other pixels representing similar features. A minimum of 0.015 percent repeats of identical VIS-IR combinations are characteristic of land and water features in a scene of 30 percent cloud cover. This increases to 0.021 percent of more when the scene is clear. Pixels having shared VIS-IR combinations less than these amounts are considered to be cloud contaminated in the cluster screening method. About twenty percent of SCi was machine indistinguishable from land features in two dimensional spectral space (VIS vs IR).

Generalization error analysis: deep convolutional neural network in mammography

NASA Astrophysics Data System (ADS)

Richter, Caleb D.; Samala, Ravi K.; Chan, Heang-Ping; Hadjiiski, Lubomir; Cha, Kenny

2018-02-01

We conducted a study to gain understanding of the generalizability of deep convolutional neural networks (DCNNs) given their inherent capability to memorize data. We examined empirically a specific DCNN trained for classification of masses on mammograms. Using a data set of 2,454 lesions from 2,242 mammographic views, a DCNN was trained to classify masses into malignant and benign classes using transfer learning from ImageNet LSVRC-2010. We performed experiments with varying amounts of label corruption and types of pixel randomization to analyze the generalization error for the DCNN. Performance was evaluated using the area under the receiver operating characteristic curve (AUC) with an N-fold cross validation. Comparisons were made between the convergence times, the inference AUCs for both the training set and the test set of the original image patches without corruption, and the root-mean-squared difference (RMSD) in the layer weights of the DCNN trained with different amounts and methods of corruption. Our experiments observed trends which revealed that the DCNN overfitted by memorizing corrupted data. More importantly, this study improved our understanding of DCNN weight updates when learning new patterns or new labels. Although we used a specific classification task with the ImageNet as example, similar methods may be useful for analysis of the DCNN learning processes, especially those that employ transfer learning for medical image analysis where sample size is limited and overfitting risk is high.

Experimental study of current loss and plasma formation in the Z machine post-hole convolute

NASA Astrophysics Data System (ADS)

Gomez, M. R.; Gilgenbach, R. M.; Cuneo, M. E.; Jennings, C. A.; McBride, R. D.; Waisman, E. M.; Hutsel, B. T.; Stygar, W. A.; Rose, D. V.; Maron, Y.

2017-01-01

The Z pulsed-power generator at Sandia National Laboratories drives high energy density physics experiments with load currents of up to 26 MA. Z utilizes a double post-hole convolute to combine the current from four parallel magnetically insulated transmission lines into a single transmission line just upstream of the load. Current loss is observed in most experiments and is traditionally attributed to inefficient convolute performance. The apparent loss current varies substantially for z-pinch loads with different inductance histories; however, a similar convolute impedance history is observed for all load types. This paper details direct spectroscopic measurements of plasma density, temperature, and apparent and actual plasma closure velocities within the convolute. Spectral measurements indicate a correlation between impedance collapse and plasma formation in the convolute. Absorption features in the spectra show the convolute plasma consists primarily of hydrogen, which likely forms from desorbed electrode contaminant species such as H2O , H2 , and hydrocarbons. Plasma densities increase from 1 ×1016 cm-3 (level of detectability) just before peak current to over 1 ×1017 cm-3 at stagnation (tens of ns later). The density seems to be highest near the cathode surface, with an apparent cathode to anode plasma velocity in the range of 35 - 50 cm /μ s . Similar plasma conditions and convolute impedance histories are observed in experiments with high and low losses, suggesting that losses are driven largely by load dynamics, which determine the voltage on the convolute.

The use of interleaving for reducing radio loss in convolutionally coded systems

NASA Technical Reports Server (NTRS)

Divsalar, D.; Simon, M. K.; Yuen, J. H.

1989-01-01

The use of interleaving after convolutional coding and deinterleaving before Viterbi decoding is proposed. This effectively reduces radio loss at low-loop Signal to Noise Ratios (SNRs) by several decibels and at high-loop SNRs by a few tenths of a decibel. Performance of the coded system can further be enhanced if the modulation index is optimized for this system. This will correspond to a reduction of bit SNR at a certain bit error rate for the overall system. The introduction of interleaving/deinterleaving into communication systems designed for future deep space missions does not substantially complicate their hardware design or increase their system cost.

Three-dimensional cascaded system analysis of a 50 µm pixel pitch wafer-scale CMOS active pixel sensor x-ray detector for digital breast tomosynthesis.

PubMed

Zhao, C; Vassiljev, N; Konstantinidis, A C; Speller, R D; Kanicki, J

2017-03-07

High-resolution, low-noise x-ray detectors based on the complementary metal-oxide-semiconductor (CMOS) active pixel sensor (APS) technology have been developed and proposed for digital breast tomosynthesis (DBT). In this study, we evaluated the three-dimensional (3D) imaging performance of a 50 µm pixel pitch CMOS APS x-ray detector named DynAMITe (Dynamic Range Adjustable for Medical Imaging Technology). The two-dimensional (2D) angle-dependent modulation transfer function (MTF), normalized noise power spectrum (NNPS), and detective quantum efficiency (DQE) were experimentally characterized and modeled using the cascaded system analysis at oblique incident angles up to 30°. The cascaded system model was extended to the 3D spatial frequency space in combination with the filtered back-projection (FBP) reconstruction method to calculate the 3D and in-plane MTF, NNPS and DQE parameters. The results demonstrate that the beam obliquity blurs the 2D MTF and DQE in the high spatial frequency range. However, this effect can be eliminated after FBP image reconstruction. In addition, impacts of the image acquisition geometry and detector parameters were evaluated using the 3D cascaded system analysis for DBT. The result shows that a wider projection angle range (e.g.  ±30°) improves the low spatial frequency (below 5 mm -1 ) performance of the CMOS APS detector. In addition, to maintain a high spatial resolution for DBT, a focal spot size of smaller than 0.3 mm should be used. Theoretical analysis suggests that a pixelated scintillator in combination with the 50 µm pixel pitch CMOS APS detector could further improve the 3D image resolution. Finally, the 3D imaging performance of the CMOS APS and an indirect amorphous silicon (a-Si:H) thin-film transistor (TFT) passive pixel sensor (PPS) detector was simulated and compared.

Three-dimensional cascaded system analysis of a 50 µm pixel pitch wafer-scale CMOS active pixel sensor x-ray detector for digital breast tomosynthesis

NASA Astrophysics Data System (ADS)

Zhao, C.; Vassiljev, N.; Konstantinidis, A. C.; Speller, R. D.; Kanicki, J.

2017-03-01

High-resolution, low-noise x-ray detectors based on the complementary metal-oxide-semiconductor (CMOS) active pixel sensor (APS) technology have been developed and proposed for digital breast tomosynthesis (DBT). In this study, we evaluated the three-dimensional (3D) imaging performance of a 50 µm pixel pitch CMOS APS x-ray detector named DynAMITe (Dynamic Range Adjustable for Medical Imaging Technology). The two-dimensional (2D) angle-dependent modulation transfer function (MTF), normalized noise power spectrum (NNPS), and detective quantum efficiency (DQE) were experimentally characterized and modeled using the cascaded system analysis at oblique incident angles up to 30°. The cascaded system model was extended to the 3D spatial frequency space in combination with the filtered back-projection (FBP) reconstruction method to calculate the 3D and in-plane MTF, NNPS and DQE parameters. The results demonstrate that the beam obliquity blurs the 2D MTF and DQE in the high spatial frequency range. However, this effect can be eliminated after FBP image reconstruction. In addition, impacts of the image acquisition geometry and detector parameters were evaluated using the 3D cascaded system analysis for DBT. The result shows that a wider projection angle range (e.g.  ±30°) improves the low spatial frequency (below 5 mm-1) performance of the CMOS APS detector. In addition, to maintain a high spatial resolution for DBT, a focal spot size of smaller than 0.3 mm should be used. Theoretical analysis suggests that a pixelated scintillator in combination with the 50 µm pixel pitch CMOS APS detector could further improve the 3D image resolution. Finally, the 3D imaging performance of the CMOS APS and an indirect amorphous silicon (a-Si:H) thin-film transistor (TFT) passive pixel sensor (PPS) detector was simulated and compared.

Deep convolutional neural network for mammographic density segmentation

NASA Astrophysics Data System (ADS)

Wei, Jun; Li, Songfeng; Chan, Heang-Ping; Helvie, Mark A.; Roubidoux, Marilyn A.; Lu, Yao; Zhou, Chuan; Hadjiiski, Lubomir; Samala, Ravi K.

2018-02-01

Breast density is one of the most significant factors for cancer risk. In this study, we proposed a supervised deep learning approach for automated estimation of percentage density (PD) on digital mammography (DM). The deep convolutional neural network (DCNN) was trained to estimate a probability map of breast density (PMD). PD was calculated as the ratio of the dense area to the breast area based on the probability of each pixel belonging to dense region or fatty region at a decision threshold of 0.5. The DCNN estimate was compared to a feature-based statistical learning approach, in which gray level, texture and morphological features were extracted from each ROI and the least absolute shrinkage and selection operator (LASSO) was used to select and combine the useful features to generate the PMD. The reference PD of each image was provided by two experienced MQSA radiologists. With IRB approval, we retrospectively collected 347 DMs from patient files at our institution. The 10-fold cross-validation results showed a strong correlation r=0.96 between the DCNN estimation and interactive segmentation by radiologists while that of the feature-based statistical learning approach vs radiologists' segmentation had a correlation r=0.78. The difference between the segmentation by DCNN and by radiologists was significantly smaller than that between the feature-based learning approach and radiologists (p < 0.0001) by two-tailed paired t-test. This study demonstrated that the DCNN approach has the potential to replace radiologists' interactive thresholding in PD estimation on DMs.

Minimal-memory realization of pearl-necklace encoders of general quantum convolutional codes

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

Houshmand, Monireh; Hosseini-Khayat, Saied

2011-02-15

Quantum convolutional codes, like their classical counterparts, promise to offer higher error correction performance than block codes of equivalent encoding complexity, and are expected to find important applications in reliable quantum communication where a continuous stream of qubits is transmitted. Grassl and Roetteler devised an algorithm to encode a quantum convolutional code with a ''pearl-necklace'' encoder. Despite their algorithm's theoretical significance as a neat way of representing quantum convolutional codes, it is not well suited to practical realization. In fact, there is no straightforward way to implement any given pearl-necklace structure. This paper closes the gap between theoretical representation andmore » practical implementation. In our previous work, we presented an efficient algorithm to find a minimal-memory realization of a pearl-necklace encoder for Calderbank-Shor-Steane (CSS) convolutional codes. This work is an extension of our previous work and presents an algorithm for turning a pearl-necklace encoder for a general (non-CSS) quantum convolutional code into a realizable quantum convolutional encoder. We show that a minimal-memory realization depends on the commutativity relations between the gate strings in the pearl-necklace encoder. We find a realization by means of a weighted graph which details the noncommutative paths through the pearl necklace. The weight of the longest path in this graph is equal to the minimal amount of memory needed to implement the encoder. The algorithm has a polynomial-time complexity in the number of gate strings in the pearl-necklace encoder.« less

Two-dimensional convolute integers for analytical instrumentation

NASA Technical Reports Server (NTRS)

Edwards, T. R.

1982-01-01

As new analytical instruments and techniques emerge with increased dimensionality, a corresponding need is seen for data processing logic which can appropriately address the data. Two-dimensional measurements reveal enhanced unknown mixture analysis capability as a result of the greater spectral information content over two one-dimensional methods taken separately. It is noted that two-dimensional convolute integers are merely an extension of the work by Savitzky and Golay (1964). It is shown that these low-pass, high-pass and band-pass digital filters are truly two-dimensional and that they can be applied in a manner identical with their one-dimensional counterpart, that is, a weighted nearest-neighbor, moving average with zero phase shifting, convoluted integer (universal number) weighting coefficients.

A convolutional neural network neutrino event classifier

DOE PAGES

Aurisano, A.; Radovic, A.; Rocco, D.; ...

2016-09-01

Here, convolutional neural networks (CNNs) have been widely applied in the computer vision community to solve complex problems in image recognition and analysis. We describe an application of the CNN technology to the problem of identifying particle interactions in sampling calorimeters used commonly in high energy physics and high energy neutrino physics in particular. Following a discussion of the core concepts of CNNs and recent innovations in CNN architectures related to the field of deep learning, we outline a specific application to the NOvA neutrino detector. This algorithm, CVN (Convolutional Visual Network) identifies neutrino interactions based on their topology withoutmore » the need for detailed reconstruction and outperforms algorithms currently in use by the NOvA collaboration.« less

Gas Classification Using Deep Convolutional Neural Networks.

PubMed

Peng, Pai; Zhao, Xiaojin; Pan, Xiaofang; Ye, Wenbin

2018-01-08

In this work, we propose a novel Deep Convolutional Neural Network (DCNN) tailored for gas classification. Inspired by the great success of DCNN in the field of computer vision, we designed a DCNN with up to 38 layers. In general, the proposed gas neural network, named GasNet, consists of: six convolutional blocks, each block consist of six layers; a pooling layer; and a fully-connected layer. Together, these various layers make up a powerful deep model for gas classification. Experimental results show that the proposed DCNN method is an effective technique for classifying electronic nose data. We also demonstrate that the DCNN method can provide higher classification accuracy than comparable Support Vector Machine (SVM) methods and Multiple Layer Perceptron (MLP).

Gas Classification Using Deep Convolutional Neural Networks

PubMed Central

Peng, Pai; Zhao, Xiaojin; Pan, Xiaofang; Ye, Wenbin

2018-01-01

In this work, we propose a novel Deep Convolutional Neural Network (DCNN) tailored for gas classification. Inspired by the great success of DCNN in the field of computer vision, we designed a DCNN with up to 38 layers. In general, the proposed gas neural network, named GasNet, consists of: six convolutional blocks, each block consist of six layers; a pooling layer; and a fully-connected layer. Together, these various layers make up a powerful deep model for gas classification. Experimental results show that the proposed DCNN method is an effective technique for classifying electronic nose data. We also demonstrate that the DCNN method can provide higher classification accuracy than comparable Support Vector Machine (SVM) methods and Multiple Layer Perceptron (MLP). PMID:29316723

A convolutional neural network neutrino event classifier

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

Aurisano, A.; Radovic, A.; Rocco, D.

Here, convolutional neural networks (CNNs) have been widely applied in the computer vision community to solve complex problems in image recognition and analysis. We describe an application of the CNN technology to the problem of identifying particle interactions in sampling calorimeters used commonly in high energy physics and high energy neutrino physics in particular. Following a discussion of the core concepts of CNNs and recent innovations in CNN architectures related to the field of deep learning, we outline a specific application to the NOvA neutrino detector. This algorithm, CVN (Convolutional Visual Network) identifies neutrino interactions based on their topology withoutmore » the need for detailed reconstruction and outperforms algorithms currently in use by the NOvA collaboration.« less

Pixels, Blocks of Pixels, and Polygons: Choosing a Spatial Unit for Thematic Accuracy Assessment

EPA Science Inventory

Pixels, polygons, and blocks of pixels are all potentially viable spatial assessment units for conducting an accuracy assessment. We develop a statistical population-based framework to examine how the spatial unit chosen affects the outcome of an accuracy assessment. The populati...

Per-Pixel, Dual-Counter Scheme for Optical Communications

NASA Technical Reports Server (NTRS)

Farr, William H.; Bimbaum, Kevin M.; Quirk, Kevin J.; Sburlan, Suzana; Sahasrabudhe, Adit

2013-01-01

Free space optical communications links from deep space are projected to fulfill future NASA communication requirements for 2020 and beyond. Accurate laser-beam pointing is required to achieve high data rates at low power levels.This innovation is a per-pixel processing scheme using a pair of three-state digital counters to implement acquisition and tracking of a dim laser beacon transmitted from Earth for pointing control of an interplanetary optical communications system using a focal plane array of single sensitive detectors. It shows how to implement dim beacon acquisition and tracking for an interplanetary optical transceiver with a method that is suitable for both achieving theoretical performance, as well as supporting additional functions of high data rate forward links and precision spacecraft ranging.

Linear diffusion-wave channel routing using a discrete Hayami convolution method

Treesearch

Li Wang; Joan Q. Wu; William J. Elliot; Fritz R. Feidler; Sergey Lapin

2014-01-01

The convolution of an input with a response function has been widely used in hydrology as a means to solve various problems analytically. Due to the high computation demand in solving the functions using numerical integration, it is often advantageous to use the discrete convolution instead of the integration of the continuous functions. This approach greatly reduces...

Charge collection properties in an irradiated pixel sensor built in a thick-film HV-SOI process

NASA Astrophysics Data System (ADS)

Hiti, B.; Cindro, V.; Gorišek, A.; Hemperek, T.; Kishishita, T.; Kramberger, G.; Krüger, H.; Mandić, I.; Mikuž, M.; Wermes, N.; Zavrtanik, M.

2017-10-01

Investigation of HV-CMOS sensors for use as a tracking detector in the ATLAS experiment at the upgraded LHC (HL-LHC) has recently been an active field of research. A potential candidate for a pixel detector built in Silicon-On-Insulator (SOI) technology has already been characterized in terms of radiation hardness to TID (Total Ionizing Dose) and charge collection after a moderate neutron irradiation. In this article we present results of an extensive irradiation hardness study with neutrons up to a fluence of 1× 1016 neq/cm2. Charge collection in a passive pixelated structure was measured by Edge Transient Current Technique (E-TCT). The evolution of the effective space charge concentration was found to be compliant with the acceptor removal model, with the minimum of the space charge concentration being reached after 5× 1014 neq/cm2. An investigation of the in-pixel uniformity of the detector response revealed parasitic charge collection by the epitaxial silicon layer characteristic for the SOI design. The results were backed by a numerical simulation of charge collection in an equivalent detector layout.

Gossip: Gaseous pixels

NASA Astrophysics Data System (ADS)

Koffeman, E. N.

2007-12-01

Several years ago a revolutionary miniature TPC was developed using a pixel chip with a Micromegas foil spanned over it. To overcome the mechanical stability problems and improve the positioning accuracy while spanning a foil on top of a small readout chip a process has been developed in which a Micromegas-like grid is applied on a CMOS wafer in a post-processing step. This aluminum grid is supported on insulating pillars that are created by etching after the grid has been made. The energy resolution (measured on the absorption of the X-rays from a 55Fe source) was remarkably good. Several geometries have since been tested and we now believe that a Gas On Slimmed Silicon Pixel chip' (Gossip) may be realized. The drift region of such a gaseous pixel detector would be reduced to a millimeter. Such a detector is potentially very radiation hard (SLHC vertexing) but aging and sparking must be eliminated.

Characterizing pixel and point patterns with a hyperuniformity disorder length

NASA Astrophysics Data System (ADS)

Chieco, A. T.; Dreyfus, R.; Durian, D. J.

2017-09-01

We introduce the concept of a "hyperuniformity disorder length" h that controls the variance of volume fraction fluctuations for randomly placed windows of fixed size. In particular, fluctuations are determined by the average number of particles within a distance h from the boundary of the window. We first compute special expectations and bounds in d dimensions, and then illustrate the range of behavior of h versus window size L by analyzing several different types of simulated two-dimensional pixel patterns—where particle positions are stored as a binary digital image in which pixels have value zero if empty and one if they contain a particle. The first are random binomial patterns, where pixels are randomly flipped from zero to one with probability equal to area fraction. These have long-ranged density fluctuations, and simulations confirm the exact result h =L /2 . Next we consider vacancy patterns, where a fraction f of particles on a lattice are randomly removed. These also display long-range density fluctuations, but with h =(L /2 )(f /d ) for small f , and h =L /2 for f →1 . And finally, for a hyperuniform system with no long-range density fluctuations, we consider "Einstein patterns," where each particle is independently displaced from a lattice site by a Gaussian-distributed amount. For these, at large L ,h approaches a constant equal to about half the root-mean-square displacement in each dimension. Then we turn to gray-scale pixel patterns that represent simulated arrangements of polydisperse particles, where the volume of a particle is encoded in the value of its central pixel. And we discuss the continuum limit of point patterns, where pixel size vanishes. In general, we thus propose to quantify particle configurations not just by the scaling of the density fluctuation spectrum but rather by the real-space spectrum of h (L ) versus L . We call this approach "hyperuniformity disorder length spectroscopy".

Characterizing pixel and point patterns with a hyperuniformity disorder length.

PubMed

Chieco, A T; Dreyfus, R; Durian, D J

2017-09-01

We introduce the concept of a "hyperuniformity disorder length" h that controls the variance of volume fraction fluctuations for randomly placed windows of fixed size. In particular, fluctuations are determined by the average number of particles within a distance h from the boundary of the window. We first compute special expectations and bounds in d dimensions, and then illustrate the range of behavior of h versus window size L by analyzing several different types of simulated two-dimensional pixel patterns-where particle positions are stored as a binary digital image in which pixels have value zero if empty and one if they contain a particle. The first are random binomial patterns, where pixels are randomly flipped from zero to one with probability equal to area fraction. These have long-ranged density fluctuations, and simulations confirm the exact result h=L/2. Next we consider vacancy patterns, where a fraction f of particles on a lattice are randomly removed. These also display long-range density fluctuations, but with h=(L/2)(f/d) for small f, and h=L/2 for f→1. And finally, for a hyperuniform system with no long-range density fluctuations, we consider "Einstein patterns," where each particle is independently displaced from a lattice site by a Gaussian-distributed amount. For these, at large L,h approaches a constant equal to about half the root-mean-square displacement in each dimension. Then we turn to gray-scale pixel patterns that represent simulated arrangements of polydisperse particles, where the volume of a particle is encoded in the value of its central pixel. And we discuss the continuum limit of point patterns, where pixel size vanishes. In general, we thus propose to quantify particle configurations not just by the scaling of the density fluctuation spectrum but rather by the real-space spectrum of h(L) versus L. We call this approach "hyperuniformity disorder length spectroscopy".

There is no MacWilliams identity for convolutional codes. [transmission gain comparison

NASA Technical Reports Server (NTRS)

Shearer, J. B.; Mceliece, R. J.

1977-01-01

An example is provided of two convolutional codes that have the same transmission gain but whose dual codes do not. This shows that no analog of the MacWilliams identity for block codes can exist relating the transmission gains of a convolutional code and its dual.

Image quality assessment using deep convolutional networks

NASA Astrophysics Data System (ADS)

Li, Yezhou; Ye, Xiang; Li, Yong

2017-12-01

This paper proposes a method of accurately assessing image quality without a reference image by using a deep convolutional neural network. Existing training based methods usually utilize a compact set of linear filters for learning features of images captured by different sensors to assess their quality. These methods may not be able to learn the semantic features that are intimately related with the features used in human subject assessment. Observing this drawback, this work proposes training a deep convolutional neural network (CNN) with labelled images for image quality assessment. The ReLU in the CNN allows non-linear transformations for extracting high-level image features, providing a more reliable assessment of image quality than linear filters. To enable the neural network to take images of any arbitrary size as input, the spatial pyramid pooling (SPP) is introduced connecting the top convolutional layer and the fully-connected layer. In addition, the SPP makes the CNN robust to object deformations to a certain extent. The proposed method taking an image as input carries out an end-to-end learning process, and outputs the quality of the image. It is tested on public datasets. Experimental results show that it outperforms existing methods by a large margin and can accurately assess the image quality on images taken by different sensors of varying sizes.

Dead pixel replacement in LWIR microgrid polarimeters.

PubMed

Ratliff, Bradley M; Tyo, J Scott; Boger, James K; Black, Wiley T; Bowers, David L; Fetrow, Matthew P

2007-06-11

LWIR imaging arrays are often affected by nonresponsive pixels, or "dead pixels." These dead pixels can severely degrade the quality of imagery and often have to be replaced before subsequent image processing and display of the imagery data. For LWIR arrays that are integrated with arrays of micropolarizers, the problem of dead pixels is amplified. Conventional dead pixel replacement (DPR) strategies cannot be employed since neighboring pixels are of different polarizations. In this paper we present two DPR schemes. The first is a modified nearest-neighbor replacement method. The second is a method based on redundancy in the polarization measurements.We find that the redundancy-based DPR scheme provides an order-of-magnitude better performance for typical LWIR polarimetric data.

Noise-enhanced convolutional neural networks.

PubMed

Audhkhasi, Kartik; Osoba, Osonde; Kosko, Bart

2016-06-01

Injecting carefully chosen noise can speed convergence in the backpropagation training of a convolutional neural network (CNN). The Noisy CNN algorithm speeds training on average because the backpropagation algorithm is a special case of the generalized expectation-maximization (EM) algorithm and because such carefully chosen noise always speeds up the EM algorithm on average. The CNN framework gives a practical way to learn and recognize images because backpropagation scales with training data. It has only linear time complexity in the number of training samples. The Noisy CNN algorithm finds a special separating hyperplane in the network's noise space. The hyperplane arises from the likelihood-based positivity condition that noise-boosts the EM algorithm. The hyperplane cuts through a uniform-noise hypercube or Gaussian ball in the noise space depending on the type of noise used. Noise chosen from above the hyperplane speeds training on average. Noise chosen from below slows it on average. The algorithm can inject noise anywhere in the multilayered network. Adding noise to the output neurons reduced the average per-iteration training-set cross entropy by 39% on a standard MNIST image test set of handwritten digits. It also reduced the average per-iteration training-set classification error by 47%. Adding noise to the hidden layers can also reduce these performance measures. The noise benefit is most pronounced for smaller data sets because the largest EM hill-climbing gains tend to occur in the first few iterations. This noise effect can assist random sampling from large data sets because it allows a smaller random sample to give the same or better performance than a noiseless sample gives. Copyright © 2015 Elsevier Ltd. All rights reserved.

Classification of Land Cover and Land Use Based on Convolutional Neural Networks

NASA Astrophysics Data System (ADS)

Yang, Chun; Rottensteiner, Franz; Heipke, Christian

2018-04-01

Land cover describes the physical material of the earth's surface, whereas land use describes the socio-economic function of a piece of land. Land use information is typically collected in geospatial databases. As such databases become outdated quickly, an automatic update process is required. This paper presents a new approach to determine land cover and to classify land use objects based on convolutional neural networks (CNN). The input data are aerial images and derived data such as digital surface models. Firstly, we apply a CNN to determine the land cover for each pixel of the input image. We compare different CNN structures, all of them based on an encoder-decoder structure for obtaining dense class predictions. Secondly, we propose a new CNN-based methodology for the prediction of the land use label of objects from a geospatial database. In this context, we present a strategy for generating image patches of identical size from the input data, which are classified by a CNN. Again, we compare different CNN architectures. Our experiments show that an overall accuracy of up to 85.7 % and 77.4 % can be achieved for land cover and land use, respectively. The classification of land cover has a positive contribution to the classification of the land use classification.

A review of advances in pixel detectors for experiments with high rate and radiation

NASA Astrophysics Data System (ADS)

Garcia-Sciveres, Maurice; Wermes, Norbert

2018-06-01

The large Hadron collider (LHC) experiments ATLAS and CMS have established hybrid pixel detectors as the instrument of choice for particle tracking and vertexing in high rate and radiation environments, as they operate close to the LHC interaction points. With the high luminosity-LHC upgrade now in sight, for which the tracking detectors will be completely replaced, new generations of pixel detectors are being devised. They have to address enormous challenges in terms of data throughput and radiation levels, ionizing and non-ionizing, that harm the sensing and readout parts of pixel detectors alike. Advances in microelectronics and microprocessing technologies now enable large scale detector designs with unprecedented performance in measurement precision (space and time), radiation hard sensors and readout chips, hybridization techniques, lightweight supports, and fully monolithic approaches to meet these challenges. This paper reviews the world-wide effort on these developments.

Characterization of a 2-mm thick, 16x16 Cadmium-Zinc-Telluride Pixel Array

NASA Technical Reports Server (NTRS)

Gaskin, Jessica; Richardson, Georgia; Mitchell, Shannon; Ramsey, Brian; Seller, Paul; Sharma, Dharma

2003-01-01

The detector under study is a 2-mm-thick, 16x16 Cadmium-Zinc-Telluride pixel array with a pixel pitch of 300 microns and inter-pixel gap of 50 microns. This detector is a precursor to that which will be used at the focal plane of the High Energy Replicated Optics (HERO) telescope currently being developed at Marshall Space Flight Center. With a telescope focal length of 6 meters, the detector needs to have a spatial resolution of around 200 microns in order to take full advantage of the HERO angular resolution. We discuss to what degree charge sharing will degrade energy resolution but will improve our spatial resolution through position interpolation. In addition, we discuss electric field modeling for this specific detector geometry and the role this mapping will play in terms of charge sharing and charge loss in the detector.

Research on image retrieval using deep convolutional neural network combining L1 regularization and PRelu activation function

NASA Astrophysics Data System (ADS)

QingJie, Wei; WenBin, Wang

2017-06-01

In this paper, the image retrieval using deep convolutional neural network combined with regularization and PRelu activation function is studied, and improves image retrieval accuracy. Deep convolutional neural network can not only simulate the process of human brain to receive and transmit information, but also contains a convolution operation, which is very suitable for processing images. Using deep convolutional neural network is better than direct extraction of image visual features for image retrieval. However, the structure of deep convolutional neural network is complex, and it is easy to over-fitting and reduces the accuracy of image retrieval. In this paper, we combine L1 regularization and PRelu activation function to construct a deep convolutional neural network to prevent over-fitting of the network and improve the accuracy of image retrieval

Measurements with MÖNCH, a 25 μm pixel pitch hybrid pixel detector

NASA Astrophysics Data System (ADS)

Ramilli, M.; Bergamaschi, A.; Andrae, M.; Brückner, M.; Cartier, S.; Dinapoli, R.; Fröjdh, E.; Greiffenberg, D.; Hutwelker, T.; Lopez-Cuenca, C.; Mezza, D.; Mozzanica, A.; Ruat, M.; Redford, S.; Schmitt, B.; Shi, X.; Tinti, G.; Zhang, J.

2017-01-01

MÖNCH is a hybrid silicon pixel detector based on charge integration and with analog readout, featuring a pixel size of 25×25 μm2. The latest working prototype consists of an array of 400×400 identical pixels for a total active area of 1×1 cm2. Its design is optimized for the single photon regime. An exhaustive characterization of this large area prototype has been carried out in the past months, and it confirms an ENC in the order of 35 electrons RMS and a dynamic range of ~4×12 keV photons in high gain mode, which increases to ~100×12 keV photons with the lowest gain setting. The low noise levels of MÖNCH make it a suitable candidate for X-ray detection at energies around 1 keV and below. Imaging applications in particular can benefit significantly from the use of MÖNCH: due to its extremely small pixel pitch, the detector intrinsically offers excellent position resolution. Moreover, in low flux conditions, charge sharing between neighboring pixels allows the use of position interpolation algorithms which grant a resolution at the micrometer-level. Its energy reconstruction and imaging capabilities have been tested for the first time at a low energy beamline at PSI, with photon energies between 1.75 keV and 3.5 keV, and results will be shown.

Deep Learning with Hierarchical Convolutional Factor Analysis

PubMed Central

Chen, Bo; Polatkan, Gungor; Sapiro, Guillermo; Blei, David; Dunson, David; Carin, Lawrence

2013-01-01

Unsupervised multi-layered (“deep”) models are considered for general data, with a particular focus on imagery. The model is represented using a hierarchical convolutional factor-analysis construction, with sparse factor loadings and scores. The computation of layer-dependent model parameters is implemented within a Bayesian setting, employing a Gibbs sampler and variational Bayesian (VB) analysis, that explicitly exploit the convolutional nature of the expansion. In order to address large-scale and streaming data, an online version of VB is also developed. The number of basis functions or dictionary elements at each layer is inferred from the data, based on a beta-Bernoulli implementation of the Indian buffet process. Example results are presented for several image-processing applications, with comparisons to related models in the literature. PMID:23787342

Co-trained convolutional neural networks for automated detection of prostate cancer in multi-parametric MRI.

PubMed

Yang, Xin; Liu, Chaoyue; Wang, Zhiwei; Yang, Jun; Min, Hung Le; Wang, Liang; Cheng, Kwang-Ting Tim

2017-12-01

Multi-parameter magnetic resonance imaging (mp-MRI) is increasingly popular for prostate cancer (PCa) detection and diagnosis. However, interpreting mp-MRI data which typically contains multiple unregistered 3D sequences, e.g. apparent diffusion coefficient (ADC) and T2-weighted (T2w) images, is time-consuming and demands special expertise, limiting its usage for large-scale PCa screening. Therefore, solutions to computer-aided detection of PCa in mp-MRI images are highly desirable. Most recent advances in automated methods for PCa detection employ a handcrafted feature based two-stage classification flow, i.e. voxel-level classification followed by a region-level classification. This work presents an automated PCa detection system which can concurrently identify the presence of PCa in an image and localize lesions based on deep convolutional neural network (CNN) features and a single-stage SVM classifier. Specifically, the developed co-trained CNNs consist of two parallel convolutional networks for ADC and T2w images respectively. Each network is trained using images of a single modality in a weakly-supervised manner by providing a set of prostate images with image-level labels indicating only the presence of PCa without priors of lesions' locations. Discriminative visual patterns of lesions can be learned effectively from clutters of prostate and surrounding tissues. A cancer response map with each pixel indicating the likelihood to be cancerous is explicitly generated at the last convolutional layer of the network for each modality. A new back-propagated error E is defined to enforce both optimized classification results and consistent cancer response maps for different modalities, which help capture highly representative PCa-relevant features during the CNN feature learning process. The CNN features of each modality are concatenated and fed into a SVM classifier. For images which are classified to contain cancers, non-maximum suppression and adaptive

Fabrication of X-ray Microcalorimeter Focal Planes Composed of Two Distinct Pixel Types

NASA Technical Reports Server (NTRS)

Wassell, Edward J.; Adams, Joseph S.; Bandler, Simon R.; Betancour-Martinez, Gabriele L; Chiao, Meng P.; Chang, Meng Ping; Chervenak, James A.; Datesman, Aaron M.; Eckart, Megan E.; Ewin, Audrey J.;

Super-Resolution for “Jilin-1” Satellite Video Imagery via a Convolutional Network

PubMed Central

Wang, Zhongyuan; Wang, Lei; Ren, Yexian

2018-01-01

Super-resolution for satellite video attaches much significance to earth observation accuracy, and the special imaging and transmission conditions on the video satellite pose great challenges to this task. The existing deep convolutional neural-network-based methods require pre-processing or post-processing to be adapted to a high-resolution size or pixel format, leading to reduced performance and extra complexity. To this end, this paper proposes a five-layer end-to-end network structure without any pre-processing and post-processing, but imposes a reshape or deconvolution layer at the end of the network to retain the distribution of ground objects within the image. Meanwhile, we formulate a joint loss function by combining the output and high-dimensional features of a non-linear mapping network to precisely learn the desirable mapping relationship between low-resolution images and their high-resolution counterparts. Also, we use satellite video data itself as a training set, which favors consistency between training and testing images and promotes the method’s practicality. Experimental results on “Jilin-1” satellite video imagery show that this method demonstrates a superior performance in terms of both visual effects and measure metrics over competing methods. PMID:29652838

Visualization of dyed NAPL concentration in transparent porous media using color space components.

PubMed

Kashuk, Sina; Mercurio, Sophia R; Iskander, Magued

2014-07-01

Finding a correlation between image pixel information and non-aqueous phase liquid (NAPL) saturation is an important issue in bench-scale geo-environmental model studies that employ optical imaging techniques. Another concern is determining the best dye color and its optimum concentration as a tracer for use in mapping NAPL zones. Most bench scale flow studies employ monochromatic gray-scale imaging to analyze the concentration of mostly red dyed NAPL tracers in porous media. However, the use of grayscale utilizes a third of the available information in color images, which typically contain three color-space components. In this study, eight color spaces consisting of 24 color-space components were calibrated against dye concentration for three color-dyes. Additionally, multiple color space components were combined to increase the correlation between color-space data and dyed NAPL concentration. This work is performed to support imaging of NAPL migration in transparent synthetic soils representing the macroscopic behavior of natural soils. The transparent soil used in this study consists of fused quartz and a matched refractive index mineral-oil solution that represents the natural aquifer. The objective is to determine the best color dye concentration and ideal color space components for rendering dyed sucrose-saturated fused quartz that represents contamination of the natural aquifer by a dense NAPL (DNAPL). Calibration was achieved for six NAPL zone lengths using 3456 images (24 color space components×3 dyes×48 NAPL combinations) of contaminants within a defined criteria expressed as peak signal to noise ratio. The effect of data filtering was also considered and a convolution average filter is recommended for image conditioning. The technology presented in this paper is fast, accurate, non-intrusive and inexpensive method for quantifying contamination zones using transparent soil models. Copyright © 2014 Elsevier B.V. All rights reserved.

A convolutional neural network to filter artifacts in spectroscopic MRI.

PubMed

Gurbani, Saumya S; Schreibmann, Eduard; Maudsley, Andrew A; Cordova, James Scott; Soher, Brian J; Poptani, Harish; Verma, Gaurav; Barker, Peter B; Shim, Hyunsuk; Cooper, Lee A D

2018-03-09

Proton MRSI is a noninvasive modality capable of generating volumetric maps of in vivo tissue metabolism without the need for ionizing radiation or injected contrast agent. Magnetic resonance spectroscopic imaging has been shown to be a viable imaging modality for studying several neuropathologies. However, a key hurdle in the routine clinical adoption of MRSI is the presence of spectral artifacts that can arise from a number of sources, possibly leading to false information. A deep learning model was developed that was capable of identifying and filtering out poor quality spectra. The core of the model used a tiled convolutional neural network that analyzed frequency-domain spectra to detect artifacts. When compared with a panel of MRS experts, our convolutional neural network achieved high sensitivity and specificity with an area under the curve of 0.95. A visualization scheme was implemented to better understand how the convolutional neural network made its judgement on single-voxel or multivoxel MRSI, and the convolutional neural network was embedded into a pipeline capable of producing whole-brain spectroscopic MRI volumes in real time. The fully automated method for assessment of spectral quality provides a valuable tool to support clinical MRSI or spectroscopic MRI studies for use in fields such as adaptive radiation therapy planning. © 2018 International Society for Magnetic Resonance in Medicine.

CVD diamond pixel detectors for LHC experiments

NASA Astrophysics Data System (ADS)

Wedenig, R.; Adam, W.; Bauer, C.; Berdermann, E.; Bergonzo, P.; Bogani, F.; Borchi, E.; Brambilla, A.; Bruzzi, M.; Colledani, C.; Conway, J.; Dabrowski, W.; Delpierre, P.; Deneuville, A.; Dulinski, W.; van Eijk, B.; Fallou, A.; Fizzotti, F.; Foulon, F.; Friedl, M.; Gan, K. K.; Gheeraert, E.; Grigoriev, E.; Hallewell, G.; Hall-Wilton, R.; Han, S.; Hartjes, F.; Hrubec, J.; Husson, D.; Kagan, H.; Kania, D.; Kaplon, J.; Karl, C.; Kass, R.; Knöpfle, K. T.; Krammer, M.; Logiudice, A.; Lu, R.; Manfredi, P. F.; Manfredotti, C.; Marshall, R. D.; Meier, D.; Mishina, M.; Oh, A.; Pan, L. S.; Palmieri, V. G.; Pernicka, M.; Peitz, A.; Pirollo, S.; Polesello, P.; Pretzl, K.; Procario, M.; Re, V.; Riester, J. L.; Roe, S.; Roff, D.; Rudge, A.; Runolfsson, O.; Russ, J.; Schnetzer, S.; Sciortino, S.; Speziali, V.; Stelzer, H.; Stone, R.; Suter, B.; Tapper, R. J.; Tesarek, R.; Trawick, M.; Trischuk, W.; Vittone, E.; Wagner, A.; Walsh, A. M.; Weilhammer, P.; White, C.; Zeuner, W.; Ziock, H.; Zoeller, M.; Blanquart, L.; Breugnion, P.; Charles, E.; Ciocio, A.; Clemens, J. C.; Dao, K.; Einsweiler, K.; Fasching, D.; Fischer, P.; Joshi, A.; Keil, M.; Klasen, V.; Kleinfelder, S.; Laugier, D.; Meuser, S.; Milgrome, O.; Mouthuy, T.; Richardson, J.; Sinervo, P.; Treis, J.; Wermes, N.; RD42 Collaboration

1999-08-01

This paper reviews the development of CVD diamond pixel detectors. The preparation of the diamond pixel sensors for bump-bonding to the pixel readout electronics for the LHC and the results from beam tests carried out at CERN are described.

On the growth and form of cortical convolutions

NASA Astrophysics Data System (ADS)

Tallinen, Tuomas; Chung, Jun Young; Rousseau, François; Girard, Nadine; Lefèvre, Julien; Mahadevan, L.

2016-06-01

The rapid growth of the human cortex during development is accompanied by the folding of the brain into a highly convoluted structure. Recent studies have focused on the genetic and cellular regulation of cortical growth, but understanding the formation of the gyral and sulcal convolutions also requires consideration of the geometry and physical shaping of the growing brain. To study this, we use magnetic resonance images to build a 3D-printed layered gel mimic of the developing smooth fetal brain; when immersed in a solvent, the outer layer swells relative to the core, mimicking cortical growth. This relative growth puts the outer layer into mechanical compression and leads to sulci and gyri similar to those in fetal brains. Starting with the same initial geometry, we also build numerical simulations of the brain modelled as a soft tissue with a growing cortex, and show that this also produces the characteristic patterns of convolutions over a realistic developmental course. All together, our results show that although many molecular determinants control the tangential expansion of the cortex, the size, shape, placement and orientation of the folds arise through iterations and variations of an elementary mechanical instability modulated by early fetal brain geometry.

Low Power Camera-on-a-Chip Using CMOS Active Pixel Sensor Technology

NASA Technical Reports Server (NTRS)

Fossum, E. R.

1995-01-01

A second generation image sensor technology has been developed at the NASA Jet Propulsion Laboratory as a result of the continuing need to miniaturize space science imaging instruments. Implemented using standard CMOS, the active pixel sensor (APS) technology permits the integration of the detector array with on-chip timing, control and signal chain electronics, including analog-to-digital conversion.

QCDNUM: Fast QCD evolution and convolution

NASA Astrophysics Data System (ADS)

Botje, M.

2011-02-01

The QCDNUM program numerically solves the evolution equations for parton densities and fragmentation functions in perturbative QCD. Un-polarised parton densities can be evolved up to next-to-next-to-leading order in powers of the strong coupling constant, while polarised densities or fragmentation functions can be evolved up to next-to-leading order. Other types of evolution can be accessed by feeding alternative sets of evolution kernels into the program. A versatile convolution engine provides tools to compute parton luminosities, cross-sections in hadron-hadron scattering, and deep inelastic structure functions in the zero-mass scheme or in generalised mass schemes. Input to these calculations are either the QCDNUM evolved densities, or those read in from an external parton density repository. Included in the software distribution are packages to calculate zero-mass structure functions in un-polarised deep inelastic scattering, and heavy flavour contributions to these structure functions in the fixed flavour number scheme. Program summaryProgram title: QCDNUM version: 17.00 Catalogue identifier: AEHV_v1_0 Program summary URL:http://cpc.cs.qub.ac.uk/summaries/AEHV_v1_0.html Program obtainable from: CPC Program Library, Queen's University, Belfast, N. Ireland Licensing provisions: GNU Public Licence No. of lines in distributed program, including test data, etc.: 45 736 No. of bytes in distributed program, including test data, etc.: 911 569 Distribution format: tar.gz Programming language: Fortran-77 Computer: All Operating system: All RAM: Typically 3 Mbytes Classification: 11.5 Nature of problem: Evolution of the strong coupling constant and parton densities, up to next-to-next-to-leading order in perturbative QCD. Computation of observable quantities by Mellin convolution of the evolved densities with partonic cross-sections. Solution method: Parametrisation of the parton densities as linear or quadratic splines on a discrete grid, and evolution of the spline

Effect of manmade pixels on the inherent dimension of natural material distributions

NASA Astrophysics Data System (ADS)

Schlamm, Ariel; Messinger, David; Basener, William

2009-05-01

The inherent dimension of hyperspectral data may be a useful metric for discriminating between the presence of manmade and natural materials in a scene without reliance on spectral signatures take from libraries. Previously, a simple geometric method for approximating the inherent dimension was introduced along with results from application to single material clusters. This method uses an estimate of the slope from a graph based on the point density estimation in the spectral space. Other information can be gathered from the plot which may aid in the discrimination between manmade and natural materials. In order to use these measures to differentiate between the two material types, the effect of the inclusion of manmade pixels on the phenomenology of the background distribution must be evaluated. Here, a procedure for injecting manmade pixels into a natural region of a scene is discussed. The results of dimension estimation on natural scenes with varying amounts of manmade pixels injected are presented here, indicating that these metrics can be sensitive to the presence of manmade phenomenology in an image.

Fully Convolutional Network-Based Multifocus Image Fusion.

PubMed

Guo, Xiaopeng; Nie, Rencan; Cao, Jinde; Zhou, Dongming; Qian, Wenhua

2018-07-01

As the optical lenses for cameras always have limited depth of field, the captured images with the same scene are not all in focus. Multifocus image fusion is an efficient technology that can synthesize an all-in-focus image using several partially focused images. Previous methods have accomplished the fusion task in spatial or transform domains. However, fusion rules are always a problem in most methods. In this letter, from the aspect of focus region detection, we propose a novel multifocus image fusion method based on a fully convolutional network (FCN) learned from synthesized multifocus images. The primary novelty of this method is that the pixel-wise focus regions are detected through a learning FCN, and the entire image, not just the image patches, are exploited to train the FCN. First, we synthesize 4500 pairs of multifocus images by repeatedly using a gaussian filter for each image from PASCAL VOC 2012, to train the FCN. After that, a pair of source images is fed into the trained FCN, and two score maps indicating the focus property are generated. Next, an inversed score map is averaged with another score map to produce an aggregative score map, which take full advantage of focus probabilities in two score maps. We implement the fully connected conditional random field (CRF) on the aggregative score map to accomplish and refine a binary decision map for the fusion task. Finally, we exploit the weighted strategy based on the refined decision map to produce the fused image. To demonstrate the performance of the proposed method, we compare its fused results with several start-of-the-art methods not only on a gray data set but also on a color data set. Experimental results show that the proposed method can achieve superior fusion performance in both human visual quality and objective assessment.

Mapping Electrical Crosstalk in Pixelated Sensor Arrays

NASA Technical Reports Server (NTRS)

Seshadri, Suresh (Inventor); Cole, David (Inventor); Smith, Roger M. (Inventor); Hancock, Bruce R. (Inventor)

2017-01-01

The effects of inter pixel capacitance in a pixilated array may be measured by first resetting all pixels in the array to a first voltage, where a first image is read out, followed by resetting only a subset of pixels in the array to a second voltage, where a second image is read out, where the difference in the first and second images provide information about the inter pixel capacitance. Other embodiments are described and claimed.

Mapping Electrical Crosstalk in Pixelated Sensor Arrays

NASA Technical Reports Server (NTRS)

Smith, Roger M (Inventor); Hancock, Bruce R. (Inventor); Cole, David (Inventor); Seshadri, Suresh (Inventor)

2013-01-01

The effects of inter pixel capacitance in a pixilated array may be measured by first resetting all pixels in the array to a first voltage, where a first image is read out, followed by resetting only a subset of pixels in the array to a second voltage, where a second image is read out, where the difference in the first and second images provide information about the inter pixel capacitance. Other embodiments are described and claimed.

Application of structured support vector machine backpropagation to a convolutional neural network for human pose estimation.

PubMed

Witoonchart, Peerajak; Chongstitvatana, Prabhas

2017-08-01

In this study, for the first time, we show how to formulate a structured support vector machine (SSVM) as two layers in a convolutional neural network, where the top layer is a loss augmented inference layer and the bottom layer is the normal convolutional layer. We show that a deformable part model can be learned with the proposed structured SVM neural network by backpropagating the error of the deformable part model to the convolutional neural network. The forward propagation calculates the loss augmented inference and the backpropagation calculates the gradient from the loss augmented inference layer to the convolutional layer. Thus, we obtain a new type of convolutional neural network called an Structured SVM convolutional neural network, which we applied to the human pose estimation problem. This new neural network can be used as the final layers in deep learning. Our method jointly learns the structural model parameters and the appearance model parameters. We implemented our method as a new layer in the existing Caffe library. Copyright © 2017 Elsevier Ltd. All rights reserved.

Pixel CdTe semiconductor module to implement a sub-MeV imaging detector for astrophysics

NASA Astrophysics Data System (ADS)

Gálvez, J.-L.; Hernanz, M.; Álvarez, L.; Artigues, B.; Álvarez, J.-M.; Ullán, M.; Pellegrini, G.; Lozano, M.; Cabruja, E.; Martínez, R.; Chmeissani, M.; Puigdengoles, C.

2017-03-01

Stellar explosions are relevant and interesting astrophysical phenomena. Since long ago we have been working on the characterization of nova and supernova explosions in X and gamma rays, with the use of space missions such as INTEGRAL, XMM-Newton and Swift. We have been also involved in feasibility studies of future instruments in the energy range from several keV up to a few MeV, in collaboration with other research institutes, such as GRI, DUAL and e-ASTROGAM. High sensitivities are essential to perform detailed studies of cosmic explosions and cosmic accelerators, e.g., Supernovae, Classical Novae, Supernova Remnants (SNRs), Gamma-Ray Bursts (GRBs). In order to fulfil the combined requirement of high detection efficiency with good spatial and energy resolution, an initial module prototype based on CdTe pixel detectors is being developed. The detector dimensions are 12.5mm x 12.5mm x 2mm, with a pixel pitch of 1mm x 1mm. Each pixel is bump bonded to a fanout board made of Sapphire substrate and routed to the corresponding input channel of the readout ASIC, to measure pixel position and pulse height for each incident gamma-ray photon. An ohmic CdTe pixel detector has been characterised by means of 57Co, 133Ba and 22Na sources. Based on this, its spectroscopic performance and the influence of charge sharing is reported here. The pixel study is complemented by the simulation of the CdTe module performance using the GEANT 4 and MEGALIB tools, which will help us to optimise the pixel size selection.

New coding advances for deep space communications

NASA Technical Reports Server (NTRS)

Yuen, Joseph H.

1987-01-01

Advances made in error-correction coding for deep space communications are described. The code believed to be the best is a (15, 1/6) convolutional code, with maximum likelihood decoding; when it is concatenated with a 10-bit Reed-Solomon code, it achieves a bit error rate of 10 to the -6th, at a bit SNR of 0.42 dB. This code outperforms the Voyager code by 2.11 dB. The use of source statics in decoding convolutionally encoded Voyager images from the Uranus encounter is investigated, and it is found that a 2 dB decoding gain can be achieved.

Shadow-free single-pixel imaging

NASA Astrophysics Data System (ADS)

Li, Shunhua; Zhang, Zibang; Ma, Xiao; Zhong, Jingang

2017-11-01

Single-pixel imaging is an innovative imaging scheme and receives increasing attention in recent years, for it is applicable for imaging at non-visible wavelengths and imaging under weak light conditions. However, as in conventional imaging, shadows would likely occur in single-pixel imaging and sometimes bring negative effects in practical uses. In this paper, the principle of shadows occurrence in single-pixel imaging is analyzed, following which a technique for shadows removal is proposed. In the proposed technique, several single-pixel detectors are used to detect the backscattered light at different locations so that the shadows in the reconstructed images corresponding to each detector shadows are complementary. Shadow-free reconstruction can be derived by fusing the shadow-complementary images using maximum selection rule. To deal with the problem of intensity mismatch in image fusion, we put forward a simple calibration. As experimentally demonstrated, the technique is able to reconstruct monochromatic and full-color shadow-free images.

A Double Dwell High Sensitivity GPS Acquisition Scheme Using Binarized Convolution Neural Network

PubMed Central

Wang, Zhen; Zhuang, Yuan; Yang, Jun; Zhang, Hengfeng; Dong, Wei; Wang, Min; Hua, Luchi; Liu, Bo; Shi, Longxing

2018-01-01

Conventional GPS acquisition methods, such as Max selection and threshold crossing (MAX/TC), estimate GPS code/Doppler by its correlation peak. Different from MAX/TC, a multi-layer binarized convolution neural network (BCNN) is proposed to recognize the GPS acquisition correlation envelope in this article. The proposed method is a double dwell acquisition in which a short integration is adopted in the first dwell and a long integration is applied in the second one. To reduce the search space for parameters, BCNN detects the possible envelope which contains the auto-correlation peak in the first dwell to compress the initial search space to 1/1023. Although there is a long integration in the second dwell, the acquisition computation overhead is still low due to the compressed search space. Comprehensively, the total computation overhead of the proposed method is only 1/5 of conventional ones. Experiments show that the proposed double dwell/correlation envelope identification (DD/CEI) neural network achieves 2 dB improvement when compared with the MAX/TC under the same specification. PMID:29747373

Building Extraction from Remote Sensing Data Using Fully Convolutional Networks

NASA Astrophysics Data System (ADS)

Bittner, K.; Cui, S.; Reinartz, P.

2017-05-01

Building detection and footprint extraction are highly demanded for many remote sensing applications. Though most previous works have shown promising results, the automatic extraction of building footprints still remains a nontrivial topic, especially in complex urban areas. Recently developed extensions of the CNN framework made it possible to perform dense pixel-wise classification of input images. Based on these abilities we propose a methodology, which automatically generates a full resolution binary building mask out of a Digital Surface Model (DSM) using a Fully Convolution Network (FCN) architecture. The advantage of using the depth information is that it provides geometrical silhouettes and allows a better separation of buildings from background as well as through its invariance to illumination and color variations. The proposed framework has mainly two steps. Firstly, the FCN is trained on a large set of patches consisting of normalized DSM (nDSM) as inputs and available ground truth building mask as target outputs. Secondly, the generated predictions from FCN are viewed as unary terms for a Fully connected Conditional Random Fields (FCRF), which enables us to create a final binary building mask. A series of experiments demonstrate that our methodology is able to extract accurate building footprints which are close to the buildings original shapes to a high degree. The quantitative and qualitative analysis show the significant improvements of the results in contrast to the multy-layer fully connected network from our previous work.

ENGage: The use of space and pixel art for increasing primary school children's interest in science, technology, engineering and mathematics

NASA Astrophysics Data System (ADS)

Roberts, Simon J.

2014-01-01

The Faculty of Engineering at The University of Nottingham, UK, has developed interdisciplinary, hands-on workshops for primary schools that introduce space technology, its relevance to everyday life and the importance of science, technology, engineering and maths. The workshop activities for 7-11 year olds highlight the roles that space and satellite technology play in observing and monitoring the Earth's biosphere as well as being vital to communications in the modern digital world. The programme also provides links to 'how science works', the environment and citizenship and uses pixel art through the medium of digital photography to demonstrate the importance of maths in a novel and unconventional manner. The interactive programme of activities provides learners with an opportunity to meet 'real' scientists and engineers, with one of the key messages from the day being that anyone can become involved in science and engineering whatever their ability or subject of interest. The methodology introduces the role of scientists and engineers using space technology themes, but it could easily be adapted for use with any inspirational topic. Analysis of learners' perceptions of science, technology, engineering and maths before and after participating in ENGage showed very positive and significant changes in their attitudes to these subjects and an increase in the number of children thinking they would be interested and capable in pursuing a career in science and engineering. This paper provides an overview of the activities, the methodology, the evaluation process and results.

WFC3 UVIS Pixel-to-Pixel QE Variations via Internal Flats Monitor

NASA Astrophysics Data System (ADS)

Bajaj, Varun

2016-10-01

The UVIS detector has a population of pixels that exhibit anomalous QE variations between anneals, characterized by a sensitivity loss that is greater in the blue than in the red. This population is randomly distributed, with evidence of clustering behavior in the UV, and is seemingly unique for each anneal cycle. This program, a continuation of cycle 23 program 14389, will aim to constrain the maximum low-sensitivity population existing before an anneal in both the UV and Visible filters. To monitor the UV behavior, internal flats with the D2 lamp will be taken through F225W and F336W. To monitor the behavior in the Visible filters, internal flats with the tungsten lamp will be taken a week before the anneal, when the population of anomalous pixels is the greatest. Internal flats with the Tungsten lamp will be taken to monitor the population in the visible filters, with data taken the week before the anneal to sample the maximum population of anomalous pixels.

A fully convolutional network for weed mapping of unmanned aerial vehicle (UAV) imagery.

PubMed

Huang, Huasheng; Deng, Jizhong; Lan, Yubin; Yang, Aqing; Deng, Xiaoling; Zhang, Lei

2018-01-01

Appropriate Site Specific Weed Management (SSWM) is crucial to ensure the crop yields. Within SSWM of large-scale area, remote sensing is a key technology to provide accurate weed distribution information. Compared with satellite and piloted aircraft remote sensing, unmanned aerial vehicle (UAV) is capable of capturing high spatial resolution imagery, which will provide more detailed information for weed mapping. The objective of this paper is to generate an accurate weed cover map based on UAV imagery. The UAV RGB imagery was collected in 2017 October over the rice field located in South China. The Fully Convolutional Network (FCN) method was proposed for weed mapping of the collected imagery. Transfer learning was used to improve generalization capability, and skip architecture was applied to increase the prediction accuracy. After that, the performance of FCN architecture was compared with Patch_based CNN algorithm and Pixel_based CNN method. Experimental results showed that our FCN method outperformed others, both in terms of accuracy and efficiency. The overall accuracy of the FCN approach was up to 0.935 and the accuracy for weed recognition was 0.883, which means that this algorithm is capable of generating accurate weed cover maps for the evaluated UAV imagery.

SU-E-T-423: Fast Photon Convolution Calculation with a 3D-Ideal Kernel On the GPU

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

Moriya, S; Sato, M; Tachibana, H

Purpose: The calculation time is a trade-off for improving the accuracy of convolution dose calculation with fine calculation spacing of the KERMA kernel. We investigated to accelerate the convolution calculation using an ideal kernel on the Graphic Processing Units (GPU). Methods: The calculation was performed on the AMD graphics hardware of Dual FirePro D700 and our algorithm was implemented using the Aparapi that convert Java bytecode to OpenCL. The process of dose calculation was separated with the TERMA and KERMA steps. The dose deposited at the coordinate (x, y, z) was determined in the process. In the dose calculation runningmore » on the central processing unit (CPU) of Intel Xeon E5, the calculation loops were performed for all calculation points. On the GPU computation, all of the calculation processes for the points were sent to the GPU and the multi-thread computation was done. In this study, the dose calculation was performed in a water equivalent homogeneous phantom with 150{sup 3} voxels (2 mm calculation grid) and the calculation speed on the GPU to that on the CPU and the accuracy of PDD were compared. Results: The calculation time for the GPU and the CPU were 3.3 sec and 4.4 hour, respectively. The calculation speed for the GPU was 4800 times faster than that for the CPU. The PDD curve for the GPU was perfectly matched to that for the CPU. Conclusion: The convolution calculation with the ideal kernel on the GPU was clinically acceptable for time and may be more accurate in an inhomogeneous region. Intensity modulated arc therapy needs dose calculations for different gantry angles at many control points. Thus, it would be more practical that the kernel uses a coarse spacing technique if the calculation is faster while keeping the similar accuracy to a current treatment planning system.« less

Color constancy using bright-neutral pixels

NASA Astrophysics Data System (ADS)

Wang, Yanfang; Luo, Yupin

2014-03-01

An effective illuminant-estimation approach for color constancy is proposed. Bright and near-neutral pixels are selected to jointly represent the illuminant color and utilized for illuminant estimation. To assess the representing capability of pixels, bright-neutral strength (BNS) is proposed by combining pixel chroma and brightness. Accordingly, a certain percentage of pixels with the largest BNS is selected to be the representative set. For every input image, a proper percentage value is determined via an iterative strategy by seeking the optimal color-corrected image. To compare various color-corrected images of an input image, image color-cast degree (ICCD) is devised using means and standard deviations of RGB channels. Experimental evaluation on standard real-world datasets validates the effectiveness of the proposed approach.

A 400 KHz line rate 2048-pixel stitched SWIR linear array

NASA Astrophysics Data System (ADS)

Anchlia, Ankur; Vinella, Rosa M.; Gielen, Daphne; Wouters, Kristof; Vervenne, Vincent; Hooylaerts, Peter; Deroo, Pieter; Ruythooren, Wouter; De Gaspari, Danny; Das, Jo; Merken, Patrick

2016-05-01

Xenics has developed a family of stitched SWIR long linear arrays that operate up to 400 KHz of line rate. These arrays serve medical and industrial applications that require high line rates as well as space applications that require long linear arrays. The arrays are based on a modular ROIC design concept: modules of 512 pixels are stitched during fabrication to achieve 512, 1024 and 2048 pixel arrays. Each 512-pixel module has its own on-chip digital sequencer, analog readout chain and 4 output buffers. This modular concept enables a long array to run at a high line rates irrespective of the array length, which limits the line rate in a traditional linear array. The ROIC is flip-chipped with InGaAs detector arrays. The FPA has a pixel pitch of 12.5μm and has two pixel flavors: square (12.5μm) and rectangular (250μm). The frontend circuit is based on Capacitive Trans-impedance Amplifier (CTIA) to attain stable detector bias, and good linearity and signal integrity, especially at high speeds. The CTIA has an input auto-zero mechanism that allows to have low detector bias (<20mV). An on-chip Correlated Double Sample (CDS) facilitates removal of CTIA KTC and 1/f noise, and other offsets, achieving low noise performance. There are five gain modes in the FPA giving the full well range from 85Ke- to 40Me-. The measured input referred noise is 35e-rms in the highest gain mode. The FPA operates in Integrate While Read mode and, at a master clock rate of 60MHz and a minimum integration time of 1.4μs, achieves the highest line rate of 400 KHz. In this paper, design details and measurements results are presented in order to demonstrate the array performance.

Traffic sign recognition based on deep convolutional neural network

NASA Astrophysics Data System (ADS)

Yin, Shi-hao; Deng, Ji-cai; Zhang, Da-wei; Du, Jing-yuan

2017-11-01

Traffic sign recognition (TSR) is an important component of automated driving systems. It is a rather challenging task to design a high-performance classifier for the TSR system. In this paper, we propose a new method for TSR system based on deep convolutional neural network. In order to enhance the expression of the network, a novel structure (dubbed block-layer below) which combines network-in-network and residual connection is designed. Our network has 10 layers with parameters (block-layer seen as a single layer): the first seven are alternate convolutional layers and block-layers, and the remaining three are fully-connected layers. We train our TSR network on the German traffic sign recognition benchmark (GTSRB) dataset. To reduce overfitting, we perform data augmentation on the training images and employ a regularization method named "dropout". The activation function we employ in our network adopts scaled exponential linear units (SELUs), which can induce self-normalizing properties. To speed up the training, we use an efficient GPU to accelerate the convolutional operation. On the test dataset of GTSRB, we achieve the accuracy rate of 99.67%, exceeding the state-of-the-art results.

Monitoring anti-angiogenic therapy in colorectal cancer murine model using dynamic contrast-enhanced MRI: comparing pixel-by-pixel with region of interest analysis.

PubMed

Haney, C R; Fan, X; Markiewicz, E; Mustafi, D; Karczmar, G S; Stadler, W M

2013-02-01

Sorafenib is a multi-kinase inhibitor that blocks cell proliferation and angiogenesis. It is currently approved for advanced hepatocellular and renal cell carcinomas in humans, where its major mechanism of action is thought to be through inhibition of vascular endothelial growth factor and platelet-derived growth factor receptors. The purpose of this study was to determine whether pixel-by-pixel analysis of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is better able to capture the heterogeneous response of Sorafenib in a murine model of colorectal tumor xenografts (as compared with region of interest analysis). MRI was performed on a 9.4 T pre-clinical scanner on the initial treatment day. Then either vehicle or drug were gavaged daily (3 days) up to the final image. Four days later, the mice were again imaged. The two-compartment model and reference tissue method of DCE-MRI were used to analyze the data. The results demonstrated that the contrast agent distribution rate constant (K(trans)) were significantly reduced (p < 0.005) at day-4 of Sorafenib treatment. In addition, the K(trans) of nearby muscle was also reduced after Sorafenib treatment. The pixel-by-pixel analysis (compared to region of interest analysis) was better able to capture the heterogeneity of the tumor and the decrease in K(trans) four days after treatment. For both methods, the volume of the extravascular extracellular space did not change significantly after treatment. These results confirm that parameters such as K(trans), could provide a non-invasive biomarker to assess the response to anti-angiogenic therapies such as Sorafenib, but that the heterogeneity of response across a tumor requires a more detailed analysis than has typically been undertaken.

Optimal convolution SOR acceleration of waveform relaxation with application to semiconductor device simulation

NASA Technical Reports Server (NTRS)

Reichelt, Mark

1993-01-01

In this paper we describe a novel generalized SOR (successive overrelaxation) algorithm for accelerating the convergence of the dynamic iteration method known as waveform relaxation. A new convolution SOR algorithm is presented, along with a theorem for determining the optimal convolution SOR parameter. Both analytic and experimental results are given to demonstrate that the convergence of the convolution SOR algorithm is substantially faster than that of the more obvious frequency-independent waveform SOR algorithm. Finally, to demonstrate the general applicability of this new method, it is used to solve the differential-algebraic system generated by spatial discretization of the time-dependent semiconductor device equations.

It's not the pixel count, you fool

NASA Astrophysics Data System (ADS)

Kriss, Michael A.

2012-01-01

The first thing a "marketing guy" asks the digital camera engineer is "how many pixels does it have, for we need as many mega pixels as possible since the other guys are killing us with their "umpteen" mega pixel pocket sized digital cameras. And so it goes until the pixels get smaller and smaller in order to inflate the pixel count in the never-ending pixel-wars. These small pixels just are not very good. The truth of the matter is that the most important feature of digital cameras in the last five years is the automatic motion control to stabilize the image on the sensor along with some very sophisticated image processing. All the rest has been hype and some "cool" design. What is the future for digital imaging and what will drive growth of camera sales (not counting the cell phone cameras which totally dominate the market in terms of camera sales) and more importantly after sales profits? Well sit in on the Dark Side of Color and find out what is being done to increase the after sales profits and don't be surprised if has been done long ago in some basement lab of a photographic company and of course, before its time.

Face recognition via Gabor and convolutional neural network

NASA Astrophysics Data System (ADS)

Lu, Tongwei; Wu, Menglu; Lu, Tao

2018-04-01

In recent years, the powerful feature learning and classification ability of convolutional neural network have attracted widely attention. Compared with the deep learning, the traditional machine learning algorithm has a good explanatory which deep learning does not have. Thus, In this paper, we propose a method to extract the feature of the traditional algorithm as the input of convolution neural network. In order to reduce the complexity of the network, the kernel function of Gabor wavelet is used to extract the feature from different position, frequency and direction of target image. It is sensitive to edge of image which can provide good direction and scale selection. The extraction of the image from eight directions on a scale are as the input of network that we proposed. The network have the advantage of weight sharing and local connection and texture feature of the input image can reduce the influence of facial expression, gesture and illumination. At the same time, we introduced a layer which combined the results of the pooling and convolution can extract deeper features. The training network used the open source caffe framework which is beneficial to feature extraction. The experiment results of the proposed method proved that the network structure effectively overcame the barrier of illumination and had a good robustness as well as more accurate and rapid than the traditional algorithm.

Sequential Syndrome Decoding of Convolutional Codes

NASA Technical Reports Server (NTRS)

Reed, I. S.; Truong, T. K.

1984-01-01

The algebraic structure of convolutional codes are reviewed and sequential syndrome decoding is applied to those codes. These concepts are then used to realize by example actual sequential decoding, using the stack algorithm. The Fano metric for use in sequential decoding is modified so that it can be utilized to sequentially find the minimum weight error sequence.

Sensitivity of landscape metrics to pixel size

Treesearch

J. D. Wickham; K. H. Riitters

1995-01-01

Analysis of diversity and evenness metrics using land cover data are becoming formalized in landscape ecology. Diversity and evenness metrics are dependent on the pixel size (scale) over which the data are collected. Aerial photography was interpreted for land cover and converted into four raster data sets with 4, 12, 28, and 80 m pixel sizes, representing pixel sizes...

Moving from pixel to object scale when inverting radiative transfer models for quantitative estimation of biophysical variables in vegetation (Invited)

NASA Astrophysics Data System (ADS)

Atzberger, C.

2013-12-01

The robust and accurate retrieval of vegetation biophysical variables using RTM is seriously hampered by the ill-posedness of the inverse problem. The contribution presents our object-based inversion approach and evaluate it against measured data. The proposed method takes advantage of the fact that nearby pixels are generally more similar than those at a larger distance. For example, within a given vegetation patch, nearby pixels often share similar leaf angular distributions. This leads to spectral co-variations in the n-dimensional spectral features space, which can be used for regularization purposes. Using a set of leaf area index (LAI) measurements (n=26) acquired over alfalfa, sugar beet and garlic crops of the Barrax test site (Spain), it is demonstrated that the proposed regularization using neighbourhood information yields more accurate results compared to the traditional pixel-based inversion. Principle of the ill-posed inverse problem and the proposed solution illustrated in the red-nIR feature space using (PROSAIL). [A] spectral trajectory ('soil trajectory') obtained for one leaf angle (ALA) and one soil brightness (αsoil), when LAI varies between 0 and 10, [B] 'soil trajectories' for 5 soil brightness values and three leaf angles, [C] ill-posed inverse problem: different combinations of ALA × αsoil yield an identical crossing point, [D] object-based RTM inversion; only one 'soil trajectory' fits all nine pixelswithin a gliding (3×3) window. The black dots (plus the rectangle=central pixel) represent the hypothetical position of nine pixels within a 3×3 (gliding) window. Assuming that over short distances (× 1 pixel) variations in soil brightness can be neglected, the proposed object-based inversion searches for one common set of ALA × αsoil so that the resulting 'soil trajectory' best fits the nine measured pixels. Ground measured vs. retrieved LAI values for three crops. Left: proposed object-based approach. Right: pixel-based inversion

Fabrication of X-ray Microcalorimeter Focal Planes Composed of Two Distinct Pixel Types.

PubMed

Wassell, E J; Adams, J S; Bandler, S R; Betancourt-Martinez, G L; Chiao, M P; Chang, M P; Chervenak, J A; Datesman, A M; Eckart, M E; Ewin, A J; Finkbeiner, F M; Ha, J Y; Kelley, R; Kilbourne, C A; Miniussi, A R; Sakai, K; Porter, F; Sadleir, J E; Smith, S J; Wakeham, N A; Yoon, W

2017-06-01

We are developing superconducting transition-edge sensor (TES) microcalorimeter focal planes for versatility in meeting specifications of X-ray imaging spectrometers including high count-rate, high energy resolution, and large field-of-view. In particular, a focal plane composed of two sub-arrays: one of fine-pitch, high count-rate devices and the other of slower, larger pixels with similar energy resolution, offers promise for the next generation of astrophysics instruments, such as the X-ray Integral Field Unit (X-IFU) instrument on the European Space Agency's Athena mission. We have based the sub-arrays of our current design on successful pixel designs that have been demonstrated separately. Pixels with an all gold X-ray absorber on 50 and 75 micron scales where the Mo/Au TES sits atop a thick metal heatsinking layer have shown high resolution and can accommodate high count-rates. The demonstrated larger pixels use a silicon nitride membrane for thermal isolation, thinner Au and an added bismuth layer in a 250 micron square absorber. To tune the parameters of each sub-array requires merging the fabrication processes of the two detector types. We present the fabrication process for dual production of different X-ray absorbers on the same substrate, thick Au on the small pixels and thinner Au with a Bi capping layer on the larger pixels to tune their heat capacities. The process requires multiple electroplating and etching steps, but the absorbers are defined in a single ion milling step. We demonstrate methods for integrating heatsinking of the two types of pixel into the same focal plane consistent with the requirements for each sub-array, including the limiting of thermal crosstalk. We also discuss fabrication process modifications for tuning the intrinsic transition temperature (T c ) of the bilayers for the different device types through variation of the bilayer thicknesses. The latest results on these "hybrid" arrays will be presented.

Fabrication of X-ray Microcalorimeter Focal Planes Composed of Two Distinct Pixel Types

PubMed Central

Wassell, E. J.; Adams, J. S.; Bandler, S. R.; Betancourt-Martinez, G. L.; Chiao, M. P.; Chang, M. P.; Chervenak, J. A.; Datesman, A. M.; Eckart, M. E.; Ewin, A. J.; Finkbeiner, F. M.; Ha, J. Y.; Kelley, R.; Kilbourne, C. A.; Miniussi, A. R.; Sakai, K.; Porter, F.; Sadleir, J. E.; Smith, S. J.; Wakeham, N. A.; Yoon, W.

2017-01-01

We are developing superconducting transition-edge sensor (TES) microcalorimeter focal planes for versatility in meeting specifications of X-ray imaging spectrometers including high count-rate, high energy resolution, and large field-of-view. In particular, a focal plane composed of two sub-arrays: one of fine-pitch, high count-rate devices and the other of slower, larger pixels with similar energy resolution, offers promise for the next generation of astrophysics instruments, such as the X-ray Integral Field Unit (X-IFU) instrument on the European Space Agency’s Athena mission. We have based the sub-arrays of our current design on successful pixel designs that have been demonstrated separately. Pixels with an all gold X-ray absorber on 50 and 75 micron scales where the Mo/Au TES sits atop a thick metal heatsinking layer have shown high resolution and can accommodate high count-rates. The demonstrated larger pixels use a silicon nitride membrane for thermal isolation, thinner Au and an added bismuth layer in a 250 micron square absorber. To tune the parameters of each sub-array requires merging the fabrication processes of the two detector types. We present the fabrication process for dual production of different X-ray absorbers on the same substrate, thick Au on the small pixels and thinner Au with a Bi capping layer on the larger pixels to tune their heat capacities. The process requires multiple electroplating and etching steps, but the absorbers are defined in a single ion milling step. We demonstrate methods for integrating heatsinking of the two types of pixel into the same focal plane consistent with the requirements for each sub-array, including the limiting of thermal crosstalk. We also discuss fabrication process modifications for tuning the intrinsic transition temperature (Tc) of the bilayers for the different device types through variation of the bilayer thicknesses. The latest results on these “hybrid” arrays will be presented. PMID:28804229

Automatic Skin Lesion Segmentation Using Deep Fully Convolutional Networks With Jaccard Distance.

PubMed

Yuan, Yading; Chao, Ming; Lo, Yeh-Chi

2017-09-01

Automatic skin lesion segmentation in dermoscopic images is a challenging task due to the low contrast between lesion and the surrounding skin, the irregular and fuzzy lesion borders, the existence of various artifacts, and various imaging acquisition conditions. In this paper, we present a fully automatic method for skin lesion segmentation by leveraging 19-layer deep convolutional neural networks that is trained end-to-end and does not rely on prior knowledge of the data. We propose a set of strategies to ensure effective and efficient learning with limited training data. Furthermore, we design a novel loss function based on Jaccard distance to eliminate the need of sample re-weighting, a typical procedure when using cross entropy as the loss function for image segmentation due to the strong imbalance between the number of foreground and background pixels. We evaluated the effectiveness, efficiency, as well as the generalization capability of the proposed framework on two publicly available databases. One is from ISBI 2016 skin lesion analysis towards melanoma detection challenge, and the other is the PH2 database. Experimental results showed that the proposed method outperformed other state-of-the-art algorithms on these two databases. Our method is general enough and only needs minimum pre- and post-processing, which allows its adoption in a variety of medical image segmentation tasks.

A Fast Numerical Method for Max-Convolution and the Application to Efficient Max-Product Inference in Bayesian Networks.

PubMed

Serang, Oliver

2015-08-01

Observations depending on sums of random variables are common throughout many fields; however, no efficient solution is currently known for performing max-product inference on these sums of general discrete distributions (max-product inference can be used to obtain maximum a posteriori estimates). The limiting step to max-product inference is the max-convolution problem (sometimes presented in log-transformed form and denoted as "infimal convolution," "min-convolution," or "convolution on the tropical semiring"), for which no O(k log(k)) method is currently known. Presented here is an O(k log(k)) numerical method for estimating the max-convolution of two nonnegative vectors (e.g., two probability mass functions), where k is the length of the larger vector. This numerical max-convolution method is then demonstrated by performing fast max-product inference on a convolution tree, a data structure for performing fast inference given information on the sum of n discrete random variables in O(nk log(nk)log(n)) steps (where each random variable has an arbitrary prior distribution on k contiguous possible states). The numerical max-convolution method can be applied to specialized classes of hidden Markov models to reduce the runtime of computing the Viterbi path from nk(2) to nk log(k), and has potential application to the all-pairs shortest paths problem.

Plants and pixels: Comparing phenologies from the ground and from space (Invited)

NASA Astrophysics Data System (ADS)

Rutishauser, T.; Stoekli, R.; Jeanneret, F.; Peñuelas, J.

2010-12-01

Changes in the seasonality of life cycles of plants as recorded in phenological observations have been widely analysed at the species level with data available for many decades back in time. At the same time, seasonality changes in satellite-based observations and prognostic phenology models comprise information at the pixel-size or landscape scale. Change analysis of satellite-based records is restricted due to relatively short satellite records that further include gaps while model-based analyses are biased due to current model deficiencies. At 30 selected sites across Europe, we analysed three different sources of plant seasonality during the 1971-2000 period. Data consisted of (1) species-specific development stages of flowering and leave-out with different species observed at each site. (2) We used a synthetic phenological metric that integrates the common interannual phenological signal across all species at one site. (3) We estimated daily Leaf Area Index with a prognostic phenology model. The prior uncertainties of the model’s empirical parameter space are constrained by assimilating the Fraction of Photosynthetically Active Radiation absorbed by vegetation (FPAR) and Leaf Area Index (LAI) from the MODerate Resolution Imaging Spectroradiometer (MODIS). We extracted the day of year when the 25%, 50% and 75% thresholds were passed each spring. The question arises how the three phenological signals compare and correlate across climate zones in Europe. Is there a match between single species observations, species-based ground-observed metrics and the landscape-scale prognostic model? Are there single key-species across Europe that best represent a landscape scale measure from the prognostic model? Can one source substitute another and serve as proxy-data? What can we learn from potential mismatches? Focusing on changes in spring this contribution presents first results of an ongoing comparison study from a number of European test sites that will be extended

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.

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.

[Application of numerical convolution in in vivo/in vitro correlation research].

PubMed

Yue, Peng

2009-01-01

This paper introduced the conception and principle of in vivo/in vitro correlation (IVIVC) and convolution/deconvolution methods, and elucidated in details the convolution strategy and method for calculating the in vivo absorption performance of the pharmaceutics according to the their pharmacokinetic data in Excel, then put the results forward to IVIVC research. Firstly, the pharmacokinetic data ware fitted by mathematical software to make up the lost points. Secondly, the parameters of the optimal fitted input function were defined by trail-and-error method according to the convolution principle in Excel under the hypothesis that all the input functions fit the Weibull functions. Finally, the IVIVC between in vivo input function and the in vitro dissolution was studied. In the examples, not only the application of this method was demonstrated in details but also its simplicity and effectiveness were proved by comparing with the compartment model method and deconvolution method. It showed to be a powerful tool for IVIVC research.

Resting State fMRI Functional Connectivity-Based Classification Using a Convolutional Neural Network Architecture

PubMed Central

Meszlényi, Regina J.; Buza, Krisztian; Vidnyánszky, Zoltán

2017-01-01

Machine learning techniques have become increasingly popular in the field of resting state fMRI (functional magnetic resonance imaging) network based classification. However, the application of convolutional networks has been proposed only very recently and has remained largely unexplored. In this paper we describe a convolutional neural network architecture for functional connectome classification called connectome-convolutional neural network (CCNN). Our results on simulated datasets and a publicly available dataset for amnestic mild cognitive impairment classification demonstrate that our CCNN model can efficiently distinguish between subject groups. We also show that the connectome-convolutional network is capable to combine information from diverse functional connectivity metrics and that models using a combination of different connectivity descriptors are able to outperform classifiers using only one metric. From this flexibility follows that our proposed CCNN model can be easily adapted to a wide range of connectome based classification or regression tasks, by varying which connectivity descriptor combinations are used to train the network. PMID:29089883

Resting State fMRI Functional Connectivity-Based Classification Using a Convolutional Neural Network Architecture.

PubMed

Meszlényi, Regina J; Buza, Krisztian; Vidnyánszky, Zoltán

2017-01-01

Machine learning techniques have become increasingly popular in the field of resting state fMRI (functional magnetic resonance imaging) network based classification. However, the application of convolutional networks has been proposed only very recently and has remained largely unexplored. In this paper we describe a convolutional neural network architecture for functional connectome classification called connectome-convolutional neural network (CCNN). Our results on simulated datasets and a publicly available dataset for amnestic mild cognitive impairment classification demonstrate that our CCNN model can efficiently distinguish between subject groups. We also show that the connectome-convolutional network is capable to combine information from diverse functional connectivity metrics and that models using a combination of different connectivity descriptors are able to outperform classifiers using only one metric. From this flexibility follows that our proposed CCNN model can be easily adapted to a wide range of connectome based classification or regression tasks, by varying which connectivity descriptor combinations are used to train the network.

Robust hepatic vessel segmentation using multi deep convolution network

NASA Astrophysics Data System (ADS)

Kitrungrotsakul, Titinunt; Han, Xian-Hua; Iwamoto, Yutaro; Foruzan, Amir Hossein; Lin, Lanfen; Chen, Yen-Wei

2017-03-01

Extraction of blood vessels of the organ is a challenging task in the area of medical image processing. It is really difficult to get accurate vessel segmentation results even with manually labeling by human being. The difficulty of vessels segmentation is the complicated structure of blood vessels and its large variations that make them hard to recognize. In this paper, we present deep artificial neural network architecture to automatically segment the hepatic vessels from computed tomography (CT) image. We proposed novel deep neural network (DNN) architecture for vessel segmentation from a medical CT volume, which consists of three deep convolution neural networks to extract features from difference planes of CT data. The three networks have share features at the first convolution layer but will separately learn their own features in the second layer. All three networks will join again at the top layer. To validate effectiveness and efficiency of our proposed method, we conduct experiments on 12 CT volumes which training data are randomly generate from 5 CT volumes and 7 using for test. Our network can yield an average dice coefficient 0.830, while 3D deep convolution neural network can yield around 0.7 and multi-scale can yield only 0.6.

Sub-pixel image classification for forest types in East Texas

NASA Astrophysics Data System (ADS)

Westbrook, Joey

Sub-pixel classification is the extraction of information about the proportion of individual materials of interest within a pixel. Landcover classification at the sub-pixel scale provides more discrimination than traditional per-pixel multispectral classifiers for pixels where the material of interest is mixed with other materials. It allows for the un-mixing of pixels to show the proportion of each material of interest. The materials of interest for this study are pine, hardwood, mixed forest and non-forest. The goal of this project was to perform a sub-pixel classification, which allows a pixel to have multiple labels, and compare the result to a traditional supervised classification, which allows a pixel to have only one label. The satellite image used was a Landsat 5 Thematic Mapper (TM) scene of the Stephen F. Austin Experimental Forest in Nacogdoches County, Texas and the four cover type classes are pine, hardwood, mixed forest and non-forest. Once classified, a multi-layer raster datasets was created that comprised four raster layers where each layer showed the percentage of that cover type within the pixel area. Percentage cover type maps were then produced and the accuracy of each was assessed using a fuzzy error matrix for the sub-pixel classifications, and the results were compared to the supervised classification in which a traditional error matrix was used. The overall accuracy of the sub-pixel classification using the aerial photo for both training and reference data had the highest (65% overall) out of the three sub-pixel classifications. This was understandable because the analyst can visually observe the cover types actually on the ground for training data and reference data, whereas using the FIA (Forest Inventory and Analysis) plot data, the analyst must assume that an entire pixel contains the exact percentage of a cover type found in a plot. An increase in accuracy was found after reclassifying each sub-pixel classification from nine classes

Method for fabricating pixelated silicon device cells

DOEpatents

Nielson, Gregory N.; Okandan, Murat; Cruz-Campa, Jose Luis; Nelson, Jeffrey S.; Anderson, Benjamin John

2015-08-18

A method, apparatus and system for flexible, ultra-thin, and high efficiency pixelated silicon or other semiconductor photovoltaic solar cell array fabrication is disclosed. A structure and method of creation for a pixelated silicon or other semiconductor photovoltaic solar cell array with interconnects is described using a manufacturing method that is simplified compared to previous versions of pixelated silicon photovoltaic cells that require more microfabrication steps.

A comparison of the convolution and TMR10 treatment planning algorithms for Gamma Knife® radiosurgery

PubMed Central

Wright, Gavin; Harrold, Natalie; Bownes, Peter

2018-01-01

Aims To compare the accuracies of the convolution and TMR10 Gamma Knife treatment planning algorithms, and assess the impact upon clinical practice of implementing convolution-based treatment planning. Methods Doses calculated by both algorithms were compared against ionisation chamber measurements in homogeneous and heterogeneous phantoms. Relative dose distributions calculated by both algorithms were compared against film-derived 2D isodose plots in a heterogeneous phantom, with distance-to-agreement (DTA) measured at the 80%, 50% and 20% isodose levels. A retrospective planning study compared 19 clinically acceptable metastasis convolution plans against TMR10 plans with matched shot times, allowing novel comparison of true dosimetric parameters rather than total beam-on-time. Gamma analysis and dose-difference analysis were performed on each pair of dose distributions. Results Both algorithms matched point dose measurement within ±1.1% in homogeneous conditions. Convolution provided superior point-dose accuracy in the heterogeneous phantom (-1.1% v 4.0%), with no discernible differences in relative dose distribution accuracy. In our study convolution-calculated plans yielded D99% 6.4% (95% CI:5.5%-7.3%,p<0.001) less than shot matched TMR10 plans. For gamma passing criteria 1%/1mm, 16% of targets had passing rates >95%. The range of dose differences in the targets was 0.2-4.6Gy. Conclusions Convolution provides superior accuracy versus TMR10 in heterogeneous conditions. Implementing convolution would result in increased target doses therefore its implementation may require a revaluation of prescription doses. PMID:29657896

Dimensionality-varied convolutional neural network for spectral-spatial classification of hyperspectral data

NASA Astrophysics Data System (ADS)

Liu, Wanjun; Liang, Xuejian; Qu, Haicheng

2017-11-01

Hyperspectral image (HSI) classification is one of the most popular topics in remote sensing community. Traditional and deep learning-based classification methods were proposed constantly in recent years. In order to improve the classification accuracy and robustness, a dimensionality-varied convolutional neural network (DVCNN) was proposed in this paper. DVCNN was a novel deep architecture based on convolutional neural network (CNN). The input of DVCNN was a set of 3D patches selected from HSI which contained spectral-spatial joint information. In the following feature extraction process, each patch was transformed into some different 1D vectors by 3D convolution kernels, which were able to extract features from spectral-spatial data. The rest of DVCNN was about the same as general CNN and processed 2D matrix which was constituted by by all 1D data. So that the DVCNN could not only extract more accurate and rich features than CNN, but also fused spectral-spatial information to improve classification accuracy. Moreover, the robustness of network on water-absorption bands was enhanced in the process of spectral-spatial fusion by 3D convolution, and the calculation was simplified by dimensionality varied convolution. Experiments were performed on both Indian Pines and Pavia University scene datasets, and the results showed that the classification accuracy of DVCNN improved by 32.87% on Indian Pines and 19.63% on Pavia University scene than spectral-only CNN. The maximum accuracy improvement of DVCNN achievement was 13.72% compared with other state-of-the-art HSI classification methods, and the robustness of DVCNN on water-absorption bands noise was demonstrated.

Performance of Serially Concatenated Convolutional Codes with Binary Modulation in AWGN and Noise Jamming over Rayleigh Fading Channels

DTIC Science & Technology

2001-09-01

Rate - compatible punctured convolutional codes (RCPC codes ) and their applications,” IEEE...ABSTRACT In this dissertation, the bit error rates for serially concatenated convolutional codes (SCCC) for both BPSK and DPSK modulation with...INTENTIONALLY LEFT BLANK i EXECUTIVE SUMMARY In this dissertation, the bit error rates of serially concatenated convolutional codes

Cascaded K-means convolutional feature learner and its application to face recognition

NASA Astrophysics Data System (ADS)

Zhou, Daoxiang; Yang, Dan; Zhang, Xiaohong; Huang, Sheng; Feng, Shu

2017-09-01

Currently, considerable efforts have been devoted to devise image representation. However, handcrafted methods need strong domain knowledge and show low generalization ability, and conventional feature learning methods require enormous training data and rich parameters tuning experience. A lightened feature learner is presented to solve these problems with application to face recognition, which shares similar topology architecture as a convolutional neural network. Our model is divided into three components: cascaded convolution filters bank learning layer, nonlinear processing layer, and feature pooling layer. Specifically, in the filters learning layer, we use K-means to learn convolution filters. Features are extracted via convoluting images with the learned filters. Afterward, in the nonlinear processing layer, hyperbolic tangent is employed to capture the nonlinear feature. In the feature pooling layer, to remove the redundancy information and incorporate the spatial layout, we exploit multilevel spatial pyramid second-order pooling technique to pool the features in subregions and concatenate them together as the final representation. Extensive experiments on four representative datasets demonstrate the effectiveness and robustness of our model to various variations, yielding competitive recognition results on extended Yale B and FERET. In addition, our method achieves the best identification performance on AR and labeled faces in the wild datasets among the comparative methods.

The Belle II DEPFET pixel detector

NASA Astrophysics Data System (ADS)

Lütticke, F.

2013-02-01

The existing Japanese Flavour Factory (KEKB) is currently being upgraded and is foreseen to be comissioned by 2014. The new e+e- collider (SuperKEKB) will have an instantaneous luminosity of 8 × 1035cm-2s-1, 40 times higher than the current world record set by KEKB. In order to handle the increased event rate and the higher background and provide high data quality, the Belle detector is upgraded to Belle II. The increased particle rate requires a new vertex pixel detector with high granularity. This silicon detector will be based on DEPFET technology and will consist of two layers of active pixel sensors. By integrating a field effect transistor into every pixel on top of a fully depleted bulk, the DEPFET technology combines detection and in-pixel amplification. This technology allows good signal to noise performance with a very low material budget.

The HEXITEC Hard X-Ray Pixelated CdTe Imager for Fast Solar Observations

NASA Technical Reports Server (NTRS)

Baumgartner, Wayne H.; Christe, Steven D.; Ryan, Daniel; Inglis, Andrew R.; Shih, Albert Y.; Gregory, Kyle; Wilson, Matt; Seller, Paul; Gaskin, Jessica; Wilson-Hodge, Colleen

2016-01-01

There is an increasing demand in solar and astrophysics for high resolution X-ray spectroscopic imaging. Such observations would present ground breaking opportunities to study the poorly understood high energy processes in our solar system and beyond, such as solar flares, X-ray binaries, and active galactic nuclei. However, such observations require a new breed of solid state detectors sensitive to high energy X-rays with fine independent pixels to sub-sample the point spread function (PSF) of the X-ray optics. For solar observations in particular, they must also be capable of handling very high count rates as photon fluxes from solar flares often cause pile up and saturation in present generation detectors. The Rutherford Appleton Laboratory (RAL) has recently developed a new cadmium telluride (CdTe) detector system, called HEXITEC (High Energy X-ray Imaging Technology). It is an 80 x 80 array of 250 micron independent pixels sensitive in the 2-200 keV band and capable of a high full frame read out rate of 10 kHz. HEXITEC provides the smallest independently read out CdTe pixels currently available, and are well matched to the few arcsecond PSF produced by current and next generation hard X-ray focusing optics. NASA's Goddard and Marshall Space Flight Centers are collaborating with RAL to develop these detectors for use on future space borne hard X-ray focusing telescopes. We show the latest results on HEXITEC's imaging capability, energy resolution, high read out rate, and reveal it to be ideal for such future instruments.

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.

Using convolutional decoding to improve time delay and phase estimation in digital communications

DOEpatents

Ormesher, Richard C [Albuquerque, NM; Mason, John J [Albuquerque, NM

2010-01-26

The time delay and/or phase of a communication signal received by a digital communication receiver can be estimated based on a convolutional decoding operation that the communication receiver performs on the received communication signal. If the original transmitted communication signal has been spread according to a spreading operation, a corresponding despreading operation can be integrated into the convolutional decoding operation.

The probabilistic convolution tree: efficient exact Bayesian inference for faster LC-MS/MS protein inference.

PubMed

Serang, Oliver

2014-01-01

Exact Bayesian inference can sometimes be performed efficiently for special cases where a function has commutative and associative symmetry of its inputs (called "causal independence"). For this reason, it is desirable to exploit such symmetry on big data sets. Here we present a method to exploit a general form of this symmetry on probabilistic adder nodes by transforming those probabilistic adder nodes into a probabilistic convolution tree with which dynamic programming computes exact probabilities. A substantial speedup is demonstrated using an illustration example that can arise when identifying splice forms with bottom-up mass spectrometry-based proteomics. On this example, even state-of-the-art exact inference algorithms require a runtime more than exponential in the number of splice forms considered. By using the probabilistic convolution tree, we reduce the runtime to O(k log(k)2) and the space to O(k log(k)) where k is the number of variables joined by an additive or cardinal operator. This approach, which can also be used with junction tree inference, is applicable to graphs with arbitrary dependency on counting variables or cardinalities and can be used on diverse problems and fields like forward error correcting codes, elemental decomposition, and spectral demixing. The approach also trivially generalizes to multiple dimensions.

The Probabilistic Convolution Tree: Efficient Exact Bayesian Inference for Faster LC-MS/MS Protein Inference

PubMed Central

Serang, Oliver

2014-01-01

Exact Bayesian inference can sometimes be performed efficiently for special cases where a function has commutative and associative symmetry of its inputs (called “causal independence”). For this reason, it is desirable to exploit such symmetry on big data sets. Here we present a method to exploit a general form of this symmetry on probabilistic adder nodes by transforming those probabilistic adder nodes into a probabilistic convolution tree with which dynamic programming computes exact probabilities. A substantial speedup is demonstrated using an illustration example that can arise when identifying splice forms with bottom-up mass spectrometry-based proteomics. On this example, even state-of-the-art exact inference algorithms require a runtime more than exponential in the number of splice forms considered. By using the probabilistic convolution tree, we reduce the runtime to and the space to where is the number of variables joined by an additive or cardinal operator. This approach, which can also be used with junction tree inference, is applicable to graphs with arbitrary dependency on counting variables or cardinalities and can be used on diverse problems and fields like forward error correcting codes, elemental decomposition, and spectral demixing. The approach also trivially generalizes to multiple dimensions. PMID:24626234

Characterization of pixelated TlBr detectors with Tl electrodes

NASA Astrophysics Data System (ADS)

Hitomi, Keitaro; Onodera, Toshiyuki; Kim, Seong-Yun; Shoji, Tadayoshi; Ishii, Keizo

2014-05-01

A 4.36-mm-thick pixelated thallium bromide (TlBr) detector with Tl electrodes was fabricated from a crystal grown by the traveling molten zone method using zone-purified material. The detector had four 1×1 mm2 pixelated anodes. The detector performance was characterized at room temperature. The mobility-lifetime products of electrons for each pixel of the TlBr detector were measured to be >2.8×10-3 cm2/V. The four pixelated anodes of the detector exhibited energy resolutions of 1.5-1.8% full width at half maximum (FWHM) for 662-keV gamma rays for single-pixel events with the depth correction method. An energy resolution of 4.5% FWHM for 662-keV gamma rays was obtained from a reconstructed energy spectrum using two-pixel events from the two pixelated anodes on the detector.

Acral melanoma detection using a convolutional neural network for dermoscopy images.

PubMed

Yu, Chanki; Yang, Sejung; Kim, Wonoh; Jung, Jinwoong; Chung, Kee-Yang; Lee, Sang Wook; Oh, Byungho

2018-01-01

Acral melanoma is the most common type of melanoma in Asians, and usually results in a poor prognosis due to late diagnosis. We applied a convolutional neural network to dermoscopy images of acral melanoma and benign nevi on the hands and feet and evaluated its usefulness for the early diagnosis of these conditions. A total of 724 dermoscopy images comprising acral melanoma (350 images from 81 patients) and benign nevi (374 images from 194 patients), and confirmed by histopathological examination, were analyzed in this study. To perform the 2-fold cross validation, we split them into two mutually exclusive subsets: half of the total image dataset was selected for training and the rest for testing, and we calculated the accuracy of diagnosis comparing it with the dermatologist's and non-expert's evaluation. The accuracy (percentage of true positive and true negative from all images) of the convolutional neural network was 83.51% and 80.23%, which was higher than the non-expert's evaluation (67.84%, 62.71%) and close to that of the expert (81.08%, 81.64%). Moreover, the convolutional neural network showed area-under-the-curve values like 0.8, 0.84 and Youden's index like 0.6795, 0.6073, which were similar score with the expert. Although further data analysis is necessary to improve their accuracy, convolutional neural networks would be helpful to detect acral melanoma from dermoscopy images of the hands and feet.

Charge Gain, Voltage Gain, and Node Capacitance of the SAPHIRA Detector Pixel by Pixel

NASA Astrophysics Data System (ADS)

Pastrana, Izabella M.; Hall, Donald N. B.; Baker, Ian M.; Jacobson, Shane M.; Goebel, Sean B.

2018-01-01

The University of Hawai`i Institute for Astronomy has partnered with Leonardo (formerly Selex) in the development of HgCdTe linear mode avalanche photodiode (L-APD) SAPHIRA detectors. The SAPHIRA (Selex Avalanche Photodiode High-speed Infra-Red Array) is ideally suited for photon-starved astronomical observations, particularly near infrared (NIR) adaptive optics (AO) wave-front sensing. I have measured the stability, and linearity with current, of a 1.7-um (10% spectral bandpass) infrared light emitting diode (IR LED) used to illuminate the SAPHIRA and have then utilized this source to determine the charge gain (in e-/ADU), voltage gain (in uV/ADU), and node capacitance (in fF) for each pixel of the 320x256@24um SAPHIRA. These have previously only been averages over some sub-array. Determined from the ratio of the temporal averaged signal level to variance under constant 1.7-um LED illumination, I present the charge gain pixel-by-pixel in a 64x64 sub-array at the center of the active area of the SAPHIRA (analyzed separately as four 32x32 sub-arrays) to be about 1.6 e-/ADU (σ=0.5 e-/ADU). Additionally, the standard technique of varying the pixel reset voltage (PRV) in 10 mV increments and recording output frames for the same 64x64 subarray found the voltage gain per pixel to be about 11.7 uV/ADU (σ=0.2 uV/ADU). Finally, node capacitance was found to be approximately 23 fF (σ=6 fF) utilizing the aforementioned charge and voltage gain measurements. I further discuss the linearity measurements of the 1.7-um LED used in the charge gain characterization procedure.

Method for hyperspectral imagery exploitation and pixel spectral unmixing

NASA Technical Reports Server (NTRS)

Lin, Ching-Fang (Inventor)

2003-01-01

An efficiently hybrid approach to exploit hyperspectral imagery and unmix spectral pixels. This hybrid approach uses a genetic algorithm to solve the abundance vector for the first pixel of a hyperspectral image cube. This abundance vector is used as initial state in a robust filter to derive the abundance estimate for the next pixel. By using Kalman filter, the abundance estimate for a pixel can be obtained in one iteration procedure which is much fast than genetic algorithm. The output of the robust filter is fed to genetic algorithm again to derive accurate abundance estimate for the current pixel. The using of robust filter solution as starting point of the genetic algorithm speeds up the evolution of the genetic algorithm. After obtaining the accurate abundance estimate, the procedure goes to next pixel, and uses the output of genetic algorithm as the previous state estimate to derive abundance estimate for this pixel using robust filter. And again use the genetic algorithm to derive accurate abundance estimate efficiently based on the robust filter solution. This iteration continues until pixels in a hyperspectral image cube end.

Nonlinearity and pixel shifting effects in HXRG infrared detectors

NASA Astrophysics Data System (ADS)

Plazas, A. A.; Shapiro, C.; Smith, R.; Rhodes, J.; Huff, E.

2017-04-01

We study the nonlinearity (NL) in the conversion from charge to voltage in infrared detectors (HXRG) for use in precision astronomy. We present laboratory measurements of the NL function of a H2RG detector and discuss the accuracy to which it would need to be calibrated in future space missions to perform cosmological measurements through the weak gravitational lensing technique. In addition, we present an analysis of archival data from the infrared H1RG detector of the Wide Field Camera 3 in the Hubble Space Telescope that provides evidence consistent with the existence of a sensor effect analogous to the ``brighter-fatter'' effect found in Charge-Coupled Devices. We propose a model in which this effect could be understood as shifts in the effective pixel boundaries, and discuss prospects of laboratory measurements to fully characterize this effect.

No-reference image quality assessment based on statistics of convolution feature maps

NASA Astrophysics Data System (ADS)

Lv, Xiaoxin; Qin, Min; Chen, Xiaohui; Wei, Guo

2018-04-01

We propose a Convolutional Feature Maps (CFM) driven approach to accurately predict image quality. Our motivation bases on the finding that the Nature Scene Statistic (NSS) features on convolution feature maps are significantly sensitive to distortion degree of an image. In our method, a Convolutional Neural Network (CNN) is trained to obtain kernels for generating CFM. We design a forward NSS layer which performs on CFM to better extract NSS features. The quality aware features derived from the output of NSS layer is effective to describe the distortion type and degree an image suffered. Finally, a Support Vector Regression (SVR) is employed in our No-Reference Image Quality Assessment (NR-IQA) model to predict a subjective quality score of a distorted image. Experiments conducted on two public databases demonstrate the promising performance of the proposed method is competitive to state of the art NR-IQA methods.

A new 9T global shutter pixel with CDS technique

NASA Astrophysics Data System (ADS)

Liu, Yang; Ma, Cheng; Zhou, Quan; Wang, Xinyang

2015-04-01

Benefiting from motion blur free, Global shutter pixel is very widely used in the design of CMOS image sensors for high speed applications such as motion vision, scientifically inspection, etc. In global shutter sensors, all pixel signal information needs to be stored in the pixel first and then waiting for readout. For higher frame rate, we need very fast operation of the pixel array. There are basically two ways for the in pixel signal storage, one is in charge domain, such as the one shown in [1], this needs complicated process during the pixel fabrication. The other one is in voltage domain, one example is the one in [2], this pixel is based on the 4T PPD technology and normally the driving of the high capacitive transfer gate limits the speed of the array operation. In this paper we report a new 9T global shutter pixel based on 3-T partially pinned photodiode (PPPD) technology. It incorporates three in-pixel storage capacitors allowing for correlated double sampling (CDS) and pipeline operation of the array (pixel exposure during the readout of the array). Only two control pulses are needed for all the pixels at the end of exposure which allows high speed exposure control.

Automatic segmentation of the clinical target volume and organs at risk in the planning CT for rectal cancer using deep dilated convolutional neural networks.

PubMed

Men, Kuo; Dai, Jianrong; Li, Yexiong

2017-12-01

Delineation of the clinical target volume (CTV) and organs at risk (OARs) is very important for radiotherapy but is time-consuming and prone to inter-observer variation. Here, we proposed a novel deep dilated convolutional neural network (DDCNN)-based method for fast and consistent auto-segmentation of these structures. Our DDCNN method was an end-to-end architecture enabling fast training and testing. Specifically, it employed a novel multiple-scale convolutional architecture to extract multiple-scale context features in the early layers, which contain the original information on fine texture and boundaries and which are very useful for accurate auto-segmentation. In addition, it enlarged the receptive fields of dilated convolutions at the end of networks to capture complementary context features. Then, it replaced the fully connected layers with fully convolutional layers to achieve pixel-wise segmentation. We used data from 278 patients with rectal cancer for evaluation. The CTV and OARs were delineated and validated by senior radiation oncologists in the planning computed tomography (CT) images. A total of 218 patients chosen randomly were used for training, and the remaining 60 for validation. The Dice similarity coefficient (DSC) was used to measure segmentation accuracy. Performance was evaluated on segmentation of the CTV and OARs. In addition, the performance of DDCNN was compared with that of U-Net. The proposed DDCNN method outperformed the U-Net for all segmentations, and the average DSC value of DDCNN was 3.8% higher than that of U-Net. Mean DSC values of DDCNN were 87.7% for the CTV, 93.4% for the bladder, 92.1% for the left femoral head, 92.3% for the right femoral head, 65.3% for the intestine, and 61.8% for the colon. The test time was 45 s per patient for segmentation of all the CTV, bladder, left and right femoral heads, colon, and intestine. We also assessed our approaches and results with those in the literature: our system showed superior

Automated red blood cells extraction from holographic images using fully convolutional neural networks.

PubMed

Yi, Faliu; Moon, Inkyu; Javidi, Bahram

2017-10-01

In this paper, we present two models for automatically extracting red blood cells (RBCs) from RBCs holographic images based on a deep learning fully convolutional neural network (FCN) algorithm. The first model, called FCN-1, only uses the FCN algorithm to carry out RBCs prediction, whereas the second model, called FCN-2, combines the FCN approach with the marker-controlled watershed transform segmentation scheme to achieve RBCs extraction. Both models achieve good segmentation accuracy. In addition, the second model has much better performance in terms of cell separation than traditional segmentation methods. In the proposed methods, the RBCs phase images are first numerically reconstructed from RBCs holograms recorded with off-axis digital holographic microscopy. Then, some RBCs phase images are manually segmented and used as training data to fine-tune the FCN. Finally, each pixel in new input RBCs phase images is predicted into either foreground or background using the trained FCN models. The RBCs prediction result from the first model is the final segmentation result, whereas the result from the second model is used as the internal markers of the marker-controlled transform algorithm for further segmentation. Experimental results show that the given schemes can automatically extract RBCs from RBCs phase images and much better RBCs separation results are obtained when the FCN technique is combined with the marker-controlled watershed segmentation algorithm.

Automated red blood cells extraction from holographic images using fully convolutional neural networks

PubMed Central

Yi, Faliu; Moon, Inkyu; Javidi, Bahram

2017-01-01

In this paper, we present two models for automatically extracting red blood cells (RBCs) from RBCs holographic images based on a deep learning fully convolutional neural network (FCN) algorithm. The first model, called FCN-1, only uses the FCN algorithm to carry out RBCs prediction, whereas the second model, called FCN-2, combines the FCN approach with the marker-controlled watershed transform segmentation scheme to achieve RBCs extraction. Both models achieve good segmentation accuracy. In addition, the second model has much better performance in terms of cell separation than traditional segmentation methods. In the proposed methods, the RBCs phase images are first numerically reconstructed from RBCs holograms recorded with off-axis digital holographic microscopy. Then, some RBCs phase images are manually segmented and used as training data to fine-tune the FCN. Finally, each pixel in new input RBCs phase images is predicted into either foreground or background using the trained FCN models. The RBCs prediction result from the first model is the final segmentation result, whereas the result from the second model is used as the internal markers of the marker-controlled transform algorithm for further segmentation. Experimental results show that the given schemes can automatically extract RBCs from RBCs phase images and much better RBCs separation results are obtained when the FCN technique is combined with the marker-controlled watershed segmentation algorithm. PMID:29082078

Enhanced line integral convolution with flow feature detection

DOT National Transportation Integrated Search

1995-01-01

Prepared ca. 1995. The Line Integral Convolution (LIC) method, which blurs white noise textures along a vector field, is an effective way to visualize overall flow patterns in a 2D domain [Cabral & Leedom '93]. The method produces a flow texture imag...

Correction of clipped pixels in color images.

PubMed

Xu, Di; Doutre, Colin; Nasiopoulos, Panos

2011-03-01

Conventional images store a very limited dynamic range of brightness. The true luma in the bright area of such images is often lost due to clipping. When clipping changes the R, G, B color ratios of a pixel, color distortion also occurs. In this paper, we propose an algorithm to enhance both the luma and chroma of the clipped pixels. Our method is based on the strong chroma spatial correlation between clipped pixels and their surrounding unclipped area. After identifying the clipped areas in the image, we partition the clipped areas into regions with similar chroma, and estimate the chroma of each clipped region based on the chroma of its surrounding unclipped region. We correct the clipped R, G, or B color channels based on the estimated chroma and the unclipped color channel(s) of the current pixel. The last step involves smoothing of the boundaries between regions of different clipping scenarios. Both objective and subjective experimental results show that our algorithm is very effective in restoring the color of clipped pixels. © 2011 IEEE

Random On-Board Pixel Sampling (ROPS) X-Ray Camera

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

Wang, Zhehui; Iaroshenko, O.; Li, S.

Recent advances in compressed sensing theory and algorithms offer new possibilities for high-speed X-ray camera design. In many CMOS cameras, each pixel has an independent on-board circuit that includes an amplifier, noise rejection, signal shaper, an analog-to-digital converter (ADC), and optional in-pixel storage. When X-ray images are sparse, i.e., when one of the following cases is true: (a.) The number of pixels with true X-ray hits is much smaller than the total number of pixels; (b.) The X-ray information is redundant; or (c.) Some prior knowledge about the X-ray images exists, sparse sampling may be allowed. Here we first illustratemore » the feasibility of random on-board pixel sampling (ROPS) using an existing set of X-ray images, followed by a discussion about signal to noise as a function of pixel size. Next, we describe a possible circuit architecture to achieve random pixel access and in-pixel storage. The combination of a multilayer architecture, sparse on-chip sampling, and computational image techniques, is expected to facilitate the development and applications of high-speed X-ray camera technology.« less

Enhancement of digital radiography image quality using a convolutional neural network.

PubMed

Sun, Yuewen; Li, Litao; Cong, Peng; Wang, Zhentao; Guo, Xiaojing

2017-01-01

Digital radiography system is widely used for noninvasive security check and medical imaging examination. However, the system has a limitation of lower image quality in spatial resolution and signal to noise ratio. In this study, we explored whether the image quality acquired by the digital radiography system can be improved with a modified convolutional neural network to generate high-resolution images with reduced noise from the original low-quality images. The experiment evaluated on a test dataset, which contains 5 X-ray images, showed that the proposed method outperformed the traditional methods (i.e., bicubic interpolation and 3D block-matching approach) as measured by peak signal to noise ratio (PSNR) about 1.3 dB while kept highly efficient processing time within one second. Experimental results demonstrated that a residual to residual (RTR) convolutional neural network remarkably improved the image quality of object structural details by increasing the image resolution and reducing image noise. Thus, this study indicated that applying this RTR convolutional neural network system was useful to improve image quality acquired by the digital radiography system.

Simulation study of pixel detector charge digitization

NASA Astrophysics Data System (ADS)

Wang, Fuyue; Nachman, Benjamin; Sciveres, Maurice; Lawrence Berkeley National Laboratory Team

2017-01-01

Reconstruction of tracks from nearly overlapping particles, called Tracking in Dense Environments (TIDE), is an increasingly important component of many physics analyses at the Large Hadron Collider as signatures involving highly boosted jets are investigated. TIDE makes use of the charge distribution inside a pixel cluster to resolve tracks that share one of more of their pixel detector hits. In practice, the pixel charge is discretized using the Time-over-Threshold (ToT) technique. More charge information is better for discrimination, but more challenging for designing and operating the detector. A model of the silicon pixels has been developed in order to study the impact of the precision of the digitized charge distribution on distinguishing multi-particle clusters. The output of the GEANT4-based simulation is used to train neutral networks that predict the multiplicity and location of particles depositing energy inside one cluster of pixels. By studying the multi-particle cluster identification efficiency and position resolution, we quantify the trade-off between the number of ToT bits and low-level tracking inputs. As both ATLAS and CMS are designing upgraded detectors, this work provides guidance for the pixel module designs to meet TIDE needs. Work funded by the China Scholarship Council and the Office of High Energy Physics of the U.S. Department of Energy under contract DE-AC02-05CH11231.

Steganography based on pixel intensity value decomposition

NASA Astrophysics Data System (ADS)

Abdulla, Alan Anwar; Sellahewa, Harin; Jassim, Sabah A.

2014-05-01

This paper focuses on steganography based on pixel intensity value decomposition. A number of existing schemes such as binary, Fibonacci, Prime, Natural, Lucas, and Catalan-Fibonacci (CF) are evaluated in terms of payload capacity and stego quality. A new technique based on a specific representation is proposed to decompose pixel intensity values into 16 (virtual) bit-planes suitable for embedding purposes. The proposed decomposition has a desirable property whereby the sum of all bit-planes does not exceed the maximum pixel intensity value, i.e. 255. Experimental results demonstrate that the proposed technique offers an effective compromise between payload capacity and stego quality of existing embedding techniques based on pixel intensity value decomposition. Its capacity is equal to that of binary and Lucas, while it offers a higher capacity than Fibonacci, Prime, Natural, and CF when the secret bits are embedded in 1st Least Significant Bit (LSB). When the secret bits are embedded in higher bit-planes, i.e., 2nd LSB to 8th Most Significant Bit (MSB), the proposed scheme has more capacity than Natural numbers based embedding. However, from the 6th bit-plane onwards, the proposed scheme offers better stego quality. In general, the proposed decomposition scheme has less effect in terms of quality on pixel value when compared to most existing pixel intensity value decomposition techniques when embedding messages in higher bit-planes.

The unitary convolution approximation for heavy ions

NASA Astrophysics Data System (ADS)

Grande, P. L.; Schiwietz, G.

2002-10-01

The convolution approximation for the impact-parameter dependent energy loss is reviewed with emphasis on the determination of the stopping force for heavy projectiles. In this method, the energy loss in different impact-parameter regions is well determined and interpolated smoothly. The physical inputs of the model are the projectile-screening function (in the case of dressed ions), the electron density and oscillators strengths of the target atoms. Moreover, the convolution approximation, in the perturbative mode (called PCA), yields remarkable agreement with full semi-classical-approximation (SCA) results for bare as well as for screened ions at all impact parameters. In the unitary mode (called UCA), the method contains some higher-order effects (yielding in some cases rather good agreement with full coupled-channel calculations) and approaches the classical regime similar as the Bohr model for large perturbations ( Z/ v≫1). The results are then used to compare with experimental values of the non-equilibrium stopping force as a function of the projectile charge as well as with the equilibrium energy loss under non-aligned and channeling conditions.

Characteristics of Monolithically Integrated InGaAs Active Pixel Imager Array

NASA Technical Reports Server (NTRS)

Kim, Q.; Cunningham, T. J.; Pain, B.; Lange, M. J.; Olsen, G. H.

2000-01-01

Switching and amplifying characteristics of a newly developed monolithic InGaAs Active Pixel Imager Array are presented. The sensor array is fabricated from InGaAs material epitaxially deposited on an InP substrate. It consists of an InGaAs photodiode connected to InP depletion-mode junction field effect transistors (JFETs) for low leakage, low power, and fast control of circuit signal amplifying, buffering, selection, and reset. This monolithically integrated active pixel sensor configuration eliminates the need for hybridization with silicon multiplexer. In addition, the configuration allows the sensor to be front illuminated, making it sensitive to visible as well as near infrared signal radiation. Adapting the existing 1.55 micrometer fiber optical communication technology, this integration will be an ideal system of optoelectronic integration for dual band (Visible/IR) applications near room temperature, for use in atmospheric gas sensing in space, and for target identification on earth. In this paper, two different types of small 4 x 1 test arrays will be described. The effectiveness of switching and amplifying circuits will be discussed in terms of circuit effectiveness (leakage, operating frequency, and temperature) in preparation for the second phase demonstration of integrated, two-dimensional monolithic InGaAs active pixel sensor arrays for applications in transportable shipboard surveillance, night vision, and emission spectroscopy.

Tandem mass spectrometry data quality assessment by self-convolution.

PubMed

Choo, Keng Wah; Tham, Wai Mun

2007-09-20

Many algorithms have been developed for deciphering the tandem mass spectrometry (MS) data sets. They can be essentially clustered into two classes. The first performs searches on theoretical mass spectrum database, while the second based itself on de novo sequencing from raw mass spectrometry data. It was noted that the quality of mass spectra affects significantly the protein identification processes in both instances. This prompted the authors to explore ways to measure the quality of MS data sets before subjecting them to the protein identification algorithms, thus allowing for more meaningful searches and increased confidence level of proteins identified. The proposed method measures the qualities of MS data sets based on the symmetric property of b- and y-ion peaks present in a MS spectrum. Self-convolution on MS data and its time-reversal copy was employed. Due to the symmetric nature of b-ions and y-ions peaks, the self-convolution result of a good spectrum would produce a highest mid point intensity peak. To reduce processing time, self-convolution was achieved using Fast Fourier Transform and its inverse transform, followed by the removal of the "DC" (Direct Current) component and the normalisation of the data set. The quality score was defined as the ratio of the intensity at the mid point to the remaining peaks of the convolution result. The method was validated using both theoretical mass spectra, with various permutations, and several real MS data sets. The results were encouraging, revealing a high percentage of positive prediction rates for spectra with good quality scores. We have demonstrated in this work a method for determining the quality of tandem MS data set. By pre-determining the quality of tandem MS data before subjecting them to protein identification algorithms, spurious protein predictions due to poor tandem MS data are avoided, giving scientists greater confidence in the predicted results. We conclude that the algorithm performs well

Studying the Kinematic Behavior of Coronal Mass Ejections and Other Solar Phenomena using the Time-Convolution Mapping Method

NASA Astrophysics Data System (ADS)

Hess Webber, Shea A.; Thompson, Barbara J.; Kwon, Ryun Young; Ireland, Jack

2018-01-01

An improved understanding of the kinematic properties of CMEs and CME-associated phenomena has several impacts: 1) a less ambiguous method of mapping propagating structures into their inner coronal manifestations, 2) a clearer view of the relationship between the “main” CME and CME-associated brightenings, and 3) an improved identification of the heliospheric sources of shocks, Type II bursts, and SEPs. We present the results of a mapping technique that facilitates the separation of CMEs and CME-associated brightenings (such as shocks) from background corona. The Time Convolution Mapping Method (TCMM) segments coronagraph data to identify the time history of coronal evolution, the advantage being that the spatiotemporal evolution profiles allow users to separate features with different propagation characteristics. For example, separating “main” CME mass from CME-associated brightenings or shocks is a well-known obstacle, which the TCMM aids in differentiating. A TCMM CME map is made by first recording the maximum value each individual pixel in the image reaches during the traversal of the CME. Then the maximum value is convolved with an index to indicate the time that the pixel reached that value. The TCMM user is then able to identify continuous “kinematic profiles,” indicating related kinematic behavior, and also identify breaks in the profiles that indicate a discontinuity in kinematic history (i.e. different structures or different propagation characteristics). The maps obtained from multiple spacecraft viewpoints (i.e., STEREO and SOHO) can then be fit with advanced structural models to obtain the 3D properties of the evolving phenomena. We will also comment on the TCMM's further applicability toward the tracking of prominences, coronal hole boundaries and coronal cavities.

Ultrasensitive Kilo-Pixel Imaging Array of Photon Noise-Limited Kinetic Inductance Detectors Over an Octave of Bandwidth for THz Astronomy

NASA Astrophysics Data System (ADS)

Bueno, J.; Murugesan, V.; Karatsu, K.; Thoen, D. J.; Baselmans, J. J. A.

2018-05-01

We present the development of a background-limited kilo-pixel imaging array of ultrawide bandwidth kinetic inductance detectors (KIDs) suitable for space-based THz astronomy applications. The array consists of 989 KIDs, in which the radiation is coupled to each KID via a leaky lens antenna, covering the frequency range between 1.4 and 2.8 THz. The single pixel performance is fully characterised using a representative small array in terms of sensitivity, optical efficiency, beam pattern and frequency response, matching very well its expected performance. The kilo-pixel array is characterised electrically, finding a yield larger than 90% and an averaged noise-equivalent power lower than 3 × 10^{-19} W/Hz^{1/2} . The interaction between the kilo-pixel array and cosmic rays is studied, with an expected dead time lower than 0.6% when operated in an L2 or a similar far-Earth orbit.

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.

1982-01-01

Practical methods of computer screening cloud-contaminated pixels from data of various satellite systems are proposed. Examples are given of the location of clouds and representative landscape features in HCMM spectral space of reflectance (VIS) vs emission (IR). Methods of screening out cloud affected HCMM are discussed. The character of subvisible absorbing-emitting atmospheric layers (subvisible cirrus or SCi) in HCMM data is considered and radiosonde soundings are examined in relation to the presence of SCi. The statistical characteristics of multispectral meteorological satellite data in clear and SCi affected areas are discussed. Examples in TIROS-N and NOAA-7 data from several states and Mexico are presented. The VIS-IR cluster screening method for removing clouds is applied to a 262, 144 pixel HCMM scene from south Texas and northeast Mexico. The SCi that remain after cluster screening are sited out by applying a statistically determined IR limit.

DeepFix: A Fully Convolutional Neural Network for Predicting Human Eye Fixations.

PubMed

Kruthiventi, Srinivas S S; Ayush, Kumar; Babu, R Venkatesh

2017-09-01

Understanding and predicting the human visual attention mechanism is an active area of research in the fields of neuroscience and computer vision. In this paper, we propose DeepFix, a fully convolutional neural network, which models the bottom-up mechanism of visual attention via saliency prediction. Unlike classical works, which characterize the saliency map using various hand-crafted features, our model automatically learns features in a hierarchical fashion and predicts the saliency map in an end-to-end manner. DeepFix is designed to capture semantics at multiple scales while taking global context into account, by using network layers with very large receptive fields. Generally, fully convolutional nets are spatially invariant-this prevents them from modeling location-dependent patterns (e.g., centre-bias). Our network handles this by incorporating a novel location-biased convolutional layer. We evaluate our model on multiple challenging saliency data sets and show that it achieves the state-of-the-art results.

Microlens performance limits in sub-2mum pixel CMOS image sensors.

PubMed

Huo, Yijie; Fesenmaier, Christian C; Catrysse, Peter B

2010-03-15

CMOS image sensors with smaller pixels are expected to enable digital imaging systems with better resolution. When pixel size scales below 2 mum, however, diffraction affects the optical performance of the pixel and its microlens, in particular. We present a first-principles electromagnetic analysis of microlens behavior during the lateral scaling of CMOS image sensor pixels. We establish for a three-metal-layer pixel that diffraction prevents the microlens from acting as a focusing element when pixels become smaller than 1.4 microm. This severely degrades performance for on and off-axis pixels in red, green and blue color channels. We predict that one-metal-layer or backside-illuminated pixels are required to extend the functionality of microlenses beyond the 1.4 microm pixel node.

Adaptive pixel-to-pixel projection intensity adjustment for measuring a shiny surface using orthogonal color fringe pattern projection

NASA Astrophysics Data System (ADS)

Chen, Chao; Gao, Nan; Wang, Xiangjun; Zhang, Zonghua

2018-05-01

Three-dimensional (3D) shape measurement based on fringe pattern projection techniques has been commonly used in various fields. One of the remaining challenges in fringe pattern projection is that camera sensor saturation may occur if there is a large range of reflectivity variation across the surface that causes measurement errors. To overcome this problem, a novel fringe pattern projection method is proposed to avoid image saturation and maintain high-intensity modulation for measuring shiny surfaces by adaptively adjusting the pixel-to-pixel projection intensity according to the surface reflectivity. First, three sets of orthogonal color fringe patterns and a sequence of uniform gray-level patterns with different gray levels are projected onto a measured surface by a projector. The patterns are deformed with respect to the object surface and captured by a camera from a different viewpoint. Subsequently, the optimal projection intensity at each pixel is determined by fusing different gray levels and transforming the camera pixel coordinate system into the projector pixel coordinate system. Finally, the adapted fringe patterns are created and used for 3D shape measurement. Experimental results on a flat checkerboard and shiny objects demonstrate that the proposed method can measure shiny surfaces with high accuracy.

High-voltage pixel sensors for ATLAS upgrade

NASA Astrophysics Data System (ADS)

Perić, I.; Kreidl, C.; Fischer, P.; Bompard, F.; Breugnon, P.; Clemens, J.-C.; Fougeron, D.; Liu, J.; Pangaud, P.; Rozanov, A.; Barbero, M.; Feigl, S.; Capeans, M.; Ferrere, D.; Pernegger, H.; Ristic, B.; Muenstermann, D.; Gonzalez Sevilla, S.; La Rosa, A.; Miucci, A.; Nessi, M.; Iacobucci, G.; Backhaus, M.; Hügging, Fabian; Krüger, H.; Hemperek, T.; Obermann, T.; Wermes, N.; Garcia-Sciveres, M.; Quadt, A.; Weingarten, J.; George, M.; Grosse-Knetter, J.; Rieger, J.; Bates, R.; Blue, A.; Buttar, C.; Hynds, D.

2014-11-01

The high-voltage (HV-) CMOS pixel sensors offer several good properties: a fast charge collection by drift, the possibility to implement relatively complex CMOS in-pixel electronics and the compatibility with commercial processes. The sensor element is a deep n-well diode in a p-type substrate. The n-well contains CMOS pixel electronics. The main charge collection mechanism is drift in a shallow, high field region, which leads to a fast charge collection and a high radiation tolerance. We are currently evaluating the use of the high-voltage detectors implemented in 180 nm HV-CMOS technology for the high-luminosity ATLAS upgrade. Our approach is replacing the existing pixel and strip sensors with the CMOS sensors while keeping the presently used readout ASICs. By intelligence we mean the ability of the sensor to recognize a particle hit and generate the address information. In this way we could benefit from the advantages of the HV sensor technology such as lower cost, lower mass, lower operating voltage, smaller pitch, smaller clusters at high incidence angles. Additionally we expect to achieve a radiation hardness necessary for ATLAS upgrade. In order to test the concept, we have designed two HV-CMOS prototypes that can be readout in two ways: using pixel and strip readout chips. In the case of the pixel readout, the connection between HV-CMOS sensor and the readout ASIC can be established capacitively.

Keypoint Density-Based Region Proposal for Fine-Grained Object Detection and Classification Using Regions with Convolutional Neural Network Features

DTIC Science & Technology

2015-12-15

Keypoint Density-based Region Proposal for Fine-Grained Object Detection and Classification using Regions with Convolutional Neural Network ... Convolutional Neural Networks (CNNs) enable them to outperform conventional techniques on standard object detection and classification tasks, their...detection accuracy and speed on the fine-grained Caltech UCSD bird dataset (Wah et al., 2011). Recently, Convolutional Neural Networks (CNNs), a deep

Wafer-scale pixelated detector system

DOEpatents

Fahim, Farah; Deptuch, Grzegorz; Zimmerman, Tom

2017-10-17

A large area, gapless, detection system comprises at least one sensor; an interposer operably connected to the at least one sensor; and at least one application specific integrated circuit operably connected to the sensor via the interposer wherein the detection system provides high dynamic range while maintaining small pixel area and low power dissipation. Thereby the invention provides methods and systems for a wafer-scale gapless and seamless detector systems with small pixels, which have both high dynamic range and low power dissipation.

Cloud Detection by Fusing Multi-Scale Convolutional Features

NASA Astrophysics Data System (ADS)

Li, Zhiwei; Shen, Huanfeng; Wei, Yancong; Cheng, Qing; Yuan, Qiangqiang

2018-04-01

Clouds detection is an important pre-processing step for accurate application of optical satellite imagery. Recent studies indicate that deep learning achieves best performance in image segmentation tasks. Aiming at boosting the accuracy of cloud detection for multispectral imagery, especially for those that contain only visible and near infrared bands, in this paper, we proposed a deep learning based cloud detection method termed MSCN (multi-scale cloud net), which segments cloud by fusing multi-scale convolutional features. MSCN was trained on a global cloud cover validation collection, and was tested in more than ten types of optical images with different resolution. Experiment results show that MSCN has obvious advantages over the traditional multi-feature combined cloud detection method in accuracy, especially when in snow and other areas covered by bright non-cloud objects. Besides, MSCN produced more detailed cloud masks than the compared deep cloud detection convolution network. The effectiveness of MSCN make it promising for practical application in multiple kinds of optical imagery.

Pixel Paradise

NASA Technical Reports Server (NTRS)

1998-01-01

PixelVision, Inc., has developed a series of integrated imaging engines capable of high-resolution image capture at dynamic speeds. This technology was used originally at Jet Propulsion Laboratory in a series of imaging engines for a NASA mission to Pluto. By producing this integrated package, Charge-Coupled Device (CCD) technology has been made accessible to a wide range of users.

Challenges of small-pixel infrared detectors: a review.

PubMed

Rogalski, A; Martyniuk, P; Kopytko, M

2016-04-01

In the last two decades, several new concepts for improving the performance of infrared detectors have been proposed. These new concepts particularly address the drive towards the so-called high operating temperature focal plane arrays (FPAs), aiming to increase detector operating temperatures, and as a consequence reduce the cost of infrared systems. In imaging systems with the above megapixel formats, pixel dimension plays a crucial role in determining critical system attributes such as system size, weight and power consumption (SWaP). The advent of smaller pixels has also resulted in the superior spatial and temperature resolution of these systems. Optimum pixel dimensions are limited by diffraction effects from the aperture, and are in turn wavelength-dependent. In this paper, the key challenges in realizing optimum pixel dimensions in FPA design including dark current, pixel hybridization, pixel delineation, and unit cell readout capacity are outlined to achieve a sufficiently adequate modulation transfer function for the ultra-small pitches involved. Both photon and thermal detectors have been considered. Concerning infrared photon detectors, the trade-offs between two types of competing technology-HgCdTe material systems and III-V materials (mainly barrier detectors)-have been investigated.

Cyclops: single-pixel imaging lidar system based on compressive sensing

NASA Astrophysics Data System (ADS)

Magalhães, F.; Correia, M. V.; Farahi, F.; Pereira do Carmo, J.; Araújo, F. M.

2017-11-01

Mars and the Moon are envisaged as major destinations of future space exploration missions in the upcoming decades. Imaging LIDARs are seen as a key enabling technology in the support of autonomous guidance, navigation and control operations, as they can provide very accurate, wide range, high-resolution distance measurements as required for the exploration missions. Imaging LIDARs can be used at critical stages of these exploration missions, such as descent and selection of safe landing sites, rendezvous and docking manoeuvres, or robotic surface navigation and exploration. Despite these devices have been commercially available and used for long in diverse metrology and ranging applications, their size, mass and power consumption are still far from being suitable and attractive for space exploratory missions. Here, we describe a compact Single-Pixel Imaging LIDAR System that is based on a compressive sensing technique. The application of the compressive codes to a DMD array enables compression of the spatial information, while the collection of timing histograms correlated to the pulsed laser source ensures image reconstruction at the ranged distances. Single-pixel cameras have been compared with raster scanning and array based counterparts in terms of noise performance, and proved to be superior. Since a single photodetector is used, a better SNR and higher reliability is expected in contrast with systems using large format photodetector arrays. Furthermore, the event of failure of one or more micromirror elements in the DMD does not prevent full reconstruction of the images. This brings additional robustness to the proposed 3D imaging LIDAR. The prototype that was implemented has three modes of operation. Range Finder: outputs the average distance between the system and the area of the target under illumination; Attitude Meter: provides the slope of the target surface based on distance measurements in three areas of the target; 3D Imager: produces 3D ranged

Super-pixel extraction based on multi-channel pulse coupled neural network

NASA Astrophysics Data System (ADS)

Xu, GuangZhu; Hu, Song; Zhang, Liu; Zhao, JingJing; Fu, YunXia; Lei, BangJun

2018-04-01

Super-pixel extraction techniques group pixels to form over-segmented image blocks according to the similarity among pixels. Compared with the traditional pixel-based methods, the image descripting method based on super-pixel has advantages of less calculation, being easy to perceive, and has been widely used in image processing and computer vision applications. Pulse coupled neural network (PCNN) is a biologically inspired model, which stems from the phenomenon of synchronous pulse release in the visual cortex of cats. Each PCNN neuron can correspond to a pixel of an input image, and the dynamic firing pattern of each neuron contains both the pixel feature information and its context spatial structural information. In this paper, a new color super-pixel extraction algorithm based on multi-channel pulse coupled neural network (MPCNN) was proposed. The algorithm adopted the block dividing idea of SLIC algorithm, and the image was divided into blocks with same size first. Then, for each image block, the adjacent pixels of each seed with similar color were classified as a group, named a super-pixel. At last, post-processing was adopted for those pixels or pixel blocks which had not been grouped. Experiments show that the proposed method can adjust the number of superpixel and segmentation precision by setting parameters, and has good potential for super-pixel extraction.

Attenuating Stereo Pixel-Locking via Affine Window Adaptation

NASA Technical Reports Server (NTRS)

Stein, Andrew N.; Huertas, Andres; Matthies, Larry H.

2006-01-01

For real-time stereo vision systems, the standard method for estimating sub-pixel stereo disparity given an initial integer disparity map involves fitting parabolas to a matching cost function aggregated over rectangular windows. This results in a phenomenon known as 'pixel-locking,' which produces artificially-peaked histograms of sub-pixel disparity. These peaks correspond to the introduction of erroneous ripples or waves in the 3D reconstruction of truly Rat surfaces. Since stereo vision is a common input modality for autonomous vehicles, these inaccuracies can pose a problem for safe, reliable navigation. This paper proposes a new method for sub-pixel stereo disparity estimation, based on ideas from Lucas-Kanade tracking and optical flow, which substantially reduces the pixel-locking effect. In addition, it has the ability to correct much larger initial disparity errors than previous approaches and is more general as it applies not only to the ground plane.

Fast Pixel Buffer For Processing With Lookup Tables

NASA Technical Reports Server (NTRS)

Fisher, Timothy E.

1992-01-01

Proposed scheme for buffering data on intensities of picture elements (pixels) of image increases rate or processing beyond that attainable when data read, one pixel at time, from main image memory. Scheme applied in design of specialized image-processing circuitry. Intended to optimize performance of processor in which electronic equivalent of address-lookup table used to address those pixels in main image memory required for processing.

Steganography on quantum pixel images using Shannon entropy

NASA Astrophysics Data System (ADS)

Laurel, Carlos Ortega; Dong, Shi-Hai; Cruz-Irisson, M.

2016-07-01

This paper presents a steganographical algorithm based on least significant bit (LSB) from the most significant bit information (MSBI) and the equivalence of a bit pixel image to a quantum pixel image, which permits to make the information communicate secretly onto quantum pixel images for its secure transmission through insecure channels. This algorithm offers higher security since it exploits the Shannon entropy for an image.

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.

Fully convolutional neural networks for polyp segmentation in colonoscopy

NASA Astrophysics Data System (ADS)

Brandao, Patrick; Mazomenos, Evangelos; Ciuti, Gastone; Caliò, Renato; Bianchi, Federico; Menciassi, Arianna; Dario, Paolo; Koulaouzidis, Anastasios; Arezzo, Alberto; Stoyanov, Danail

2017-03-01

Colorectal cancer (CRC) is one of the most common and deadliest forms of cancer, accounting for nearly 10% of all forms of cancer in the world. Even though colonoscopy is considered the most effective method for screening and diagnosis, the success of the procedure is highly dependent on the operator skills and level of hand-eye coordination. In this work, we propose to adapt fully convolution neural networks (FCN), to identify and segment polyps in colonoscopy images. We converted three established networks into a fully convolution architecture and fine-tuned their learned representations to the polyp segmentation task. We validate our framework on the 2015 MICCAI polyp detection challenge dataset, surpassing the state-of-the-art in automated polyp detection. Our method obtained high segmentation accuracy and a detection precision and recall of 73.61% and 86.31%, respectively.

Hit efficiency study of CMS prototype forward pixel detectors

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

Kim, Dongwook; /Johns Hopkins U.

2006-01-01

In this paper the author describes the measurement of the hit efficiency of a prototype pixel device for the CMS forward pixel detector. These pixel detectors were FM type sensors with PSI46V1 chip readout. The data were taken with the 120 GeV proton beam at Fermilab during the period of December 2004 to February 2005. The detectors proved to be highly efficient (99.27 {+-} 0.02%). The inefficiency was primarily located near the corners of the individual pixels.

Simulation of ICD-9 to ICD-10-CM Transition for Family Medicine: Simple or Convoluted?

PubMed

Grief, Samuel N; Patel, Jesal; Kochendorfer, Karl M; Green, Lee A; Lussier, Yves A; Li, Jianrong; Burton, Michael; Boyd, Andrew D

2016-01-01

The objective of this study was to examine the impact of the transition from International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM), to Interactional Classification of Diseases, 10th Revision, Clinical Modification (ICD-10-CM), on family medicine and to identify areas where additional training might be required. Family medicine ICD-9-CM codes were obtained from an Illinois Medicaid data set (113,000 patient visits and $5.5 million in claims). Using the science of networks, we evaluated each ICD-9-CM code used by family medicine physicians to determine whether the transition was simple or convoluted. A simple transition is defined as 1 ICD-9-CM code mapping to 1 ICD-10-CM code, or 1 ICD-9-CM code mapping to multiple ICD-10-CM codes. A convoluted transition is where the transitions between coding systems is nonreciprocal and complex, with multiple codes for which definitions become intertwined. Three family medicine physicians evaluated the most frequently encountered complex mappings for clinical accuracy. Of the 1635 diagnosis codes used by family medicine physicians, 70% of the codes were categorized as simple, 27% of codes were convoluted, and 3% had no mapping. For the visits, 75%, 24%, and 1% corresponded with simple, convoluted, and no mapping, respectively. Payment for submitted claims was similarly aligned. Of the frequently encountered convoluted codes, 3 diagnosis codes were clinically incorrect, but they represent only <0.1% of the overall diagnosis codes. The transition to ICD-10-CM is simple for 70% or more of diagnosis codes, visits, and reimbursement for a family medicine physician. However, some frequently used codes for disease management are convoluted and incorrect, and for which additional resources need to be invested to ensure a successful transition to ICD-10-CM. © Copyright 2016 by the American Board of Family Medicine.

Simulation of ICD-9 to ICD-10-CM transition for family medicine: simple or convoluted?

PubMed Central

Grief, Samuel N.; Patel, Jesal; Lussier, Yves A.; Li, Jianrong; Burton, Michael; Boyd, Andrew D.

2017-01-01

Objectives The objective of this study was to examine the impact of the transition from International Classification of Disease Version Nine Clinical Modification (ICD-9-CM) to Interactional Classification of Disease Version Ten Clinical Modification (ICD-10-CM) on family medicine and identify areas where additional training might be required. Methods Family medicine ICD-9-CM codes were obtained from an Illinois Medicaid data set (113,000 patient visits and $5.5 million dollars in claims). Using the science of networks we evaluated each ICD-9-CM code used by family medicine physicians to determine if the transition was simple or convoluted.1 A simple translation is defined as one ICD-9-CM code mapping to one ICD-10-CM code or one ICD-9-CM code mapping to multiple ICD-10-CM codes. A convoluted transition is where the transitions between coding systems is non-reciprocal and complex with multiple codes where definitions become intertwined. Three family medicine physicians evaluated the most frequently encountered complex mappings for clinical accuracy. Results Of the 1635 diagnosis codes used by the family medicine physicians, 70% of the codes were categorized as simple, 27% of the diagnosis codes were convoluted and 3% were found to have no mapping. For the visits, 75%, 24%, and 1% corresponded with simple, convoluted, and no mapping, respectively. Payment for submitted claims were similarly aligned. Of the frequently encountered convoluted codes, 3 diagnosis codes were clinically incorrect, but they represent only < 0.1% of the overall diagnosis codes. Conclusions The transition to ICD-10-CM is simple for 70% or more of diagnosis codes, visits, and reimbursement for a family medicine physician. However, some frequently used codes for disease management are convoluted and incorrect, where additional resources need to be invested to ensure a successful transition to ICD-10-CM. PMID:26769875

Effects of Convoluted Divergent Flap Contouring on the Performance of a Fixed-Geometry Nonaxisymmetric Exhaust Nozzle

NASA Technical Reports Server (NTRS)

Asbury, Scott C.; Hunter, Craig A.

1999-01-01

An investigation was conducted in the model preparation area of the Langley 16-Foot Transonic Tunnel to determine the effects of convoluted divergent-flap contouring on the internal performance of a fixed-geometry, nonaxisymmetric, convergent-divergent exhaust nozzle. Testing was conducted at static conditions using a sub-scale nozzle model with one baseline and four convoluted configurations. All tests were conducted with no external flow at nozzle pressure ratios from 1.25 to approximately 9.50. Results indicate that baseline nozzle performance was dominated by unstable, shock-induced, boundary-layer separation at overexpanded conditions. Convoluted configurations were found to significantly reduce, and in some cases totally alleviate separation at overexpanded conditions. This result was attributed to the ability of convoluted contouring to energize and improve the condition of the nozzle boundary layer. Separation alleviation offers potential for installed nozzle aeropropulsive (thrust-minus-drag) performance benefits by reducing drag at forward flight speeds, even though this may reduce nozzle thrust ratio as much as 6.4% at off-design conditions. At on-design conditions, nozzle thrust ratio for the convoluted configurations ranged from 1% to 2.9% below the baseline configuration; this was a result of increased skin friction and oblique shock losses inside the nozzle.

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.

Superpixel-based graph cuts for accurate stereo matching

NASA Astrophysics Data System (ADS)

Feng, Liting; Qin, Kaihuai

2017-06-01

Estimating the surface normal vector and disparity of a pixel simultaneously, also known as three-dimensional label method, has been widely used in recent continuous stereo matching problem to achieve sub-pixel accuracy. However, due to the infinite label space, it’s extremely hard to assign each pixel an appropriate label. In this paper, we present an accurate and efficient algorithm, integrating patchmatch with graph cuts, to approach this critical computational problem. Besides, to get robust and precise matching cost, we use a convolutional neural network to learn a similarity measure on small image patches. Compared with other MRF related methods, our method has several advantages: its sub-modular property ensures a sub-problem optimality which is easy to perform in parallel; graph cuts can simultaneously update multiple pixels, avoiding local minima caused by sequential optimizers like belief propagation; it uses segmentation results for better local expansion move; local propagation and randomization can easily generate the initial solution without using external methods. Middlebury experiments show that our method can get higher accuracy than other MRF-based algorithms.

a Novel Deep Convolutional Neural Network for Spectral-Spatial Classification of Hyperspectral Data

NASA Astrophysics Data System (ADS)

Li, N.; Wang, C.; Zhao, H.; Gong, X.; Wang, D.

2018-04-01

Spatial and spectral information are obtained simultaneously by hyperspectral remote sensing. Joint extraction of these information of hyperspectral image is one of most import methods for hyperspectral image classification. In this paper, a novel deep convolutional neural network (CNN) is proposed, which extracts spectral-spatial information of hyperspectral images correctly. The proposed model not only learns sufficient knowledge from the limited number of samples, but also has powerful generalization ability. The proposed framework based on three-dimensional convolution can extract spectral-spatial features of labeled samples effectively. Though CNN has shown its robustness to distortion, it cannot extract features of different scales through the traditional pooling layer that only have one size of pooling window. Hence, spatial pyramid pooling (SPP) is introduced into three-dimensional local convolutional filters for hyperspectral classification. Experimental results with a widely used hyperspectral remote sensing dataset show that the proposed model provides competitive performance.

Theory and applications of structured light single pixel imaging

NASA Astrophysics Data System (ADS)

Stokoe, Robert J.; Stockton, Patrick A.; Pezeshki, Ali; Bartels, Randy A.

2018-02-01

Many single-pixel imaging techniques have been developed in recent years. Though the methods of image acquisition vary considerably, the methods share unifying features that make general analysis possible. Furthermore, the methods developed thus far are based on intuitive processes that enable simple and physically-motivated reconstruction algorithms, however, this approach may not leverage the full potential of single-pixel imaging. We present a general theoretical framework of single-pixel imaging based on frame theory, which enables general, mathematically rigorous analysis. We apply our theoretical framework to existing single-pixel imaging techniques, as well as provide a foundation for developing more-advanced methods of image acquisition and reconstruction. The proposed frame theoretic framework for single-pixel imaging results in improved noise robustness, decrease in acquisition time, and can take advantage of special properties of the specimen under study. By building on this framework, new methods of imaging with a single element detector can be developed to realize the full potential associated with single-pixel imaging.

Quantifying the interplay effect in prostate IMRT delivery using a convolution-based method.

PubMed

Li, Haisen S; Chetty, Indrin J; Solberg, Timothy D

2008-05-01

The authors present a segment-based convolution method to account for the interplay effect between intrafraction organ motion and the multileaf collimator position for each particular segment in intensity modulated radiation therapy (IMRT) delivered in a step-and-shoot manner. In this method, the static dose distribution attributed to each segment is convolved with the probability density function (PDF) of motion during delivery of the segment, whereas in the conventional convolution method ("average-based convolution"), the static dose distribution is convolved with the PDF averaged over an entire fraction, an entire treatment course, or even an entire patient population. In the case of IMRT delivered in a step-and-shoot manner, the average-based convolution method assumes that in each segment the target volume experiences the same motion pattern (PDF) as that of population. In the segment-based convolution method, the dose during each segment is calculated by convolving the static dose with the motion PDF specific to that segment, allowing both intrafraction motion and the interplay effect to be accounted for in the dose calculation. Intrafraction prostate motion data from a population of 35 patients tracked using the Calypso system (Calypso Medical Technologies, Inc., Seattle, WA) was used to generate motion PDFs. These were then convolved with dose distributions from clinical prostate IMRT plans. For a single segment with a small number of monitor units, the interplay effect introduced errors of up to 25.9% in the mean CTV dose compared against the planned dose evaluated by using the PDF of the entire fraction. In contrast, the interplay effect reduced the minimum CTV dose by 4.4%, and the CTV generalized equivalent uniform dose by 1.3%, in single fraction plans. For entire treatment courses delivered in either a hypofractionated (five fractions) or conventional (> 30 fractions) regimen, the discrepancy in total dose due to interplay effect was negligible.

Concatenated coding systems employing a unit-memory convolutional code and a byte-oriented decoding algorithm

NASA Technical Reports Server (NTRS)

Lee, L.-N.

1977-01-01

Concatenated coding systems utilizing a convolutional code as the inner code and a Reed-Solomon code as the outer code are considered. In order to obtain very reliable communications over a very noisy channel with relatively modest coding complexity, it is proposed to concatenate a byte-oriented unit-memory convolutional code with an RS outer code whose symbol size is one byte. It is further proposed to utilize a real-time minimal-byte-error probability decoding algorithm, together with feedback from the outer decoder, in the decoder for the inner convolutional code. The performance of the proposed concatenated coding system is studied, and the improvement over conventional concatenated systems due to each additional feature is isolated.

Concatenated coding systems employing a unit-memory convolutional code and a byte-oriented decoding algorithm

NASA Technical Reports Server (NTRS)

Lee, L. N.

1976-01-01

Concatenated coding systems utilizing a convolutional code as the inner code and a Reed-Solomon code as the outer code are considered. In order to obtain very reliable communications over a very noisy channel with relatively small coding complexity, it is proposed to concatenate a byte oriented unit memory convolutional code with an RS outer code whose symbol size is one byte. It is further proposed to utilize a real time minimal byte error probability decoding algorithm, together with feedback from the outer decoder, in the decoder for the inner convolutional code. The performance of the proposed concatenated coding system is studied, and the improvement over conventional concatenated systems due to each additional feature is isolated.

Combining morphometric features and convolutional networks fusion for glaucoma diagnosis

NASA Astrophysics Data System (ADS)

Perdomo, Oscar; Arevalo, John; González, Fabio A.

2017-11-01

Glaucoma is an eye condition that leads to loss of vision and blindness. Ophthalmoscopy exam evaluates the shape, color and proportion between the optic disc and physiologic cup, but the lack of agreement among experts is still the main diagnosis problem. The application of deep convolutional neural networks combined with automatic extraction of features such as: the cup-to-disc distance in the four quadrants, the perimeter, area, eccentricity, the major radio, the minor radio in optic disc and cup, in addition to all the ratios among the previous parameters may help with a better automatic grading of glaucoma. This paper presents a strategy to merge morphological features and deep convolutional neural networks as a novel methodology to support the glaucoma diagnosis in eye fundus images.

Airplane detection in remote sensing images using convolutional neural networks

NASA Astrophysics Data System (ADS)

Ouyang, Chao; Chen, Zhong; Zhang, Feng; Zhang, Yifei

2018-03-01

Airplane detection in remote sensing images remains a challenging problem and has also been taking a great interest to researchers. In this paper we propose an effective method to detect airplanes in remote sensing images using convolutional neural networks. Deep learning methods show greater advantages than the traditional methods with the rise of deep neural networks in target detection, and we give an explanation why this happens. To improve the performance on detection of airplane, we combine a region proposal algorithm with convolutional neural networks. And in the training phase, we divide the background into multi classes rather than one class, which can reduce false alarms. Our experimental results show that the proposed method is effective and robust in detecting airplane.

Convolute laminations — a theoretical analysis: example of a Pennsylvanian sandstone

NASA Astrophysics Data System (ADS)

Visher, Glenn S.; Cunningham, Russ D.

1981-03-01

Data from an outcropping laminated interval were collected and analyzed to test the applicability of a theoretical model describing instability of layered systems. Rayleigh—Taylor wave perturbations result at the interface between fluids of contrasting density, viscosity, and thickness. In the special case where reverse density and viscosity interlaminations are developed, the deformation response produces a single wave with predictable amplitudes, wavelengths, and amplification rates. Physical measurements from both the outcropping section and modern sediments suggest the usefulness of the model for the interpretation of convolute laminations. Internal characteristics of the stratigraphic interval, and the developmental sequence of convoluted beds, are used to document the developmental history of these structures.

Active pixel imagers incorporating pixel-level amplifiers based on polycrystalline-silicon thin-film transistors

PubMed Central

El-Mohri, Youcef; Antonuk, Larry E.; Koniczek, Martin; Zhao, Qihua; Li, Yixin; Street, Robert A.; Lu, Jeng-Ping

2009-01-01

Active matrix, flat-panel imagers (AMFPIs) employing a 2D matrix of a-Si addressing TFTs have become ubiquitous in many x-ray imaging applications due to their numerous advantages. However, under conditions of low exposures and∕or high spatial resolution, their signal-to-noise performance is constrained by the modest system gain relative to the electronic additive noise. In this article, a strategy for overcoming this limitation through the incorporation of in-pixel amplification circuits, referred to as active pixel (AP) architectures, using polycrystalline-silicon (poly-Si) TFTs is reported. Compared to a-Si, poly-Si offers substantially higher mobilities, enabling higher TFT currents and the possibility of sophisticated AP designs based on both n- and p-channel TFTs. Three prototype indirect detection arrays employing poly-Si TFTs and a continuous a-Si photodiode structure were characterized. The prototypes consist of an array (PSI-1) that employs a pixel architecture with a single TFT, as well as two arrays (PSI-2 and PSI-3) that employ AP architectures based on three and five TFTs, respectively. While PSI-1 serves as a reference with a design similar to that of conventional AMFPI arrays, PSI-2 and PSI-3 incorporate additional in-pixel amplification circuitry. Compared to PSI-1, results of x-ray sensitivity demonstrate signal gains of ∼10.7 and 20.9 for PSI-2 and PSI-3, respectively. These values are in reasonable agreement with design expectations, demonstrating that poly-Si AP circuits can be tailored to provide a desired level of signal gain. PSI-2 exhibits the same high levels of charge trapping as those observed for PSI-1 and other conventional arrays employing a continuous photodiode structure. For PSI-3, charge trapping was found to be significantly lower and largely independent of the bias voltage applied across the photodiode. MTF results indicate that the use of a continuous photodiode structure in PSI-1, PSI-2, and PSI-3 results in optical fill

Active pixel imagers incorporating pixel-level amplifiers based on polycrystalline-silicon thin-film transistors.

PubMed

El-Mohri, Youcef; Antonuk, Larry E; Koniczek, Martin; Zhao, Qihua; Li, Yixin; Street, Robert A; Lu, Jeng-Ping

2009-07-01

Active matrix, flat-panel imagers (AMFPIs) employing a 2D matrix of a-Si addressing TFTs have become ubiquitous in many x-ray imaging applications due to their numerous advantages. However, under conditions of low exposures and/or high spatial resolution, their signal-to-noise performance is constrained by the modest system gain relative to the electronic additive noise. In this article, a strategy for overcoming this limitation through the incorporation of in-pixel amplification circuits, referred to as active pixel (AP) architectures, using polycrystalline-silicon (poly-Si) TFTs is reported. Compared to a-Si, poly-Si offers substantially higher mobilities, enabling higher TFT currents and the possibility of sophisticated AP designs based on both n- and p-channel TFTs. Three prototype indirect detection arrays employing poly-Si TFTs and a continuous a-Si photodiode structure were characterized. The prototypes consist of an array (PSI-1) that employs a pixel architecture with a single TFT, as well as two arrays (PSI-2 and PSI-3) that employ AP architectures based on three and five TFTs, respectively. While PSI-1 serves as a reference with a design similar to that of conventional AMFPI arrays, PSI-2 and PSI-3 incorporate additional in-pixel amplification circuitry. Compared to PSI-1, results of x-ray sensitivity demonstrate signal gains of approximately 10.7 and 20.9 for PSI-2 and PSI-3, respectively. These values are in reasonable agreement with design expectations, demonstrating that poly-Si AP circuits can be tailored to provide a desired level of signal gain. PSI-2 exhibits the same high levels of charge trapping as those observed for PSI-1 and other conventional arrays employing a continuous photodiode structure. For PSI-3, charge trapping was found to be significantly lower and largely independent of the bias voltage applied across the photodiode. MTF results indicate that the use of a continuous photodiode structure in PSI-1, PSI-2, and PSI-3 results in

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.

Mapping Electrical Crosstalk in Pixelated Sensor Arrays

NASA Technical Reports Server (NTRS)

Seshadri, S.; Cole, D. M.; Hancock, B. R.; Smith, R. M.

2008-01-01

Electronic coupling effects such as Inter-Pixel Capacitance (IPC) affect the quantitative interpretation of image data from CMOS, hybrid visible and infrared imagers alike. Existing methods of characterizing IPC do not provide a map of the spatial variation of IPC over all pixels. We demonstrate a deterministic method that provides a direct quantitative map of the crosstalk across an imager. The approach requires only the ability to reset single pixels to an arbitrary voltage, different from the rest of the imager. No illumination source is required. Mapping IPC independently for each pixel is also made practical by the greater S/N ratio achievable for an electrical stimulus than for an optical stimulus, which is subject to both Poisson statistics and diffusion effects of photo-generated charge. The data we present illustrates a more complex picture of IPC in Teledyne HgCdTe and HyViSi focal plane arrays than is presently understood, including the presence of a newly discovered, long range IPC in the HyViSi FPA that extends tens of pixels in distance, likely stemming from extended field effects in the fully depleted substrate. The sensitivity of the measurement approach has been shown to be good enough to distinguish spatial structure in IPC of the order of 0.1%.

Active-Pixel Image Sensor With Analog-To-Digital Converters

NASA Technical Reports Server (NTRS)

Fossum, Eric R.; Mendis, Sunetra K.; Pain, Bedabrata; Nixon, Robert H.

1995-01-01

Proposed single-chip integrated-circuit image sensor contains 128 x 128 array of active pixel sensors at 50-micrometer pitch. Output terminals of all pixels in each given column connected to analog-to-digital (A/D) converter located at bottom of column. Pixels scanned in semiparallel fashion, one row at time; during time allocated to scanning row, outputs of all active pixel sensors in row fed to respective A/D converters. Design of chip based on complementary metal oxide semiconductor (CMOS) technology, and individual circuit elements fabricated according to 2-micrometer CMOS design rules. Active pixel sensors designed to operate at video rate of 30 frames/second, even at low light levels. A/D scheme based on first-order Sigma-Delta modulation.

ASIC-based architecture for the real-time computation of 2D convolution with large kernel size

NASA Astrophysics Data System (ADS)

Shao, Rui; Zhong, Sheng; Yan, Luxin

2015-12-01

Bidimensional convolution is a low-level processing algorithm of interest in many areas, but its high computational cost constrains the size of the kernels, especially in real-time embedded systems. This paper presents a hardware architecture for the ASIC-based implementation of 2-D convolution with medium-large kernels. Aiming to improve the efficiency of storage resources on-chip, reducing off-chip bandwidth of these two issues, proposed construction of a data cache reuse. Multi-block SPRAM to cross cached images and the on-chip ping-pong operation takes full advantage of the data convolution calculation reuse, design a new ASIC data scheduling scheme and overall architecture. Experimental results show that the structure can achieve 40× 32 size of template real-time convolution operations, and improve the utilization of on-chip memory bandwidth and on-chip memory resources, the experimental results show that the structure satisfies the conditions to maximize data throughput output , reducing the need for off-chip memory bandwidth.

Exploring the Hidden Structure of Astronomical Images: A "Pixelated" View of Solar System and Deep Space Features!

ERIC Educational Resources Information Center

Ward, R. Bruce; Sienkiewicz, Frank; Sadler, Philip; Antonucci, Paul; Miller, Jaimie

2013-01-01

We describe activities created to help student participants in Project ITEAMS (Innovative Technology-Enabled Astronomy for Middle Schools) develop a deeper understanding of picture elements (pixels), image creation, and analysis of the recorded data. ITEAMS is an out-of-school time (OST) program funded by the National Science Foundation (NSF) with…

Computational analysis of current-loss mechanisms in a post-hole convolute driven by magnetically insulated transmission lines

DOE PAGES

Rose, D.  V.; Madrid, E.  A.; Welch, D.  R.; ...

2015-03-04

Numerical simulations of a vacuum post-hole convolute driven by magnetically insulated vacuum transmission lines (MITLs) are used to study current losses due to charged particle emission from the MITL-convolute-system electrodes. This work builds on the results of a previous study [E.A. Madrid et al. Phys. Rev. ST Accel. Beams 16, 120401 (2013)] and adds realistic power pulses, Ohmic heating of anode surfaces, and a model for the formation and evolution of cathode plasmas. The simulations suggest that modestly larger anode-cathode gaps in the MITLs upstream of the convolute result in significantly less current loss. In addition, longer pulse durations leadmore » to somewhat greater current loss due to cathode-plasma expansion. These results can be applied to the design of future MITL-convolute systems for high-current pulsed-power systems.« less

Deep Convolutional Extreme Learning Machine and Its Application in Handwritten Digit Classification

PubMed Central

Yang, Xinyi

2016-01-01

In recent years, some deep learning methods have been developed and applied to image classification applications, such as convolutional neuron network (CNN) and deep belief network (DBN). However they are suffering from some problems like local minima, slow convergence rate, and intensive human intervention. In this paper, we propose a rapid learning method, namely, deep convolutional extreme learning machine (DC-ELM), which combines the power of CNN and fast training of ELM. It uses multiple alternate convolution layers and pooling layers to effectively abstract high level features from input images. Then the abstracted features are fed to an ELM classifier, which leads to better generalization performance with faster learning speed. DC-ELM also introduces stochastic pooling in the last hidden layer to reduce dimensionality of features greatly, thus saving much training time and computation resources. We systematically evaluated the performance of DC-ELM on two handwritten digit data sets: MNIST and USPS. Experimental results show that our method achieved better testing accuracy with significantly shorter training time in comparison with deep learning methods and other ELM methods. PMID:27610128

Alcoholism Detection by Data Augmentation and Convolutional Neural Network with Stochastic Pooling.

PubMed

Wang, Shui-Hua; Lv, Yi-Ding; Sui, Yuxiu; Liu, Shuai; Wang, Su-Jing; Zhang, Yu-Dong

2017-11-17

Alcohol use disorder (AUD) is an important brain disease. It alters the brain structure. Recently, scholars tend to use computer vision based techniques to detect AUD. We collected 235 subjects, 114 alcoholic and 121 non-alcoholic. Among the 235 image, 100 images were used as training set, and data augmentation method was used. The rest 135 images were used as test set. Further, we chose the latest powerful technique-convolutional neural network (CNN) based on convolutional layer, rectified linear unit layer, pooling layer, fully connected layer, and softmax layer. We also compared three different pooling techniques: max pooling, average pooling, and stochastic pooling. The results showed that our method achieved a sensitivity of 96.88%, a specificity of 97.18%, and an accuracy of 97.04%. Our method was better than three state-of-the-art approaches. Besides, stochastic pooling performed better than other max pooling and average pooling. We validated CNN with five convolution layers and two fully connected layers performed the best. The GPU yielded a 149× acceleration in training and a 166× acceleration in test, compared to CPU.

Deep Convolutional Extreme Learning Machine and Its Application in Handwritten Digit Classification.

PubMed

Pang, Shan; Yang, Xinyi

2016-01-01

In recent years, some deep learning methods have been developed and applied to image classification applications, such as convolutional neuron network (CNN) and deep belief network (DBN). However they are suffering from some problems like local minima, slow convergence rate, and intensive human intervention. In this paper, we propose a rapid learning method, namely, deep convolutional extreme learning machine (DC-ELM), which combines the power of CNN and fast training of ELM. It uses multiple alternate convolution layers and pooling layers to effectively abstract high level features from input images. Then the abstracted features are fed to an ELM classifier, which leads to better generalization performance with faster learning speed. DC-ELM also introduces stochastic pooling in the last hidden layer to reduce dimensionality of features greatly, thus saving much training time and computation resources. We systematically evaluated the performance of DC-ELM on two handwritten digit data sets: MNIST and USPS. Experimental results show that our method achieved better testing accuracy with significantly shorter training time in comparison with deep learning methods and other ELM methods.

Method of fabrication of display pixels driven by silicon thin film transistors

DOEpatents

Carey, Paul G.; Smith, Patrick M.

1999-01-01

Display pixels driven by silicon thin film transistors are fabricated on plastic substrates for use in active matrix displays, such as flat panel displays. The process for forming the pixels involves a prior method for forming individual silicon thin film transistors on low-temperature plastic substrates. Low-temperature substrates are generally considered as being incapable of withstanding sustained processing temperatures greater than about 200.degree. C. The pixel formation process results in a complete pixel and active matrix pixel array. A pixel (or picture element) in an active matrix display consists of a silicon thin film transistor (TFT) and a large electrode, which may control a liquid crystal light valve, an emissive material (such as a light emitting diode or LED), or some other light emitting or attenuating material. The pixels can be connected in arrays wherein rows of pixels contain common gate electrodes and columns of pixels contain common drain electrodes. The source electrode of each pixel TFT is connected to its pixel electrode, and is electrically isolated from every other circuit element in the pixel array.

A neighbor pixel communication filtering structure for Dynamic Vision Sensors

NASA Astrophysics Data System (ADS)

Xu, Yuan; Liu, Shiqi; Lu, Hehui; Zhang, Zilong

2017-02-01

For Dynamic Vision Sensors (DVS), thermal noise and junction leakage current induced Background Activity (BA) is the major cause of the deterioration of images quality. Inspired by the smoothing filtering principle of horizontal cells in vertebrate retina, A DVS pixel with Neighbor Pixel Communication (NPC) filtering structure is proposed to solve this issue. The NPC structure is designed to judge the validity of pixel's activity through the communication between its 4 adjacent pixels. The pixel's outputs will be suppressed if its activities are determined not real. The proposed pixel's area is 23.76×24.71μm2 and only 3ns output latency is introduced. In order to validate the effectiveness of the structure, a 5×5 pixel array has been implemented in SMIC 0.13μm CIS process. 3 test cases of array's behavioral model show that the NPC-DVS have an ability of filtering the BA.

Hierarchical Recurrent Neural Hashing for Image Retrieval With Hierarchical Convolutional Features.

PubMed

Lu, Xiaoqiang; Chen, Yaxiong; Li, Xuelong

Hashing has been an important and effective technology in image retrieval due to its computational efficiency and fast search speed. The traditional hashing methods usually learn hash functions to obtain binary codes by exploiting hand-crafted features, which cannot optimally represent the information of the sample. Recently, deep learning methods can achieve better performance, since deep learning architectures can learn more effective image representation features. However, these methods only use semantic features to generate hash codes by shallow projection but ignore texture details. In this paper, we proposed a novel hashing method, namely hierarchical recurrent neural hashing (HRNH), to exploit hierarchical recurrent neural network to generate effective hash codes. There are three contributions of this paper. First, a deep hashing method is proposed to extensively exploit both spatial details and semantic information, in which, we leverage hierarchical convolutional features to construct image pyramid representation. Second, our proposed deep network can exploit directly convolutional feature maps as input to preserve the spatial structure of convolutional feature maps. Finally, we propose a new loss function that considers the quantization error of binarizing the continuous embeddings into the discrete binary codes, and simultaneously maintains the semantic similarity and balanceable property of hash codes. Experimental results on four widely used data sets demonstrate that the proposed HRNH can achieve superior performance over other state-of-the-art hashing methods.Hashing has been an important and effective technology in image retrieval due to its computational efficiency and fast search speed. The traditional hashing methods usually learn hash functions to obtain binary codes by exploiting hand-crafted features, which cannot optimally represent the information of the sample. Recently, deep learning methods can achieve better performance, since deep

Collaborative identification method for sea battlefield target based on deep convolutional neural networks

NASA Astrophysics Data System (ADS)

Zheng, Guangdi; Pan, Mingbo; Liu, Wei; Wu, Xuetong

2018-03-01

The target identification of the sea battlefield is the prerequisite for the judgment of the enemy in the modern naval battle. In this paper, a collaborative identification method based on convolution neural network is proposed to identify the typical targets of sea battlefields. Different from the traditional single-input/single-output identification method, the proposed method constructs a multi-input/single-output co-identification architecture based on optimized convolution neural network and weighted D-S evidence theory. The simulation results show that

Impact of defective pixels in AMLCDs on the perception of medical images

NASA Astrophysics Data System (ADS)

Kimpe, Tom; Sneyders, Yuri

2006-03-01

With LCD displays, each pixel has its own individual transistor that controls the transmittance of that pixel. Occasionally, these individual transistors will short or alternatively malfunction, resulting in a defective pixel that always shows the same brightness. With ever increasing resolution of displays the number of defect pixels per display increases accordingly. State of the art processes are capable of producing displays with no more than one faulty transistor out of 3 million. A five Mega Pixel medical LCD panel contains 15 million individual sub pixels (3 sub pixels per pixel), each having an individual transistor. This means that a five Mega Pixel display on average will have 5 failing pixels. This paper investigates the visibility of defective pixels and analyzes the possible impact of defective pixels on the perception of medical images. JND simulations were done to study the effect of defective pixels on medical images. Our results indicate that defective LCD pixels can mask subtle features in medical images in an unexpectedly broad area around the defect and therefore may reduce the quality of diagnosis for specific high-demanding areas such as mammography. As a second contribution an innovative solution is proposed. A specialized image processing algorithm can make defective pixels completely invisible and moreover can also recover the information of the defect so that the radiologist perceives the medical image correctly. This correction algorithm has been validated with both JND simulations and psycho visual tests.

A model of traffic signs recognition with convolutional neural network

NASA Astrophysics Data System (ADS)

Hu, Haihe; Li, Yujian; Zhang, Ting; Huo, Yi; Kuang, Wenqing

2016-10-01

In real traffic scenes, the quality of captured images are generally low due to some factors such as lighting conditions, and occlusion on. All of these factors are challengeable for automated recognition algorithms of traffic signs. Deep learning has provided a new way to solve this kind of problems recently. The deep network can automatically learn features from a large number of data samples and obtain an excellent recognition performance. We therefore approach this task of recognition of traffic signs as a general vision problem, with few assumptions related to road signs. We propose a model of Convolutional Neural Network (CNN) and apply the model to the task of traffic signs recognition. The proposed model adopts deep CNN as the supervised learning model, directly takes the collected traffic signs image as the input, alternates the convolutional layer and subsampling layer, and automatically extracts the features for the recognition of the traffic signs images. The proposed model includes an input layer, three convolutional layers, three subsampling layers, a fully-connected layer, and an output layer. To validate the proposed model, the experiments are implemented using the public dataset of China competition of fuzzy image processing. Experimental results show that the proposed model produces a recognition accuracy of 99.01 % on the training dataset, and yield a record of 92% on the preliminary contest within the fourth best.

Active pixel sensor pixel having a photodetector whose output is coupled to an output transistor gate

NASA Technical Reports Server (NTRS)

Fossum, Eric R. (Inventor); Nakamura, Junichi (Inventor); Kemeny, Sabrina E. (Inventor)

2005-01-01

An imaging device formed as a monolithic complementary metal oxide semiconductor integrated circuit in an industry standard complementary metal oxide semiconductor process, the integrated circuit including a focal plane array of pixel cells, each one of the cells including a photogate overlying the substrate for accumulating photo-generated charge in an underlying portion of the substrate and a charge coupled device section formed on the substrate adjacent the photogate having a sensing node and at least one charge coupled device stage for transferring charge from the underlying portion of the substrate to the sensing node. There is also a readout circuit, part of which can be disposed at the bottom of each column of cells and be common to all the cells in the column. A Simple Floating Gate (SFG) pixel structure could also be employed in the imager to provide a non-destructive readout and smaller pixel sizes.

Image Classification Using Biomimetic Pattern Recognition with Convolutional Neural Networks Features

PubMed Central

Huo, Guanying

2017-01-01

As a typical deep-learning model, Convolutional Neural Networks (CNNs) can be exploited to automatically extract features from images using the hierarchical structure inspired by mammalian visual system. For image classification tasks, traditional CNN models employ the softmax function for classification. However, owing to the limited capacity of the softmax function, there are some shortcomings of traditional CNN models in image classification. To deal with this problem, a new method combining Biomimetic Pattern Recognition (BPR) with CNNs is proposed for image classification. BPR performs class recognition by a union of geometrical cover sets in a high-dimensional feature space and therefore can overcome some disadvantages of traditional pattern recognition. The proposed method is evaluated on three famous image classification benchmarks, that is, MNIST, AR, and CIFAR-10. The classification accuracies of the proposed method for the three datasets are 99.01%, 98.40%, and 87.11%, respectively, which are much higher in comparison with the other four methods in most cases. PMID:28316614

Visual mining business service using pixel bar charts

NASA Astrophysics Data System (ADS)

Hao, Ming C.; Dayal, Umeshwar; Casati, Fabio

2004-06-01

Basic bar charts have been commonly available, but they only show highly aggregated data. Finding the valuable information hidden in the data is essential to the success of business. We describe a new visualization technique called pixel bar charts, which are derived from regular bar charts. The basic idea of a pixel bar chart is to present all data values directly instead of aggregating them into a few data values. Pixel bar charts provide data distribution and exceptions besides aggregated data. The approach is to represent each data item (e.g. a business transaction) by a single pixel in the bar chart. The attribute of each data item is encoded into the pixel color and can be accessed and drilled down to the detail information as needed. Different color mappings are used to represent multiple attributes. This technique has been prototyped in three business service applications-Business Operation Analysis, Sales Analysis, and Service Level Agreement Analysis at Hewlett Packard Laboratories. Our applications show the wide applicability and usefulness of this new idea.

CMOS VLSI Active-Pixel Sensor for Tracking

NASA Technical Reports Server (NTRS)

Pain, Bedabrata; Sun, Chao; Yang, Guang; Heynssens, Julie

2004-01-01

An architecture for a proposed active-pixel sensor (APS) and a design to implement the architecture in a complementary metal oxide semiconductor (CMOS) very-large-scale integrated (VLSI) circuit provide for some advanced features that are expected to be especially desirable for tracking pointlike features of stars. The architecture would also make this APS suitable for robotic- vision and general pointing and tracking applications. CMOS imagers in general are well suited for pointing and tracking because they can be configured for random access to selected pixels and to provide readout from windows of interest within their fields of view. However, until now, the architectures of CMOS imagers have not supported multiwindow operation or low-noise data collection. Moreover, smearing and motion artifacts in collected images have made prior CMOS imagers unsuitable for tracking applications. The proposed CMOS imager (see figure) would include an array of 1,024 by 1,024 pixels containing high-performance photodiode-based APS circuitry. The pixel pitch would be 9 m. The operations of the pixel circuits would be sequenced and otherwise controlled by an on-chip timing and control block, which would enable the collection of image data, during a single frame period, from either the full frame (that is, all 1,024 1,024 pixels) or from within as many as 8 different arbitrarily placed windows as large as 8 by 8 pixels each. A typical prior CMOS APS operates in a row-at-a-time ( grolling-shutter h) readout mode, which gives rise to exposure skew. In contrast, the proposed APS would operate in a sample-first/readlater mode, suppressing rolling-shutter effects. In this mode, the analog readout signals from the pixels corresponding to the windows of the interest (which windows, in the star-tracking application, would presumably contain guide stars) would be sampled rapidly by routing them through a programmable diagonal switch array to an on-chip parallel analog memory array. The

Highly Reflective Multi-stable Electrofluidic Display Pixels

NASA Astrophysics Data System (ADS)

Yang, Shu

Electronic papers (E-papers) refer to the displays that mimic the appearance of printed papers, but still owning the features of conventional electronic displays, such as the abilities of browsing websites and playing videos. The motivation of creating paper-like displays is inspired by the truths that reading on a paper caused least eye fatigue due to the paper's reflective and light diffusive nature, and, unlike the existing commercial displays, there is no cost of any form of energy for sustaining the displayed image. To achieve the equivalent visual effect of a paper print, an ideal E-paper has to be a highly reflective with good contrast ratio and full-color capability. To sustain the image with zero power consumption, the display pixels need to be bistable, which means the "on" and "off" states are both lowest energy states. Pixel can change its state only when sufficient external energy is given. There are many emerging technologies competing to demonstrate the first ideal E-paper device. However, none is able to achieve satisfactory visual effect, bistability and video speed at the same time. Challenges come from either the inherent physical/chemical properties or the fabrication process. Electrofluidic display is one of the most promising E-paper technologies. It has successfully demonstrated high reflectivity, brilliant color and video speed operation by moving colored pigment dispersion between visible and invisible places with electrowetting force. However, the pixel design did not allow the image bistability. Presented in this dissertation are the multi-stable electrofluidic display pixels that are able to sustain grayscale levels without any power consumption, while keeping the favorable features of the previous generation electrofluidic display. The pixel design, fabrication method using multiple layer dry film photoresist lamination, and physical/optical characterizations are discussed in details. Based on the pixel structure, the preliminary

Automatic Fabric Defect Detection with a Multi-Scale Convolutional Denoising Autoencoder Network Model.

PubMed

Mei, Shuang; Wang, Yudan; Wen, Guojun

2018-04-02

Fabric defect detection is a necessary and essential step of quality control in the textile manufacturing industry. Traditional fabric inspections are usually performed by manual visual methods, which are low in efficiency and poor in precision for long-term industrial applications. In this paper, we propose an unsupervised learning-based automated approach to detect and localize fabric defects without any manual intervention. This approach is used to reconstruct image patches with a convolutional denoising autoencoder network at multiple Gaussian pyramid levels and to synthesize detection results from the corresponding resolution channels. The reconstruction residual of each image patch is used as the indicator for direct pixel-wise prediction. By segmenting and synthesizing the reconstruction residual map at each resolution level, the final inspection result can be generated. This newly developed method has several prominent advantages for fabric defect detection. First, it can be trained with only a small amount of defect-free samples. This is especially important for situations in which collecting large amounts of defective samples is difficult and impracticable. Second, owing to the multi-modal integration strategy, it is relatively more robust and accurate compared to general inspection methods (the results at each resolution level can be viewed as a modality). Third, according to our results, it can address multiple types of textile fabrics, from simple to more complex. Experimental results demonstrate that the proposed model is robust and yields good overall performance with high precision and acceptable recall rates.

Finding strong lenses in CFHTLS using convolutional neural networks

NASA Astrophysics Data System (ADS)

Jacobs, C.; Glazebrook, K.; Collett, T.; More, A.; McCarthy, C.

2017-10-01

We train and apply convolutional neural networks, a machine learning technique developed to learn from and classify image data, to Canada-France-Hawaii Telescope Legacy Survey (CFHTLS) imaging for the identification of potential strong lensing systems. An ensemble of four convolutional neural networks was trained on images of simulated galaxy-galaxy lenses. The training sets consisted of a total of 62 406 simulated lenses and 64 673 non-lens negative examples generated with two different methodologies. An ensemble of trained networks was applied to all of the 171 deg2 of the CFHTLS wide field image data, identifying 18 861 candidates including 63 known and 139 other potential lens candidates. A second search of 1.4 million early-type galaxies selected from the survey catalogue as potential deflectors, identified 2465 candidates including 117 previously known lens candidates, 29 confirmed lenses/high-quality lens candidates, 266 novel probable or potential lenses and 2097 candidates we classify as false positives. For the catalogue-based search we estimate a completeness of 21-28 per cent with respect to detectable lenses and a purity of 15 per cent, with a false-positive rate of 1 in 671 images tested. We predict a human astronomer reviewing candidates produced by the system would identify 20 probable lenses and 100 possible lenses per hour in a sample selected by the robot. Convolutional neural networks are therefore a promising tool for use in the search for lenses in current and forthcoming surveys such as the Dark Energy Survey and the Large Synoptic Survey Telescope.

A unitary convolution approximation for the impact-parameter dependent electronic energy loss

NASA Astrophysics Data System (ADS)

Schiwietz, G.; Grande, P. L.

1999-06-01

In this work, we propose a simple method to calculate the impact-parameter dependence of the electronic energy loss of bare ions for all impact parameters. This perturbative convolution approximation (PCA) is based on first-order perturbation theory, and thus, it is only valid for fast particles with low projectile charges. Using Bloch's stopping-power result and a simple scaling, we get rid of the restriction to low charge states and derive the unitary convolution approximation (UCA). Results of the UCA are then compared with full quantum-mechanical coupled-channel calculations for the impact-parameter dependent electronic energy loss.

Access To The PMM's Pixel Database

NASA Astrophysics Data System (ADS)

Monet, D.; Levine, S.

1999-12-01

The U.S. Naval Observatory Flagstaff Station is in the process of enabling access to the Precision Measuring Machine (PMM) program's pixel database. The initial release will include the pixels from the PMM's scans of the Palomar Observatory Sky Survey I (POSS-I) -O and -E surveys, the Whiteoak Extension, the European Southern Observatory-R survey, the Science and Engineering Council-J, -EJ, and -ER surveys, and the Anglo- Australian Observatory-R survey. (The SERC-ER and AAO-R surveys are currently incomplete.) As time allows, access to the POSS-II -J, -F, and -N surveys, the Palomar Infrared Milky Way Atlas, the Yale/San Juan Southern Proper Motion survey, and plates rejected by various surveys will be added. (POSS-II -J and -F are complete, but -N was never finished.) Eventually, some 10 Tbytes of pixel data will be available. Due to funding and technology limitations, the initial interface will have only limited functionality, and access time will be slow since the archive is stored on Digital Linear Tape (DLT). Usage of the pixel data will be restricted to non-commercial, scientific applications, and agreements on copyright issues have yet to be finalized. The poster presentation will give the URL.

Encoding Sequential Information in Semantic Space Models: Comparing Holographic Reduced Representation and Random Permutation

PubMed Central

Recchia, Gabriel; Sahlgren, Magnus; Kanerva, Pentti; Jones, Michael N.

2015-01-01

Circular convolution and random permutation have each been proposed as neurally plausible binding operators capable of encoding sequential information in semantic memory. We perform several controlled comparisons of circular convolution and random permutation as means of encoding paired associates as well as encoding sequential information. Random permutations outperformed convolution with respect to the number of paired associates that can be reliably stored in a single memory trace. Performance was equal on semantic tasks when using a small corpus, but random permutations were ultimately capable of achieving superior performance due to their higher scalability to large corpora. Finally, “noisy” permutations in which units are mapped to other units arbitrarily (no one-to-one mapping) perform nearly as well as true permutations. These findings increase the neurological plausibility of random permutations and highlight their utility in vector space models of semantics. PMID:25954306

Classification of stroke disease using convolutional neural network

NASA Astrophysics Data System (ADS)

Marbun, J. T.; Seniman; Andayani, U.

2018-03-01

Stroke is a condition that occurs when the blood supply stop flowing to the brain because of a blockage or a broken blood vessel. A symptoms that happen when experiencing stroke, some of them is a dropped consciousness, disrupted vision and paralyzed body. The general examination is being done to get a picture of the brain part that have stroke using Computerized Tomography (CT) Scan. The image produced from CT will be manually checked and need a proper lighting by doctor to get a type of stroke. That is why it needs a method to classify stroke from CT image automatically. A method proposed in this research is Convolutional Neural Network. CT image of the brain is used as the input for image processing. The stage before classification are image processing (Grayscaling, Scaling, Contrast Limited Adaptive Histogram Equalization, then the image being classified with Convolutional Neural Network. The result then showed that the method significantly conducted was able to be used as a tool to classify stroke disease in order to distinguish the type of stroke from CT image.

Evaluation of a single-pixel one-transistor active pixel sensor for fingerprint imaging

NASA Astrophysics Data System (ADS)

Xu, Man; Ou, Hai; Chen, Jun; Wang, Kai

2015-08-01

Since it first appeared in iPhone 5S in 2013, fingerprint identification (ID) has rapidly gained popularity among consumers. Current fingerprint-enabled smartphones unanimously consists of a discrete sensor to perform fingerprint ID. This architecture not only incurs higher material and manufacturing cost, but also provides only static identification and limited authentication. Hence as the demand for a thinner, lighter, and more secure handset grows, we propose a novel pixel architecture that is a photosensitive device embedded in a display pixel and detects the reflected light from the finger touch for high resolution, high fidelity and dynamic biometrics. To this purpose, an amorphous silicon (a-Si:H) dual-gate photo TFT working in both fingerprint-imaging mode and display-driving mode will be developed.

Microradiography with Semiconductor Pixel Detectors

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

Jakubek, Jan; Cejnarova, Andrea; Dammer, Jiri

High resolution radiography (with X-rays, neutrons, heavy charged particles, ...) often exploited also in tomographic mode to provide 3D images stands as a powerful imaging technique for instant and nondestructive visualization of fine internal structure of objects. Novel types of semiconductor single particle counting pixel detectors offer many advantages for radiation imaging: high detection efficiency, energy discrimination or direct energy measurement, noiseless digital integration (counting), high frame rate and virtually unlimited dynamic range. This article shows the application and potential of pixel detectors (such as Medipix2 or TimePix) in different fields of radiation imaging.

Research on ionospheric tomography based on variable pixel height

NASA Astrophysics Data System (ADS)

Zheng, Dunyong; Li, Peiqing; He, Jie; Hu, Wusheng; Li, Chaokui

2016-05-01

A novel ionospheric tomography technique based on variable pixel height was developed for the tomographic reconstruction of the ionospheric electron density distribution. The method considers the height of each pixel as an unknown variable, which is retrieved during the inversion process together with the electron density values. In contrast to conventional computerized ionospheric tomography (CIT), which parameterizes the model with a fixed pixel height, the variable-pixel-height computerized ionospheric tomography (VHCIT) model applies a disturbance to the height of each pixel. In comparison with conventional CIT models, the VHCIT technique achieved superior results in a numerical simulation. A careful validation of the reliability and superiority of VHCIT was performed. According to the results of the statistical analysis of the average root mean square errors, the proposed model offers an improvement by 15% compared with conventional CIT models.

Deep learning in breast cancer risk assessment: evaluation of fine-tuned convolutional neural networks on a clinical dataset of FFDMs

NASA Astrophysics Data System (ADS)

Li, Hui; Mendel, Kayla R.; Lee, John H.; Lan, Li; Giger, Maryellen L.

2018-02-01

We evaluated the potential of deep learning in the assessment of breast cancer risk using convolutional neural networks (CNNs) fine-tuned on full-field digital mammographic (FFDM) images. This study included 456 clinical FFDM cases from two high-risk datasets: BRCA1/2 gene-mutation carriers (53 cases) and unilateral cancer patients (75 cases), and a low-risk dataset as the control group (328 cases). All FFDM images (12-bit quantization and 100 micron pixel) were acquired with a GE Senographe 2000D system and were retrospectively collected under an IRB-approved, HIPAA-compliant protocol. Regions of interest of 256x256 pixels were selected from the central breast region behind the nipple in the craniocaudal projection. VGG19 pre-trained on the ImageNet dataset was used to classify the images either as high-risk or as low-risk subjects. The last fully-connected layer of pre-trained VGG19 was fine-tuned on FFDM images for breast cancer risk assessment. Performance was evaluated using the area under the receiver operating characteristic (ROC) curve (AUC) in the task of distinguishing between high-risk and low-risk subjects. AUC values of 0.84 (SE=0.05) and 0.72 (SE=0.06) were obtained in the task of distinguishing between the BRCA1/2 gene-mutation carriers and low-risk women and between unilateral cancer patients and low-risk women, respectively. Deep learning with CNNs appears to be able to extract parenchymal characteristics directly from FFDMs which are relevant to the task of distinguishing between cancer risk populations, and therefore has potential to aid clinicians in assessing mammographic parenchymal patterns for cancer risk assessment.

Offline signature verification using convolution Siamese network

NASA Astrophysics Data System (ADS)

Xing, Zi-Jian; Yin, Fei; Wu, Yi-Chao; Liu, Cheng-Lin

2018-04-01

This paper presents an offline signature verification approach using convolutional Siamese neural network. Unlike the existing methods which consider feature extraction and metric learning as two independent stages, we adopt a deepleaning based framework which combines the two stages together and can be trained end-to-end. The experimental results on two offline public databases (GPDSsynthetic and CEDAR) demonstrate the superiority of our method on the offline signature verification problem.

Mapping Capacitive Coupling Among Pixels in a Sensor Array

NASA Technical Reports Server (NTRS)

Seshadri, Suresh; Cole, David M.; Smith, Roger M.

2010-01-01

An improved method of mapping the capacitive contribution to cross-talk among pixels in an imaging array of sensors (typically, an imaging photodetector array) has been devised for use in calibrating and/or characterizing such an array. The method involves a sequence of resets of subarrays of pixels to specified voltages and measurement of the voltage responses of neighboring non-reset pixels.

Two-level convolution formula for nuclear structure function

NASA Astrophysics Data System (ADS)

Ma, Boqiang

1990-05-01

A two-level convolution formula for the nuclear structure function is derived in considering the nucleus as a composite system of baryon-mesons which are also composite systems of quark-gluons again. The results show that the European Muon Colaboration effect can not be explained by the nuclear effects as nucleon Fermi motion and nuclear binding contributions.

Self-adjusting threshold mechanism for pixel detectors

NASA Astrophysics Data System (ADS)

Heim, Timon; Garcia-Sciveres, Maurice

2017-09-01

Readout chips of hybrid pixel detectors use a low power amplifier and threshold discrimination to process charge deposited in semiconductor sensors. Due to transistor mismatch each pixel circuit needs to be calibrated individually to achieve response uniformity. Traditionally this is addressed by programmable threshold trimming in each pixel, but requires robustness against radiation effects, temperature, and time. In this paper a self-adjusting threshold mechanism is presented, which corrects the threshold for both spatial inequality and time variation and maintains a constant response. It exploits the electrical noise as relative measure for the threshold and automatically adjust the threshold of each pixel to always achieve a uniform frequency of noise hits. A digital implementation of the method in the form of an up/down counter and combinatorial logic filter is presented. The behavior of this circuit has been simulated to evaluate its performance and compare it to traditional calibration results. The simulation results show that this mechanism can perform equally well, but eliminates instability over time and is immune to single event upsets.

A multi-scale convolutional neural network for phenotyping high-content cellular images.

PubMed

Godinez, William J; Hossain, Imtiaz; Lazic, Stanley E; Davies, John W; Zhang, Xian

2017-07-01

Identifying phenotypes based on high-content cellular images is challenging. Conventional image analysis pipelines for phenotype identification comprise multiple independent steps, with each step requiring method customization and adjustment of multiple parameters. Here, we present an approach based on a multi-scale convolutional neural network (M-CNN) that classifies, in a single cohesive step, cellular images into phenotypes by using directly and solely the images' pixel intensity values. The only parameters in the approach are the weights of the neural network, which are automatically optimized based on training images. The approach requires no a priori knowledge or manual customization, and is applicable to single- or multi-channel images displaying single or multiple cells. We evaluated the classification performance of the approach on eight diverse benchmark datasets. The approach yielded overall a higher classification accuracy compared with state-of-the-art results, including those of other deep CNN architectures. In addition to using the network to simply obtain a yes-or-no prediction for a given phenotype, we use the probability outputs calculated by the network to quantitatively describe the phenotypes. This study shows that these probability values correlate with chemical treatment concentrations. This finding validates further our approach and enables chemical treatment potency estimation via CNNs. The network specifications and solver definitions are provided in Supplementary Software 1. william_jose.godinez_navarro@novartis.com or xian-1.zhang@novartis.com. Supplementary data are available at Bioinformatics online. © The Author 2017. Published by Oxford University Press. All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com

SVGA AMOLED with world's highest pixel pitch

NASA Astrophysics Data System (ADS)

Prache, Olivier; Wacyk, Ihor

2006-05-01

We present the design and early evaluation results of the world's highest pixel pitch full-color 800x3x600- pixel, active matrix organic light emitting diode (AMOLED) color microdisplay for consumer and environmentally demanding applications. The design premises were aimed at improving small area uniformity as well as reducing the pixel size while expanding the functionality found in existing eMagin Corporations' microdisplay products without incurring any power consumption degradation when compared to existing OLED microdisplays produced by eMagin. The initial results of the first silicon prototype presented here demonstrate compliance with all major objectives as well as the validation of a new adaptive gamma correction technique that can operate automatically over temperature.

Comparison of the convolution quadrature method and enhanced inverse FFT with application in elastodynamic boundary element method

NASA Astrophysics Data System (ADS)

Schanz, Martin; Ye, Wenjing; Xiao, Jinyou

2016-04-01

Transient problems can often be solved with transformation methods, where the inverse transformation is usually performed numerically. Here, the discrete Fourier transform in combination with the exponential window method is compared with the convolution quadrature method formulated as inverse transformation. Both are inverse Laplace transforms, which are formally identical but use different complex frequencies. A numerical study is performed, first with simple convolution integrals and, second, with a boundary element method (BEM) for elastodynamics. Essentially, when combined with the BEM, the discrete Fourier transform needs less frequency calculations, but finer mesh compared to the convolution quadrature method to obtain the same level of accuracy. If further fast methods like the fast multipole method are used to accelerate the boundary element method the convolution quadrature method is better, because the iterative solver needs much less iterations to converge. This is caused by the larger real part of the complex frequencies necessary for the calculation, which improves the conditions of system matrix.

A convolution model for computing the far-field directivity of a parametric loudspeaker array.

PubMed

Shi, Chuang; Kajikawa, Yoshinobu

2015-02-01

This paper describes a method to compute the far-field directivity of a parametric loudspeaker array (PLA), whereby the steerable parametric loudspeaker can be implemented when phased array techniques are applied. The convolution of the product directivity and the Westervelt's directivity is suggested, substituting for the past practice of using the product directivity only. Computed directivity of a PLA using the proposed convolution model achieves significant improvement in agreement to measured directivity at a negligible computational cost.

Improving deep convolutional neural networks with mixed maxout units.

PubMed

Zhao, Hui-Zhen; Liu, Fu-Xian; Li, Long-Yue

2017-01-01

Motivated by insights from the maxout-units-based deep Convolutional Neural Network (CNN) that "non-maximal features are unable to deliver" and "feature mapping subspace pooling is insufficient," we present a novel mixed variant of the recently introduced maxout unit called a mixout unit. Specifically, we do so by calculating the exponential probabilities of feature mappings gained by applying different convolutional transformations over the same input and then calculating the expected values according to their exponential probabilities. Moreover, we introduce the Bernoulli distribution to balance the maximum values with the expected values of the feature mappings subspace. Finally, we design a simple model to verify the pooling ability of mixout units and a Mixout-units-based Network-in-Network (NiN) model to analyze the feature learning ability of the mixout models. We argue that our proposed units improve the pooling ability and that mixout models can achieve better feature learning and classification performance.

Improving deep convolutional neural networks with mixed maxout units

PubMed Central

Liu, Fu-xian; Li, Long-yue

2017-01-01

Motivated by insights from the maxout-units-based deep Convolutional Neural Network (CNN) that “non-maximal features are unable to deliver” and “feature mapping subspace pooling is insufficient,” we present a novel mixed variant of the recently introduced maxout unit called a mixout unit. Specifically, we do so by calculating the exponential probabilities of feature mappings gained by applying different convolutional transformations over the same input and then calculating the expected values according to their exponential probabilities. Moreover, we introduce the Bernoulli distribution to balance the maximum values with the expected values of the feature mappings subspace. Finally, we design a simple model to verify the pooling ability of mixout units and a Mixout-units-based Network-in-Network (NiN) model to analyze the feature learning ability of the mixout models. We argue that our proposed units improve the pooling ability and that mixout models can achieve better feature learning and classification performance. PMID:28727737

Chemical-induced disease relation extraction via convolutional neural network.

PubMed

Gu, Jinghang; Sun, Fuqing; Qian, Longhua; Zhou, Guodong

2017-01-01

This article describes our work on the BioCreative-V chemical-disease relation (CDR) extraction task, which employed a maximum entropy (ME) model and a convolutional neural network model for relation extraction at inter- and intra-sentence level, respectively. In our work, relation extraction between entity concepts in documents was simplified to relation extraction between entity mentions. We first constructed pairs of chemical and disease mentions as relation instances for training and testing stages, then we trained and applied the ME model and the convolutional neural network model for inter- and intra-sentence level, respectively. Finally, we merged the classification results from mention level to document level to acquire the final relations between chemical and disease concepts. The evaluation on the BioCreative-V CDR corpus shows the effectiveness of our proposed approach. http://www.biocreative.org/resources/corpora/biocreative-v-cdr-corpus/. © The Author(s) 2017. Published by Oxford University Press.

Pixel-By Estimation of Scene Motion in Video

NASA Astrophysics Data System (ADS)

Tashlinskii, A. G.; Smirnov, P. V.; Tsaryov, M. G.

2017-05-01

The paper considers the effectiveness of motion estimation in video using pixel-by-pixel recurrent algorithms. The algorithms use stochastic gradient decent to find inter-frame shifts of all pixels of a frame. These vectors form shift vectors' field. As estimated parameters of the vectors the paper studies their projections and polar parameters. It considers two methods for estimating shift vectors' field. The first method uses stochastic gradient descent algorithm to sequentially process all nodes of the image row-by-row. It processes each row bidirectionally i.e. from the left to the right and from the right to the left. Subsequent joint processing of the results allows compensating inertia of the recursive estimation. The second method uses correlation between rows to increase processing efficiency. It processes rows one after the other with the change in direction after each row and uses obtained values to form resulting estimate. The paper studies two criteria of its formation: gradient estimation minimum and correlation coefficient maximum. The paper gives examples of experimental results of pixel-by-pixel estimation for a video with a moving object and estimation of a moving object trajectory using shift vectors' field.

Transfer Learning with Convolutional Neural Networks for Classification of Abdominal Ultrasound Images.

PubMed

Cheng, Phillip M; Malhi, Harshawn S

2017-04-01

The purpose of this study is to evaluate transfer learning with deep convolutional neural networks for the classification of abdominal ultrasound images. Grayscale images from 185 consecutive clinical abdominal ultrasound studies were categorized into 11 categories based on the text annotation specified by the technologist for the image. Cropped images were rescaled to 256 × 256 resolution and randomized, with 4094 images from 136 studies constituting the training set, and 1423 images from 49 studies constituting the test set. The fully connected layers of two convolutional neural networks based on CaffeNet and VGGNet, previously trained on the 2012 Large Scale Visual Recognition Challenge data set, were retrained on the training set. Weights in the convolutional layers of each network were frozen to serve as fixed feature extractors. Accuracy on the test set was evaluated for each network. A radiologist experienced in abdominal ultrasound also independently classified the images in the test set into the same 11 categories. The CaffeNet network classified 77.3% of the test set images accurately (1100/1423 images), with a top-2 accuracy of 90.4% (1287/1423 images). The larger VGGNet network classified 77.9% of the test set accurately (1109/1423 images), with a top-2 accuracy of VGGNet was 89.7% (1276/1423 images). The radiologist classified 71.7% of the test set images correctly (1020/1423 images). The differences in classification accuracies between both neural networks and the radiologist were statistically significant (p < 0.001). The results demonstrate that transfer learning with convolutional neural networks may be used to construct effective classifiers for abdominal ultrasound images.

A pre-trained convolutional neural network based method for thyroid nodule diagnosis.

PubMed

Ma, Jinlian; Wu, Fa; Zhu, Jiang; Xu, Dong; Kong, Dexing

2017-01-01

In ultrasound images, most thyroid nodules are in heterogeneous appearances with various internal components and also have vague boundaries, so it is difficult for physicians to discriminate malignant thyroid nodules from benign ones. In this study, we propose a hybrid method for thyroid nodule diagnosis, which is a fusion of two pre-trained convolutional neural networks (CNNs) with different convolutional layers and fully-connected layers. Firstly, the two networks pre-trained with ImageNet database are separately trained. Secondly, we fuse feature maps learned by trained convolutional filters, pooling and normalization operations of the two CNNs. Finally, with the fused feature maps, a softmax classifier is used to diagnose thyroid nodules. The proposed method is validated on 15,000 ultrasound images collected from two local hospitals. Experiment results show that the proposed CNN based methods can accurately and effectively diagnose thyroid nodules. In addition, the fusion of the two CNN based models lead to significant performance improvement, with an accuracy of 83.02%±0.72%. These demonstrate the potential clinical applications of this method. Copyright © 2016 Elsevier B.V. All rights reserved.

WFC3 UVIS Pixel-to-Pixel QE Variations via Internal Flats Monitor

NASA Astrophysics Data System (ADS)

Mckay, Myles

2017-08-01

The UVIS detector has a population of pixels that exhibit anomalous QE variations between anneals, characterized by a sensitivity loss that is greater in the shorter wavelengths(blue) than in the longer wavelengths(red). This population is distributed randomly, with evidence of grouping behavior in the UV, and is seemingly different for each anneal cycle. This program, a continuation of cycle 24 program 14546, will aim to constrain the maximum low-sensitivity population existing before an anneal in both the UV and Visible filters. To monitor the UV behavior, internal flats with the Deuterium D2 lamp will be taken through the filters, F225W and F336W. To monitor the behavior in the Visible filters, internal flats with the tungsten lamp will be taken through F814W and F438W a week before the anneal, when the population of anomalous pixels is the greatest.

Improving Pixel Level Cloud Optical Property Retrieval using Monte Carlo Simulations

NASA Technical Reports Server (NTRS)

Oreopoulos, Lazaros; Marshak, Alexander; Cahalan, Robert F.

1999-01-01

The accurate pixel-by-pixel retrieval of cloud optical properties from space is influenced by radiative smoothing due to high order photon scattering and radiative roughening due to low order scattering events. Both are caused by cloud heterogeneity and the three-dimensional nature of radiative transfer and can be studied with the aid of computer simulations. We use Monte Carlo simulations on variable 1-D and 2-D model cloud fields to seek for dependencies of smoothing and roughening phenomena on single scattering albedo, solar zenith angle, and cloud characteristics. The results are discussed in the context of high resolution satellite (such as Landsat) retrieval applications. The current work extends the investigation on the inverse NIPA (Non-local Independent Pixel Approximation) as a tool for removing smoothing and improving retrievals of cloud optical depth. This is accomplished by: (1) Delineating the limits of NIPA applicability; (2) Exploring NIPA parameter dependences on cloud macrostructural features, such as mean cloud optical depth and geometrical thickness, degree of extinction and cloud top height variability. We also compare parameter values from empirical and theoretical considerations; (3) Examining the differences between applying NIPA on radiation quantities vs direct application on optical properties; (4) Studying the radiation budget importance of the NIPA corrections as a function of scale. Finally, we discuss fundamental adjustments that need to be considered for successful radiance inversion at non-conservative wavelengths and oblique Sun angles. These adjustments are necessary to remove roughening signatures which become more prominent with increasing absorption and solar zenith angle.

Model-based optimization of near-field binary-pixelated beam shapers

DOE PAGES

Dorrer, C.; Hassett, J.

2017-01-23

The optimization of components that rely on spatially dithered distributions of transparent or opaque pixels and an imaging system with far-field filtering for transmission control is demonstrated. The binary-pixel distribution can be iteratively optimized to lower an error function that takes into account the design transmission and the characteristics of the required far-field filter. Simulations using a design transmission chosen in the context of high-energy lasers show that the beam-fluence modulation at an image plane can be reduced by a factor of 2, leading to performance similar to using a non-optimized spatial-dithering algorithm with pixels of size reduced by amore » factor of 2 without the additional fabrication complexity or cost. The optimization process preserves the pixel distribution statistical properties. Analysis shows that the optimized pixel distribution starting from a high-noise distribution defined by a random-draw algorithm should be more resilient to fabrication errors than the optimized pixel distributions starting from a low-noise, error-diffusion algorithm, while leading to similar beamshaping performance. Furthermore, this is confirmed by experimental results obtained with various pixel distributions and induced fabrication errors.« less

Human vision-based algorithm to hide defective pixels in LCDs

NASA Astrophysics Data System (ADS)

Kimpe, Tom; Coulier, Stefaan; Van Hoey, Gert

2006-02-01

Producing displays without pixel defects or repairing defective pixels is technically not possible at this moment. This paper presents a new approach to solve this problem: defects are made invisible for the user by using image processing algorithms based on characteristics of the human eye. The performance of this new algorithm has been evaluated using two different methods. First of all the theoretical response of the human eye was analyzed on a series of images and this before and after applying the defective pixel compensation algorithm. These results show that indeed it is possible to mask a defective pixel. A second method was to perform a psycho-visual test where users were asked whether or not a defective pixel could be perceived. The results of these user tests also confirm the value of the new algorithm. Our "defective pixel correction" algorithm can be implemented very efficiently and cost-effectively as pixel-dataprocessing algorithms inside the display in for instance an FPGA, a DSP or a microprocessor. The described techniques are also valid for both monochrome and color displays ranging from high-quality medical displays to consumer LCDTV applications.

3D reconstructions with pixel-based images are made possible by digitally clearing plant and animal tissue

USDA-ARS?s Scientific Manuscript database

Reconstruction of 3D images from a series of 2D images has been restricted by the limited capacity to decrease the opacity of surrounding tissue. Commercial software that allows color-keying and manipulation of 2D images in true 3D space allowed us to produce 3D reconstructions from pixel based imag...

Accelerating Convolutional Sparse Coding for Curvilinear Structures Segmentation by Refining SCIRD-TS Filter Banks.

PubMed

Annunziata, Roberto; Trucco, Emanuele

2016-11-01

Deep learning has shown great potential for curvilinear structure (e.g., retinal blood vessels and neurites) segmentation as demonstrated by a recent auto-context regression architecture based on filter banks learned by convolutional sparse coding. However, learning such filter banks is very time-consuming, thus limiting the amount of filters employed and the adaptation to other data sets (i.e., slow re-training). We address this limitation by proposing a novel acceleration strategy to speed-up convolutional sparse coding filter learning for curvilinear structure segmentation. Our approach is based on a novel initialisation strategy (warm start), and therefore it is different from recent methods improving the optimisation itself. Our warm-start strategy is based on carefully designed hand-crafted filters (SCIRD-TS), modelling appearance properties of curvilinear structures which are then refined by convolutional sparse coding. Experiments on four diverse data sets, including retinal blood vessels and neurites, suggest that the proposed method reduces significantly the time taken to learn convolutional filter banks (i.e., up to -82%) compared to conventional initialisation strategies. Remarkably, this speed-up does not worsen performance; in fact, filters learned with the proposed strategy often achieve a much lower reconstruction error and match or exceed the segmentation performance of random and DCT-based initialisation, when used as input to a random forest classifier.

An Interactive Graphics Program for Assistance in Learning Convolution.

ERIC Educational Resources Information Center

Frederick, Dean K.; Waag, Gary L.

1980-01-01

A program has been written for the interactive computer graphics facility at Rensselaer Polytechnic Institute that is designed to assist the user in learning the mathematical technique of convolving two functions. Because convolution can be represented graphically by a sequence of steps involving folding, shifting, multiplying, and integration, it…

Monolithic Active Pixel Sensors

NASA Astrophysics Data System (ADS)

Lutz, P.

In close collaboration with the group from Strasbourg, Saclay has been developing fast monolithic active pixel sensors for future vertex detectors. This presentation gives some recent results from the MIMOSA serie, emphazising the participation of the group.

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

Readout and DAQ for Pixel Detectors

NASA Astrophysics Data System (ADS)

Platkevic, Michal

2010-01-01

Data readout and acquisition control of pixel detectors demand the transfer of significantly a large amounts of bits between the detector and the computer. For this purpose dedicated interfaces are used which are designed with focus on features like speed, small dimensions or flexibility of use such as digital signal processors, field-programmable gate arrays (FPGA) and USB communication ports. This work summarizes the readout and DAQ system built for state-of-the-art pixel detectors of the Medipix family.

Realistic full wave modeling of focal plane array pixels

DOE PAGES

Campione, Salvatore; Warne, Larry K.; Jorgenson, Roy E.; ...

2017-11-01

Here, we investigate full-wave simulations of realistic implementations of multifunctional nanoantenna enabled detectors (NEDs). We focus on a 2x2 pixelated array structure that supports two wavelengths of operation. We design each resonating structure independently using full-wave simulations with periodic boundary conditions mimicking the whole infinite array. We then construct a supercell made of a 2x2 pixelated array with periodic boundary conditions mimicking the full NED; in this case, however, each pixel comprises 10-20 antennas per side. In this way, the cross-talk between contiguous pixels is accounted for in our simulations. We observe that, even though there are finite extent effects,more » the pixels work as designed, each responding at the respective wavelength of operation. This allows us to stress that realistic simulations of multifunctional NEDs need to be performed to verify the design functionality by taking into account finite extent and cross-talk effects.« less

Applications of deep convolutional neural networks to digitized natural history collections.

PubMed

Schuettpelz, Eric; Frandsen, Paul B; Dikow, Rebecca B; Brown, Abel; Orli, Sylvia; Peters, Melinda; Metallo, Adam; Funk, Vicki A; Dorr, Laurence J

2017-01-01

Natural history collections contain data that are critical for many scientific endeavors. Recent efforts in mass digitization are generating large datasets from these collections that can provide unprecedented insight. Here, we present examples of how deep convolutional neural networks can be applied in analyses of imaged herbarium specimens. We first demonstrate that a convolutional neural network can detect mercury-stained specimens across a collection with 90% accuracy. We then show that such a network can correctly distinguish two morphologically similar plant families 96% of the time. Discarding the most challenging specimen images increases accuracy to 94% and 99%, respectively. These results highlight the importance of mass digitization and deep learning approaches and reveal how they can together deliver powerful new investigative tools.

Applications of deep convolutional neural networks to digitized natural history collections

PubMed Central

Frandsen, Paul B.; Dikow, Rebecca B.; Brown, Abel; Orli, Sylvia; Peters, Melinda; Metallo, Adam; Funk, Vicki A.; Dorr, Laurence J.

2017-01-01

Abstract Natural history collections contain data that are critical for many scientific endeavors. Recent efforts in mass digitization are generating large datasets from these collections that can provide unprecedented insight. Here, we present examples of how deep convolutional neural networks can be applied in analyses of imaged herbarium specimens. We first demonstrate that a convolutional neural network can detect mercury-stained specimens across a collection with 90% accuracy. We then show that such a network can correctly distinguish two morphologically similar plant families 96% of the time. Discarding the most challenging specimen images increases accuracy to 94% and 99%, respectively. These results highlight the importance of mass digitization and deep learning approaches and reveal how they can together deliver powerful new investigative tools. PMID:29200929

Active Pixel Sensors: Are CCD's Dinosaurs?

NASA Technical Reports Server (NTRS)

Fossum, Eric R.

1993-01-01

Charge-coupled devices (CCD's) are presently the technology of choice for most imaging applications. In the 23 years since their invention in 1970, they have evolved to a sophisticated level of performance. However, as with all technologies, we can be certain that they will be supplanted someday. In this paper, the Active Pixel Sensor (APS) technology is explored as a possible successor to the CCD. An active pixel is defined as a detector array technology that has at least one active transistor within the pixel unit cell. The APS eliminates the need for nearly perfect charge transfer -- the Achilles' heel of CCDs. This perfect charge transfer makes CCD's radiation 'soft,' difficult to use under low light conditions, difficult to manufacture in large array sizes, difficult to integrate with on-chip electronics, difficult to use at low temperatures, difficult to use at high frame rates, and difficult to manufacture in non-silicon materials that extend wavelength response.

Bio-Inspired Asynchronous Pixel Event Tricolor Vision Sensor.

PubMed

Lenero-Bardallo, Juan Antonio; Bryn, D H; Hafliger, Philipp

2014-06-01

This article investigates the potential of the first ever prototype of a vision sensor that combines tricolor stacked photo diodes with the bio-inspired asynchronous pixel event communication protocol known as Address Event Representation (AER). The stacked photo diodes are implemented in a 22 × 22 pixel array in a standard STM 90 nm CMOS process. Dynamic range is larger than 60 dB and pixels fill factor is 28%. The pixels employ either simple pulse frequency modulation (PFM) or a Time-to-First-Spike (TFS) mode. A heuristic linear combination of the chip's inherent pseudo colors serves to approximate RGB color representation. Furthermore, the sensor outputs can be processed to represent the radiation in the near infrared (NIR) band without employing external filters, and to color-encode direction of motion due to an asymmetry in the update rates of the different diode layers.

Performance comparison of small-pixel CdZnTe radiation detectors with gold contacts formed by sputter and electroless deposition

NASA Astrophysics Data System (ADS)

Bell, S. J.; Baker, M. A.; Duarte, D. D.; Schneider, A.; Seller, P.; Sellin, P. J.; Veale, M. C.; Wilson, M. D.

2017-06-01

Recent improvements in the growth of wide-bandgap semiconductors, such as cadmium zinc telluride (CdZnTe or CZT), has enabled spectroscopic X/γ-ray imaging detectors to be developed. These detectors have applications covering homeland security, industrial analysis, space science and medical imaging. At the Rutherford Appleton Laboratory (RAL) a promising range of spectroscopic, position sensitive, small-pixel Cd(Zn)Te detectors have been developed. The challenge now is to improve the quality of metal contacts on CdZnTe in order to meet the demanding energy and spatial resolution requirements of these applications. The choice of metal deposition method and fabrication process are of fundamental importance. Presented is a comparison of two CdZnTe detectors with contacts formed by sputter and electroless deposition. The detectors were fabricated with a 74 × 74 array of 200 μm pixels on a 250 μm pitch and bump-bonded to the HEXITEC ASIC. The X/γ-ray emissions from an 241Am source were measured to form energy spectra for comparison. It was found that the detector with contacts formed by electroless deposition produced the best uniformity and energy resolution; the best pixel produced a FWHM of 560 eV at 59.54 keV and 50% of pixels produced a FWHM better than 1.7 keV . This compared with a FWHM of 1.5 keV for the best pixel and 50% of pixels better than 4.4 keV for the detector with sputtered contacts.

Spatial and Time Domain Feature of ERP Speller System Extracted via Convolutional Neural Network.

PubMed

Yoon, Jaehong; Lee, Jungnyun; Whang, Mincheol

2018-01-01

Feature of event-related potential (ERP) has not been completely understood and illiteracy problem remains unsolved. To this end, P300 peak has been used as the feature of ERP in most brain-computer interface applications, but subjects who do not show such peak are common. Recent development of convolutional neural network provides a way to analyze spatial and temporal features of ERP. Here, we train the convolutional neural network with 2 convolutional layers whose feature maps represented spatial and temporal features of event-related potential. We have found that nonilliterate subjects' ERP show high correlation between occipital lobe and parietal lobe, whereas illiterate subjects only show correlation between neural activities from frontal lobe and central lobe. The nonilliterates showed peaks in P300, P500, and P700, whereas illiterates mostly showed peaks in around P700. P700 was strong in both subjects. We found that P700 peak may be the key feature of ERP as it appears in both illiterate and nonilliterate subjects.

Spatial and Time Domain Feature of ERP Speller System Extracted via Convolutional Neural Network

PubMed Central

2018-01-01

Feature of event-related potential (ERP) has not been completely understood and illiteracy problem remains unsolved. To this end, P300 peak has been used as the feature of ERP in most brain–computer interface applications, but subjects who do not show such peak are common. Recent development of convolutional neural network provides a way to analyze spatial and temporal features of ERP. Here, we train the convolutional neural network with 2 convolutional layers whose feature maps represented spatial and temporal features of event-related potential. We have found that nonilliterate subjects' ERP show high correlation between occipital lobe and parietal lobe, whereas illiterate subjects only show correlation between neural activities from frontal lobe and central lobe. The nonilliterates showed peaks in P300, P500, and P700, whereas illiterates mostly showed peaks in around P700. P700 was strong in both subjects. We found that P700 peak may be the key feature of ERP as it appears in both illiterate and nonilliterate subjects.

Efficient phase contrast imaging in STEM using a pixelated detector. Part 1: Experimental demonstration at atomic resolution

DOE PAGES

Pennycook, Timothy J.; Lupini, Andrew R.; Yang, Hao; ...

2014-10-15

In this paper, we demonstrate a method to achieve high efficiency phase contrast imaging in aberration corrected scanning transmission electron microscopy (STEM) with a pixelated detector. The pixelated detector is used to record the Ronchigram as a function of probe position which is then analyzed with ptychography. Ptychography has previously been used to provide super-resolution beyond the diffraction limit of the optics, alongside numerically correcting for spherical aberration. Here we rely on a hardware aberration corrector to eliminate aberrations, but use the pixelated detector data set to utilize the largest possible volume of Fourier space to create high efficiency phasemore » contrast images. The use of ptychography to diagnose the effects of chromatic aberration is also demonstrated. In conclusion, the four dimensional dataset is used to compare different bright field detector configurations from the same scan for a sample of bilayer graphene. Our method of high efficiency ptychography produces the clearest images, while annular bright field produces almost no contrast for an in-focus aberration-corrected probe.« less

Characterization of pixel sensor designed in 180 nm SOI CMOS technology

NASA Astrophysics Data System (ADS)

Benka, T.; Havranek, M.; Hejtmanek, M.; Jakovenko, J.; Janoska, Z.; Marcisovska, M.; Marcisovsky, M.; Neue, G.; Tomasek, L.; Vrba, V.

2018-01-01

A new type of X-ray imaging Monolithic Active Pixel Sensor (MAPS), X-CHIP-02, was developed using a 180 nm deep submicron Silicon On Insulator (SOI) CMOS commercial technology. Two pixel matrices were integrated into the prototype chip, which differ by the pixel pitch of 50 μm and 100 μm. The X-CHIP-02 contains several test structures, which are useful for characterization of individual blocks. The sensitive part of the pixel integrated in the handle wafer is one of the key structures designed for testing. The purpose of this structure is to determine the capacitance of the sensitive part (diode in the MAPS pixel). The measured capacitance is 2.9 fF for 50 μm pixel pitch and 4.8 fF for 100 μm pixel pitch at -100 V (default operational voltage). This structure was used to measure the IV characteristics of the sensitive diode. In this work, we report on a circuit designed for precise determination of sensor capacitance and IV characteristics of both pixel types with respect to X-ray irradiation. The motivation for measurement of the sensor capacitance was its importance for the design of front-end amplifier circuits. The design of pixel elements, as well as circuit simulation and laboratory measurement techniques are described. The experimental results are of great importance for further development of MAPS sensors in this technology.

Fully 3D-Integrated Pixel Detectors for X-Rays

DOE PAGES

Deptuch, Grzegorz W.; Gabriella, Carini; Enquist, Paul; ...

2016-01-01

The vertically integrated photon imaging chip (VIPIC1) pixel detector is a stack consisting of a 500-μm-thick silicon sensor, a two-tier 34-μm-thick integrated circuit, and a host printed circuit board (PCB). The integrated circuit tiers were bonded using the direct bonding technology with copper, and each tier features 1-μm-diameter through-silicon vias that were used for connections to the sensor on one side, and to the host PCB on the other side. The 80-μm-pixel-pitch sensor was the direct bonding technology with nickel bonded to the integrated circuit. The stack was mounted on the board using Sn–Pb balls placed on a 320-μm pitch,more » yielding an entirely wire-bond-less structure. The analog front-end features a pulse response peaking at below 250 ns, and the power consumption per pixel is 25 μW. We successful completed the 3-D integration and have reported here. Additionally, all pixels in the matrix of 64 × 64 pixels were responding on well-bonded devices. Correct operation of the sparsified readout, allowing a single 153-ns bunch timing resolution, was confirmed in the tests on a synchrotron beam of 10-keV X-rays. An equivalent noise charge of 36.2 e - rms and a conversion gain of 69.5 μV/e - with 2.6 e - rms and 2.7 μV/e - rms pixel-to-pixel variations, respectively, were measured.« less

CMOS Active Pixel Sensor Technology and Reliability Characterization Methodology

NASA Technical Reports Server (NTRS)

Chen, Yuan; Guertin, Steven M.; Pain, Bedabrata; Kayaii, Sammy

2006-01-01

This paper describes the technology, design features and reliability characterization methodology of a CMOS Active Pixel Sensor. Both overall chip reliability and pixel reliability are projected for the imagers.

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

Spatial aggregation of holistically-nested convolutional neural networks for automated pancreas localization and segmentation.

PubMed

Roth, Holger R; Lu, Le; Lay, Nathan; Harrison, Adam P; Farag, Amal; Sohn, Andrew; Summers, Ronald M

2018-04-01

Accurate and automatic organ segmentation from 3D radiological scans is an important yet challenging problem for medical image analysis. Specifically, as a small, soft, and flexible abdominal organ, the pancreas demonstrates very high inter-patient anatomical variability in both its shape and volume. This inhibits traditional automated segmentation methods from achieving high accuracies, especially compared to the performance obtained for other organs, such as the liver, heart or kidneys. To fill this gap, we present an automated system from 3D computed tomography (CT) volumes that is based on a two-stage cascaded approach-pancreas localization and pancreas segmentation. For the first step, we localize the pancreas from the entire 3D CT scan, providing a reliable bounding box for the more refined segmentation step. We introduce a fully deep-learning approach, based on an efficient application of holistically-nested convolutional networks (HNNs) on the three orthogonal axial, sagittal, and coronal views. The resulting HNN per-pixel probability maps are then fused using pooling to reliably produce a 3D bounding box of the pancreas that maximizes the recall. We show that our introduced localizer compares favorably to both a conventional non-deep-learning method and a recent hybrid approach based on spatial aggregation of superpixels using random forest classification. The second, segmentation, phase operates within the computed bounding box and integrates semantic mid-level cues of deeply-learned organ interior and boundary maps, obtained by two additional and separate realizations of HNNs. By integrating these two mid-level cues, our method is capable of generating boundary-preserving pixel-wise class label maps that result in the final pancreas segmentation. Quantitative evaluation is performed on a publicly available dataset of 82 patient CT scans using 4-fold cross-validation (CV). We achieve a (mean  ±  std. dev.) Dice similarity coefficient (DSC) of 81.27�

Performance assessment of a compressive sensing single-pixel imaging system

NASA Astrophysics Data System (ADS)

Du Bosq, Todd W.; Preece, Bradley L.

2017-04-01

Conventional sensors measure the light incident at each pixel in a focal plane array. Compressive sensing (CS) involves capturing a smaller number of unconventional measurements from the scene, and then using a companion process to recover the image. CS has the potential to acquire imagery with equivalent information content to a large format array while using smaller, cheaper, and lower bandwidth components. However, the benefits of CS do not come without compromise. The CS architecture chosen must effectively balance between physical considerations, reconstruction accuracy, and reconstruction speed to meet operational requirements. Performance modeling of CS imagers is challenging due to the complexity and nonlinearity of the system and reconstruction algorithm. To properly assess the value of such systems, it is necessary to fully characterize the image quality, including artifacts and sensitivity to noise. Imagery of a two-handheld object target set was collected using an shortwave infrared single-pixel CS camera for various ranges and number of processed measurements. Human perception experiments were performed to determine the identification performance within the trade space. The performance of the nonlinear CS camera was modeled by mapping the nonlinear degradations to an equivalent linear shift invariant model. Finally, the limitations of CS modeling techniques are discussed.

A Mathematical Motivation for Complex-Valued Convolutional Networks.

PubMed

Tygert, Mark; Bruna, Joan; Chintala, Soumith; LeCun, Yann; Piantino, Serkan; Szlam, Arthur

2016-05-01

A complex-valued convolutional network (convnet) implements the repeated application of the following composition of three operations, recursively applying the composition to an input vector of nonnegative real numbers: (1) convolution with complex-valued vectors, followed by (2) taking the absolute value of every entry of the resulting vectors, followed by (3) local averaging. For processing real-valued random vectors, complex-valued convnets can be viewed as data-driven multiscale windowed power spectra, data-driven multiscale windowed absolute spectra, data-driven multiwavelet absolute values, or (in their most general configuration) data-driven nonlinear multiwavelet packets. Indeed, complex-valued convnets can calculate multiscale windowed spectra when the convnet filters are windowed complex-valued exponentials. Standard real-valued convnets, using rectified linear units (ReLUs), sigmoidal (e.g., logistic or tanh) nonlinearities, or max pooling, for example, do not obviously exhibit the same exact correspondence with data-driven wavelets (whereas for complex-valued convnets, the correspondence is much more than just a vague analogy). Courtesy of the exact correspondence, the remarkably rich and rigorous body of mathematical analysis for wavelets applies directly to (complex-valued) convnets.

Multichannel Convolutional Neural Network for Biological Relation Extraction.

PubMed

Quan, Chanqin; Hua, Lei; Sun, Xiao; Bai, Wenjun

2016-01-01

The plethora of biomedical relations which are embedded in medical logs (records) demands researchers' attention. Previous theoretical and practical focuses were restricted on traditional machine learning techniques. However, these methods are susceptible to the issues of "vocabulary gap" and data sparseness and the unattainable automation process in feature extraction. To address aforementioned issues, in this work, we propose a multichannel convolutional neural network (MCCNN) for automated biomedical relation extraction. The proposed model has the following two contributions: (1) it enables the fusion of multiple (e.g., five) versions in word embeddings; (2) the need for manual feature engineering can be obviated by automated feature learning with convolutional neural network (CNN). We evaluated our model on two biomedical relation extraction tasks: drug-drug interaction (DDI) extraction and protein-protein interaction (PPI) extraction. For DDI task, our system achieved an overall f -score of 70.2% compared to the standard linear SVM based system (e.g., 67.0%) on DDIExtraction 2013 challenge dataset. And for PPI task, we evaluated our system on Aimed and BioInfer PPI corpus; our system exceeded the state-of-art ensemble SVM system by 2.7% and 5.6% on f -scores.

Spectral characteristics of convolutionally coded digital signals

NASA Technical Reports Server (NTRS)

Divsalar, D.

1979-01-01

The power spectral density of the output symbol sequence of a convolutional encoder is computed for two different input symbol stream source models, namely, an NRZ signaling format and a first order Markov source. In the former, the two signaling states of the binary waveform are not necessarily assumed to occur with equal probability. The effects of alternate symbol inversion on this spectrum are also considered. The mathematical results are illustrated with many examples corresponding to optimal performance codes.

Sub-pixel localization of highways in AVIRIS images

NASA Technical Reports Server (NTRS)

Salu, Yehuda

1995-01-01

Roads and highways show up clearly in many bands of AVIRIS images. A typical lane in the U.S. is 12 feet wide, and the total width of a four lane highway, including 18 feet of paved shoulders, is 19.8 m. Such a highway will cover only a portion of any 20x20 m AVIRIS pixel that it traverses. The other portion of these pixels wil be usually covered by vegetation. An interesting problem is to precisely determine the location of a highway within the AVIRIS pixels that it traverses. This information may be used for alignment and spatial calibration of AVIRIS images. Also, since the reflection properties of highway surfaces do not change with time, and they can be determined once and for all, such information can be of help in calculating and filtering out the atmospheric noise that contaminates AVIRIS measurements. The purpose of this report is to describe a method for sub-pixel localization of highways.

Polarized-pixel performance model for DoFP polarimeter

NASA Astrophysics Data System (ADS)

Feng, Bin; Shi, Zelin; Liu, Haizheng; Liu, Li; Zhao, Yaohong; Zhang, Junchao

2018-06-01

A division of a focal plane (DoFP) polarimeter is manufactured by placing a micropolarizer array directly onto the focal plane array (FPA) of a detector. Each element of the DoFP polarimeter is a polarized pixel. This paper proposes a performance model for a polarized pixel. The proposed model characterizes the optical and electronic performance of a polarized pixel by three parameters. They are respectively major polarization responsivity, minor polarization responsivity and polarization orientation. Each parameter corresponds to an intuitive physical feature of a polarized pixel. This paper further extends this model to calibrate polarization images from a DoFP (division of focal plane) polarimeter. This calibration work is evaluated quantitatively by a developed DoFP polarimeter under varying illumination intensity and angle of linear polarization. The experiment proves that our model reduces nonuniformity to 6.79% of uncalibrated DoLP (degree of linear polarization) images, and significantly improves the visual effect of DoLP images.

The effect of split pixel HDR image sensor technology on MTF measurements

NASA Astrophysics Data System (ADS)

Deegan, Brian M.

2014-03-01

Split-pixel HDR sensor technology is particularly advantageous in automotive applications, because the images are captured simultaneously rather than sequentially, thereby reducing motion blur. However, split pixel technology introduces artifacts in MTF measurement. To achieve a HDR image, raw images are captured from both large and small sub-pixels, and combined to make the HDR output. In some cases, a large sub-pixel is used for long exposure captures, and a small sub-pixel for short exposures, to extend the dynamic range. The relative size of the photosensitive area of the pixel (fill factor) plays a very significant role in the output MTF measurement. Given an identical scene, the MTF will be significantly different, depending on whether you use the large or small sub-pixels i.e. a smaller fill factor (e.g. in the short exposure sub-pixel) will result in higher MTF scores, but significantly greater aliasing. Simulations of split-pixel sensors revealed that, when raw images from both sub-pixels are combined, there is a significant difference in rising edge (i.e. black-to-white transition) and falling edge (white-to-black) reproduction. Experimental results showed a difference of ~50% in measured MTF50 between the falling and rising edges of a slanted edge test chart.

A brain MRI bias field correction method created in the Gaussian multi-scale space

NASA Astrophysics Data System (ADS)

Chen, Mingsheng; Qin, Mingxin

2017-07-01

A pre-processing step is needed to correct for the bias field signal before submitting corrupted MR images to such image-processing algorithms. This study presents a new bias field correction method. The method creates a Gaussian multi-scale space by the convolution of the inhomogeneous MR image with a two-dimensional Gaussian function. In the multi-Gaussian space, the method retrieves the image details from the differentiation of the original image and convolution image. Then, it obtains an image whose inhomogeneity is eliminated by the weighted sum of image details in each layer in the space. Next, the bias field-corrected MR image is retrieved after the Υ correction, which enhances the contrast and brightness of the inhomogeneity-eliminated MR image. We have tested the approach on T1 MRI and T2 MRI with varying bias field levels and have achieved satisfactory results. Comparison experiments with popular software have demonstrated superior performance of the proposed method in terms of quantitative indices, especially an improvement in subsequent image segmentation.

A CMOS image sensor with programmable pixel-level analog processing.

PubMed

Massari, Nicola; Gottardi, Massimo; Gonzo, Lorenzo; Stoppa, David; Simoni, Andrea

2005-11-01

A prototype of a 34 x 34 pixel image sensor, implementing real-time analog image processing, is presented. Edge detection, motion detection, image amplification, and dynamic-range boosting are executed at pixel level by means of a highly interconnected pixel architecture based on the absolute value of the difference among neighbor pixels. The analog operations are performed over a kernel of 3 x 3 pixels. The square pixel, consisting of 30 transistors, has a pitch of 35 microm with a fill-factor of 20%. The chip was fabricated in a 0.35 microm CMOS technology, and its power consumption is 6 mW with 3.3 V power supply. The device was fully characterized and achieves a dynamic range of 50 dB with a light power density of 150 nW/mm2 and a frame rate of 30 frame/s. The measured fixed pattern noise corresponds to 1.1% of the saturation level. The sensor's dynamic range can be extended up to 96 dB using the double-sampling technique.

Deep convolutional neural networks for automatic classification of gastric carcinoma using whole slide images in digital histopathology.

PubMed

Sharma, Harshita; Zerbe, Norman; Klempert, Iris; Hellwich, Olaf; Hufnagl, Peter

2017-11-01

Deep learning using convolutional neural networks is an actively emerging field in histological image analysis. This study explores deep learning methods for computer-aided classification in H&E stained histopathological whole slide images of gastric carcinoma. An introductory convolutional neural network architecture is proposed for two computerized applications, namely, cancer classification based on immunohistochemical response and necrosis detection based on the existence of tumor necrosis in the tissue. Classification performance of the developed deep learning approach is quantitatively compared with traditional image analysis methods in digital histopathology requiring prior computation of handcrafted features, such as statistical measures using gray level co-occurrence matrix, Gabor filter-bank responses, LBP histograms, gray histograms, HSV histograms and RGB histograms, followed by random forest machine learning. Additionally, the widely known AlexNet deep convolutional framework is comparatively analyzed for the corresponding classification problems. The proposed convolutional neural network architecture reports favorable results, with an overall classification accuracy of 0.6990 for cancer classification and 0.8144 for necrosis detection. Copyright © 2017 Elsevier Ltd. All rights reserved.

Modulation transfer function of a triangular pixel array detector.

PubMed

Karimzadeh, Ayatollah

2014-07-01

The modulation transfer function (MTF) is the main parameter that is used to evaluate image quality in electro-optical systems. Detector sampling MTF in most electro-optical systems determines the cutoff frequency of the system. The MTF of the detector depends on its pixel shape. In this work, we calculated the MTF of a detector with an equilateral triangular pixel shape. Some new results were found in deriving the MTF for the equilateral triangular pixel shape.

ID card number detection algorithm based on convolutional neural network

NASA Astrophysics Data System (ADS)

Zhu, Jian; Ma, Hanjie; Feng, Jie; Dai, Leiyan

2018-04-01

In this paper, a new detection algorithm based on Convolutional Neural Network is presented in order to realize the fast and convenient ID information extraction in multiple scenarios. The algorithm uses the mobile device equipped with Android operating system to locate and extract the ID number; Use the special color distribution of the ID card, select the appropriate channel component; Use the image threshold segmentation, noise processing and morphological processing to take the binary processing for image; At the same time, the image rotation and projection method are used for horizontal correction when image was tilting; Finally, the single character is extracted by the projection method, and recognized by using Convolutional Neural Network. Through test shows that, A single ID number image from the extraction to the identification time is about 80ms, the accuracy rate is about 99%, It can be applied to the actual production and living environment.

An enhanced multi-view vertical line locus matching algorithm of object space ground primitives based on positioning consistency for aerial and space images

NASA Astrophysics Data System (ADS)

Zhang, Ka; Sheng, Yehua; Wang, Meizhen; Fu, Suxia

2018-05-01

The traditional multi-view vertical line locus (TMVLL) matching method is an object-space-based method that is commonly used to directly acquire spatial 3D coordinates of ground objects in photogrammetry. However, the TMVLL method can only obtain one elevation and lacks an accurate means of validating the matching results. In this paper, we propose an enhanced multi-view vertical line locus (EMVLL) matching algorithm based on positioning consistency for aerial or space images. The algorithm involves three components: confirming candidate pixels of the ground primitive in the base image, multi-view image matching based on the object space constraints for all candidate pixels, and validating the consistency of the object space coordinates with the multi-view matching result. The proposed algorithm was tested using actual aerial images and space images. Experimental results show that the EMVLL method successfully solves the problems associated with the TMVLL method, and has greater reliability, accuracy and computing efficiency.

Esophagus segmentation in CT via 3D fully convolutional neural network and random walk.

PubMed

Fechter, Tobias; Adebahr, Sonja; Baltas, Dimos; Ben Ayed, Ismail; Desrosiers, Christian; Dolz, Jose

2017-12-01

the results of this model can be refined by a random walk step taking pixel intensities and neighborhood relationships into account. One of the main advantages of our network over previous methods is that it performs 3D convolutions, thus fully exploiting the 3D spatial context and performing an efficient volume-wise prediction. The whole segmentation process is fully automatic and yields esophagus delineations in very good agreement with the gold standard, showing that it can compete with previously published methods. © 2017 American Association of Physicists in Medicine.

Pixel super resolution using wavelength scanning

DTIC Science & Technology

2016-04-08

the light source is adjusted to ~20 μW. The image sensor chip is a color CMOS sensor chip with a pixel size of 1.12 μm manufactured for cellphone...pitch (that is, ~ 1 μm in Figure 3a, using a CMOS sensor that has a 1.12-μm pixel pitch). For the same configuration depicted in Figure 3, utilizing...section). The a Lens-free raw holograms captured by 1.12 μm CMOS image sensor Field of view ≈ 20.5 mm2 Angle change directions for synthetic aperture

Detecting of foreign object debris on airfield pavement using convolution neural network

NASA Astrophysics Data System (ADS)

Cao, Xiaoguang; Gu, Yufeng; Bai, Xiangzhi

2017-11-01

It is of great practical significance to detect foreign object debris (FOD) timely and accurately on the airfield pavement, because the FOD is a fatal threaten for runway safety in airport. In this paper, a new FOD detection framework based on Single Shot MultiBox Detector (SSD) is proposed. Two strategies include making the detection network lighter and using dilated convolution, which are proposed to better solve the FOD detection problem. The advantages mainly include: (i) the network structure becomes lighter to speed up detection task and enhance detection accuracy; (ii) dilated convolution is applied in network structure to handle smaller FOD. Thus, we get a faster and more accurate detection system.

Video-based convolutional neural networks for activity recognition from robot-centric videos

NASA Astrophysics Data System (ADS)

Ryoo, M. S.; Matthies, Larry

2016-05-01

In this evaluation paper, we discuss convolutional neural network (CNN)-based approaches for human activity recognition. In particular, we investigate CNN architectures designed to capture temporal information in videos and their applications to the human activity recognition problem. There have been multiple previous works to use CNN-features for videos. These include CNNs using 3-D XYT convolutional filters, CNNs using pooling operations on top of per-frame image-based CNN descriptors, and recurrent neural networks to learn temporal changes in per-frame CNN descriptors. We experimentally compare some of these different representatives CNNs while using first-person human activity videos. We especially focus on videos from a robots viewpoint, captured during its operations and human-robot interactions.

Convolution-based estimation of organ dose in tube current modulated CT

NASA Astrophysics Data System (ADS)

Tian, Xiaoyu; Segars, W. Paul; Dixon, Robert L.; Samei, Ehsan

2016-05-01

Estimating organ dose for clinical patients requires accurate modeling of the patient anatomy and the dose field of the CT exam. The modeling of patient anatomy can be achieved using a library of representative computational phantoms (Samei et al 2014 Pediatr. Radiol. 44 460-7). The modeling of the dose field can be challenging for CT exams performed with a tube current modulation (TCM) technique. The purpose of this work was to effectively model the dose field for TCM exams using a convolution-based method. A framework was further proposed for prospective and retrospective organ dose estimation in clinical practice. The study included 60 adult patients (age range: 18-70 years, weight range: 60-180 kg). Patient-specific computational phantoms were generated based on patient CT image datasets. A previously validated Monte Carlo simulation program was used to model a clinical CT scanner (SOMATOM Definition Flash, Siemens Healthcare, Forchheim, Germany). A practical strategy was developed to achieve real-time organ dose estimation for a given clinical patient. CTDIvol-normalized organ dose coefficients ({{h}\\text{Organ}} ) under constant tube current were estimated and modeled as a function of patient size. Each clinical patient in the library was optimally matched to another computational phantom to obtain a representation of organ location/distribution. The patient organ distribution was convolved with a dose distribution profile to generate {{≤ft(\\text{CTD}{{\\text{I}}\\text{vol}}\\right)}\\text{organ, \\text{convolution}}} values that quantified the regional dose field for each organ. The organ dose was estimated by multiplying {{≤ft(\\text{CTD}{{\\text{I}}\\text{vol}}\\right)}\\text{organ, \\text{convolution}}} with the organ dose coefficients ({{h}\\text{Organ}} ). To validate the accuracy of this dose estimation technique, the organ dose of the original clinical patient was estimated using Monte Carlo program with TCM profiles explicitly modeled. The

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

Techniques for precise energy calibration of particle pixel detectors

NASA Astrophysics Data System (ADS)

Kroupa, M.; Campbell-Ricketts, T.; Bahadori, A.; Empl, A.

2017-03-01

We demonstrate techniques to improve the accuracy of the energy calibration of Timepix pixel detectors, used for the measurement of energetic particles. The typical signal from such particles spreads among many pixels due to charge sharing effects. As a consequence, the deposited energy in each pixel cannot be reconstructed unless the detector is calibrated, limiting the usability of such signals for calibration. To avoid this shortcoming, we calibrate using low energy X-rays. However, charge sharing effects still occur, resulting in part of the energy being deposited in adjacent pixels and possibly lost. This systematic error in the calibration process results in an error of about 5% in the energy measurements of calibrated devices. We use FLUKA simulations to assess the magnitude of charge sharing effects, allowing a corrected energy calibration to be performed on several Timepix pixel detectors and resulting in substantial improvement in energy deposition measurements. Next, we address shortcomings in calibration associated with the huge range (from kiloelectron-volts to megaelectron-volts) of energy deposited per pixel which result in a nonlinear energy response over the full range. We introduce a new method to characterize the non-linear response of the Timepix detectors at high input energies. We demonstrate improvement using a broad range of particle types and energies, showing that the new method reduces the energy measurement errors, in some cases by more than 90%.

Techniques for precise energy calibration of particle pixel detectors.

PubMed

Kroupa, M; Campbell-Ricketts, T; Bahadori, A; Empl, A

2017-03-01

We demonstrate techniques to improve the accuracy of the energy calibration of Timepix pixel detectors, used for the measurement of energetic particles. The typical signal from such particles spreads among many pixels due to charge sharing effects. As a consequence, the deposited energy in each pixel cannot be reconstructed unless the detector is calibrated, limiting the usability of such signals for calibration. To avoid this shortcoming, we calibrate using low energy X-rays. However, charge sharing effects still occur, resulting in part of the energy being deposited in adjacent pixels and possibly lost. This systematic error in the calibration process results in an error of about 5% in the energy measurements of calibrated devices. We use FLUKA simulations to assess the magnitude of charge sharing effects, allowing a corrected energy calibration to be performed on several Timepix pixel detectors and resulting in substantial improvement in energy deposition measurements. Next, we address shortcomings in calibration associated with the huge range (from kiloelectron-volts to megaelectron-volts) of energy deposited per pixel which result in a nonlinear energy response over the full range. We introduce a new method to characterize the non-linear response of the Timepix detectors at high input energies. We demonstrate improvement using a broad range of particle types and energies, showing that the new method reduces the energy measurement errors, in some cases by more than 90%.

Electroencephalography Based Fusion Two-Dimensional (2D)-Convolution Neural Networks (CNN) Model for Emotion Recognition System.

PubMed

Kwon, Yea-Hoon; Shin, Sae-Byuk; Kim, Shin-Dug

2018-04-30

The purpose of this study is to improve human emotional classification accuracy using a convolution neural networks (CNN) model and to suggest an overall method to classify emotion based on multimodal data. We improved classification performance by combining electroencephalogram (EEG) and galvanic skin response (GSR) signals. GSR signals are preprocessed using by the zero-crossing rate. Sufficient EEG feature extraction can be obtained through CNN. Therefore, we propose a suitable CNN model for feature extraction by tuning hyper parameters in convolution filters. The EEG signal is preprocessed prior to convolution by a wavelet transform while considering time and frequency simultaneously. We use a database for emotion analysis using the physiological signals open dataset to verify the proposed process, achieving 73.4% accuracy, showing significant performance improvement over the current best practice models.

Transfer Function Bounds for Partial-unit-memory Convolutional Codes Based on Reduced State Diagram

NASA Technical Reports Server (NTRS)

Lee, P. J.

1984-01-01

The performance of a coding system consisting of a convolutional encoder and a Viterbi decoder is analytically found by the well-known transfer function bounding technique. For the partial-unit-memory byte-oriented convolutional encoder with m sub 0 binary memory cells and (k sub 0 m sub 0) inputs, a state diagram of 2(K) (sub 0) was for the transfer function bound. A reduced state diagram of (2 (m sub 0) +1) is used for easy evaluation of transfer function bounds for partial-unit-memory codes.

CMOS Active Pixel Sensor Star Tracker with Regional Electronic Shutter

NASA Technical Reports Server (NTRS)

Yadid-Pecht, Orly; Pain, Bedabrata; Staller, Craig; Clark, Christopher; Fossum, Eric

1996-01-01

The guidance system in a spacecraft determines spacecraft attitude by matching an observed star field to a star catalog....An APS(active pixel sensor)-based system can reduce mass and power consumption and radiation effects compared to a CCD(charge-coupled device)-based system...This paper reports an APS (active pixel sensor) with locally variable times, achieved through individual pixel reset (IPR).

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.

1.5 Meter Per Pixel View of Boulders in Ganges Chasma

NASA Technical Reports Server (NTRS)

1999-01-01

The Mars Orbiter Camera (MOC) on board the Mars Global Surveyor (MGS)spacecraft was designed to be able to take pictures that 'bridge the gap' between what could be seen by the Mariner 9 and Viking Orbiters from space and what could be seen by landers from the ground. In other words, MOC was designed to be able to see boulders of sizes similar to and larger than those named 'Yogi' at the Mars Pathfinder site and 'Big Joe' at the Viking 1 landing site. To see such boulders, a resolution of at least 1.5 meters (5 feet) per pixel was required.

With the start of the MGS Mapping Phase of the mission during the second week of March 1999, the MOC team is pleased to report that 'the gap is bridged.' This image shows a field of boulders on the surface of a landslide deposit in Ganges Chasma. Ganges Chasma is one of the valleys in the Valles Marineris canyon system. The image resolution is 1.5 meters per pixel. The boulders shown here range in size from about 2 meters (7 feet) to about 20 meters (66 feet) in size. The image covers an area 1 kilometer (0.62 miles) across, and illumination is from the upper left.

Malin Space Science Systems and the California Institute of Technology built the MOC using spare hardware from the Mars Observer mission. MSSS operates the camera from its facilities in San Diego, CA. The Jet Propulsion Laboratory's Mars Surveyor Operations Project operates the Mars Global Surveyor spacecraft with its industrial partner, Lockheed Martin Astronautics, from facilities in Pasadena, CA and Denver, CO.

Three-terminal RGB full-color OLED pixels for ultrahigh density displays.

PubMed

Fröbel, Markus; Fries, Felix; Schwab, Tobias; Lenk, Simone; Leo, Karl; Gather, Malte C; Reineke, Sebastian

2018-06-26

In recent years, the organic light-emitting diode (OLED) technology has been a rapidly evolving field of research, successfully making the transition to commercial applications such as mobile phones and other small portable devices. OLEDs provide efficient generation of light, excellent color quality, and allow for innovative display designs, e.g., curved shapes, mechanically flexible and/or transparent devices. Especially their self emissive nature is a highly desirable feature for display applications. In this work, we demonstrate an approach for full-color OLED pixels that are fabricated by vertical stacking of a red-, green-, and blue-emitting unit. Each unit can be addressed separately which allows for efficient generation of every color that is accessible by superpositioning the spectra of the individual emission units. Here, we use a combination of time division multiplexing and pulse width modulation to achieve efficient color mixing. The presented device design requires only three independently addressable electrodes, simplifying both fabrication and electrical driving. The device is built in a top-emission geometry, which is highly desirable for display fabrication as the pixel can be directly deposited onto back-plane electronics. Despite the top-emission design and the application of three silver layers within the device, there is only a minor color shift even for large viewing angles. The color space spanned by the three emission sub-units exceeds the sRGB space, providing more saturated green/yellow/red colors. Furthermore, the electrical performance of each individual unit is on par with standard single emission unit OLEDs, showing very low leakage currents and achieving brightness levels above 1000 cd/m 2 at moderate voltages of around 3-4 V.

A new LTPS TFT AC pixel circuit for an AMOLED

NASA Astrophysics Data System (ADS)

Yongwen, Zhang; Wenbin, Chen

2013-01-01

This work presents a new voltage programmed pixel circuit for an active-matrix organic light-emitting diode (AMOLED) display. The proposed pixel circuit consists of six low temperature polycrystalline silicon thin-film transistors (LTPS TFTs), one storage capacitor, and one OLED, and is verified by simulation work using HSPICE software. Besides effectively compensating for the threshold voltage variation of the driving TFT and OLED, the proposed pixel circuit offers an AC driving mode for the OLED, which can suppress the degradation of the OLED. Moreover, a high contrast ratio can be achieved by the proposed pixel circuit since the OLED does not emit any light except for the emission period.

Analog pixel array detectors.

PubMed

Ercan, A; Tate, M W; Gruner, S M

2006-03-01

X-ray pixel array detectors (PADs) are generally thought of as either digital photon counters (DPADs) or X-ray analog-integrating pixel array detectors (APADs). Experiences with APADs, which are especially well suited for X-ray imaging experiments where transient or high instantaneous flux events must be recorded, are reported. The design, characterization and experimental applications of several APAD designs developed at Cornell University are discussed. The simplest design is a ;flash' architecture, wherein successive integrated X-ray images, as short as several hundred nanoseconds in duration, are stored in the detector chips for later off-chip digitization. Radiography experiments using a prototype flash APAD are summarized. Another design has been implemented that combines flash capability with the ability to continuously stream X-ray images at slower (e.g. milliseconds) rates. Progress is described towards radiation-hardened APADs that can be tiled to cover a large area. A mixed-mode PAD, design by combining many of the attractive features of both APADs and DPADs, is also described.

Using a pulsed laser beam to investigate the feasibility of sub-pixel position resolution with time-correlated transient signals in 3D pixelated CdZnTe detectors

DOE PAGES

Giraldo, L. Ocampo; Bolotnikov, A. E.; Camarda, G. S.; ...

2017-04-20

For this study, we evaluated the X-Y position resolution achievable in 3D pixelated detectors by processing the signal waveforms readout from neighboring pixels. In these measurements we used a focused light beam, down to 10 μm, generated by a ~1 mW pulsed laser (650 nm) to carry out raster scans over selected 3×3 pixel areas, while recording the charge signals from the 9 pixels and the cathode using two synchronized digital oscilloscopes.

Fully depleted CMOS pixel sensor development and potential applications

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

Baudot, J.; Kachel, M.; CNRS, UMR7178, 67037 Strasbourg

CMOS pixel sensors are often opposed to hybrid pixel sensors due to their very different sensitive layer. In standard CMOS imaging processes, a thin (about 20 μm) low resistivity epitaxial layer acts as the sensitive volume and charge collection is mostly driven by thermal agitation. In contrast, the so-called hybrid pixel technology exploits a thick (typically 300 μm) silicon sensor with high resistivity allowing for the depletion of this volume, hence charges drift toward collecting electrodes. But this difference is fading away with the recent availability of some CMOS imaging processes based on a relatively thick (about 50 μm) highmore » resistivity epitaxial layer which allows for full depletion. This evolution extents the range of applications for CMOS pixel sensors where their known assets, high sensitivity and granularity combined with embedded signal treatment, could potentially foster breakthrough in detection performances for specific scientific instruments. One such domain is the Xray detection for soft energies, typically below 10 keV, where the thin sensitive layer was previously severely impeding CMOS sensor usage. Another application becoming realistic for CMOS sensors, is the detection in environment with a high fluence of non-ionizing radiation, such as hadron colliders. However, when considering highly demanding applications, it is still to be proven that micro-circuits required to uniformly deplete the sensor at the pixel level, do not mitigate the sensitivity and efficiency required. Prototype sensors in two different technologies with resistivity higher than 1 kΩ, sensitive layer between 40 and 50 μm and featuring pixel pitch in the range 25 to 50 μm, have been designed and fabricated. Various biasing architectures were adopted to reach full depletion with only a few volts. Laboratory investigations with three types of sources (X-rays, β-rays and infrared light) demonstrated the validity of the approach with respect to depletion

Generalized type II hybrid ARQ scheme using punctured convolutional coding

NASA Astrophysics Data System (ADS)

Kallel, Samir; Haccoun, David

1990-11-01

A method is presented to construct rate-compatible convolutional (RCC) codes from known high-rate punctured convolutional codes, obtained from best-rate 1/2 codes. The construction method is rather simple and straightforward, and still yields good codes. Moreover, low-rate codes can be obtained without any limit on the lowest achievable code rate. Based on the RCC codes, a generalized type-II hybrid ARQ scheme, which combines the benefits of the modified type-II hybrid ARQ strategy of Hagenauer (1988) with the code-combining ARQ strategy of Chase (1985), is proposed and analyzed. With the proposed generalized type-II hybrid ARQ strategy, the throughput increases as the starting coding rate increases, and as the channel degrades, it tends to merge with the throughput of rate 1/2 type-II hybrid ARQ schemes with code combining, thus allowing the system to be flexible and adaptive to channel conditions, even under wide noise variations and severe degradations.

Spatiotemporal Recurrent Convolutional Networks for Traffic Prediction in Transportation Networks

PubMed Central

Yu, Haiyang; Wu, Zhihai; Wang, Shuqin; Wang, Yunpeng; Ma, Xiaolei

2017-01-01

Predicting large-scale transportation network traffic has become an important and challenging topic in recent decades. Inspired by the domain knowledge of motion prediction, in which the future motion of an object can be predicted based on previous scenes, we propose a network grid representation method that can retain the fine-scale structure of a transportation network. Network-wide traffic speeds are converted into a series of static images and input into a novel deep architecture, namely, spatiotemporal recurrent convolutional networks (SRCNs), for traffic forecasting. The proposed SRCNs inherit the advantages of deep convolutional neural networks (DCNNs) and long short-term memory (LSTM) neural networks. The spatial dependencies of network-wide traffic can be captured by DCNNs, and the temporal dynamics can be learned by LSTMs. An experiment on a Beijing transportation network with 278 links demonstrates that SRCNs outperform other deep learning-based algorithms in both short-term and long-term traffic prediction. PMID:28672867

Spatiotemporal Recurrent Convolutional Networks for Traffic Prediction in Transportation Networks.

PubMed

Yu, Haiyang; Wu, Zhihai; Wang, Shuqin; Wang, Yunpeng; Ma, Xiaolei

2017-06-26

Predicting large-scale transportation network traffic has become an important and challenging topic in recent decades. Inspired by the domain knowledge of motion prediction, in which the future motion of an object can be predicted based on previous scenes, we propose a network grid representation method that can retain the fine-scale structure of a transportation network. Network-wide traffic speeds are converted into a series of static images and input into a novel deep architecture, namely, spatiotemporal recurrent convolutional networks (SRCNs), for traffic forecasting. The proposed SRCNs inherit the advantages of deep convolutional neural networks (DCNNs) and long short-term memory (LSTM) neural networks. The spatial dependencies of network-wide traffic can be captured by DCNNs, and the temporal dynamics can be learned by LSTMs. An experiment on a Beijing transportation network with 278 links demonstrates that SRCNs outperform other deep learning-based algorithms in both short-term and long-term traffic prediction.

Dynamics of Hierarchical Urban Green Space Patches and Implications for Management Policy.

PubMed

Yu, Zhoulu; Wang, Yaohui; Deng, Jinsong; Shen, Zhangquan; Wang, Ke; Zhu, Jinxia; Gan, Muye

2017-06-06

Accurately quantifying the variation of urban green space is the prerequisite for fully understanding its ecosystem services. However, knowledge about the spatiotemporal dynamics of urban green space is still insufficient due to multiple challenges that remain in mapping green spaces within heterogeneous urban environments. This paper uses the city of Hangzhou to demonstrate an analysis methodology that integrates sub-pixel mapping technology and landscape analysis to fully investigate the spatiotemporal pattern and variation of hierarchical urban green space patches. Firstly, multiple endmember spectral mixture analysis was applied to time series Landsat data to derive green space coverage at the sub-pixel level. Landscape metric analysis was then employed to characterize the variation pattern of urban green space patches. Results indicate that Hangzhou has experienced a significant loss of urban greenness, producing a more fragmented and isolated vegetation landscape. Additionally, a remarkable amelioration of urban greenness occurred in the city core from 2002 to 2013, characterized by the significant increase of small-sized green space patches. The green space network has been formed as a consequence of new urban greening strategies in Hangzhou. These strategies have greatly fragmented the built-up areas and enriched the diversity of the urban landscape. Gradient analysis further revealed a distinct pattern of urban green space landscape variation in the process of urbanization. By integrating both sub-pixel mapping technology and landscape analysis, our approach revealed the subtle variation of urban green space patches which are otherwise easy to overlook. Findings from this study will help us to refine our understanding of the evolution of heterogeneous urban environments.

Evaluating White Space of a Malaysian Secondary ELT Textbook

ERIC Educational Resources Information Center

Jin, Ng Yu

2010-01-01

Students are believed to learn at optimum level with materials with a lot of white space since the existence of vast number of words and limited amount of white space, especially in textbooks may contribute to the increase in anxiety among learners (Tomlinson, 1998). This study evaluates the amount of white space in terms of pixels in the…

Convolution neural-network-based detection of lung structures

NASA Astrophysics Data System (ADS)

Hasegawa, Akira; Lo, Shih-Chung B.; Freedman, Matthew T.; Mun, Seong K.

1994-05-01

Chest radiography is one of the most primary and widely used techniques in diagnostic imaging. Nowadays with the advent of digital radiology, the digital medical image processing techniques for digital chest radiographs have attracted considerable attention, and several studies on the computer-aided diagnosis (CADx) as well as on the conventional image processing techniques for chest radiographs have been reported. In the automatic diagnostic process for chest radiographs, it is important to outline the areas of the lungs, the heart, and the diaphragm. This is because the original chest radiograph is composed of important anatomic structures and, without knowing exact positions of the organs, the automatic diagnosis may result in unexpected detections. The automatic extraction of an anatomical structure from digital chest radiographs can be a useful tool for (1) the evaluation of heart size, (2) automatic detection of interstitial lung diseases, (3) automatic detection of lung nodules, and (4) data compression, etc. Based on the clearly defined boundaries of heart area, rib spaces, rib positions, and rib cage extracted, one should be able to use this information to facilitate the tasks of the CADx on chest radiographs. In this paper, we present an automatic scheme for the detection of lung field from chest radiographs by using a shift-invariant convolution neural network. A novel algorithm for smoothing boundaries of lungs is also presented.

The Convolutional Visual Network for Identification and Reconstruction of NOvA Events

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

Psihas, Fernanda

In 2016 the NOvA experiment released results for the observation of oscillations in the vμ and ve channels as well as ve cross section measurements using neutrinos from Fermilab’s NuMI beam. These and other measurements in progress rely on the accurate identification and reconstruction of the neutrino flavor and energy recorded by our detectors. This presentation describes the first application of convolutional neural network technology for event identification and reconstruction in particle detectors like NOvA. The Convolutional Visual Network (CVN) Algorithm was developed for identification, categorization, and reconstruction of NOvA events. It increased the selection efficiency of the ve appearancemore » signal by 40% and studies show potential impact to the vμ disappearance analysis.« less

Learning Traffic as Images: A Deep Convolutional Neural Network for Large-Scale Transportation Network Speed Prediction.

PubMed

Ma, Xiaolei; Dai, Zhuang; He, Zhengbing; Ma, Jihui; Wang, Yong; Wang, Yunpeng

2017-04-10

This paper proposes a convolutional neural network (CNN)-based method that learns traffic as images and predicts large-scale, network-wide traffic speed with a high accuracy. Spatiotemporal traffic dynamics are converted to images describing the time and space relations of traffic flow via a two-dimensional time-space matrix. A CNN is applied to the image following two consecutive steps: abstract traffic feature extraction and network-wide traffic speed prediction. The effectiveness of the proposed method is evaluated by taking two real-world transportation networks, the second ring road and north-east transportation network in Beijing, as examples, and comparing the method with four prevailing algorithms, namely, ordinary least squares, k-nearest neighbors, artificial neural network, and random forest, and three deep learning architectures, namely, stacked autoencoder, recurrent neural network, and long-short-term memory network. The results show that the proposed method outperforms other algorithms by an average accuracy improvement of 42.91% within an acceptable execution time. The CNN can train the model in a reasonable time and, thus, is suitable for large-scale transportation networks.

Learning Traffic as Images: A Deep Convolutional Neural Network for Large-Scale Transportation Network Speed Prediction

PubMed Central

Ma, Xiaolei; Dai, Zhuang; He, Zhengbing; Ma, Jihui; Wang, Yong; Wang, Yunpeng

2017-01-01

This paper proposes a convolutional neural network (CNN)-based method that learns traffic as images and predicts large-scale, network-wide traffic speed with a high accuracy. Spatiotemporal traffic dynamics are converted to images describing the time and space relations of traffic flow via a two-dimensional time-space matrix. A CNN is applied to the image following two consecutive steps: abstract traffic feature extraction and network-wide traffic speed prediction. The effectiveness of the proposed method is evaluated by taking two real-world transportation networks, the second ring road and north-east transportation network in Beijing, as examples, and comparing the method with four prevailing algorithms, namely, ordinary least squares, k-nearest neighbors, artificial neural network, and random forest, and three deep learning architectures, namely, stacked autoencoder, recurrent neural network, and long-short-term memory network. The results show that the proposed method outperforms other algorithms by an average accuracy improvement of 42.91% within an acceptable execution time. The CNN can train the model in a reasonable time and, thus, is suitable for large-scale transportation networks. PMID:28394270

Deep convolutional neural networks as strong gravitational lens detectors

NASA Astrophysics Data System (ADS)

Schaefer, C.; Geiger, M.; Kuntzer, T.; Kneib, J.-P.

2018-03-01

Context. Future large-scale surveys with high-resolution imaging will provide us with approximately 105 new strong galaxy-scale lenses. These strong-lensing systems will be contained in large data amounts, however, which are beyond the capacity of human experts to visually classify in an unbiased way. Aim. We present a new strong gravitational lens finder based on convolutional neural networks (CNNs). The method was applied to the strong-lensing challenge organized by the Bologna Lens Factory. It achieved first and third place, respectively, on the space-based data set and the ground-based data set. The goal was to find a fully automated lens finder for ground-based and space-based surveys that minimizes human inspection. Methods: We compared the results of our CNN architecture and three new variations ("invariant" "views" and "residual") on the simulated data of the challenge. Each method was trained separately five times on 17 000 simulated images, cross-validated using 3000 images, and then applied to a test set with 100 000 images. We used two different metrics for evaluation, the area under the receiver operating characteristic curve (AUC) score, and the recall with no false positive (Recall0FP). Results: For ground-based data, our best method achieved an AUC score of 0.977 and a Recall0FP of 0.50. For space-based data, our best method achieved an AUC score of 0.940 and a Recall0FP of 0.32. Adding dihedral invariance to the CNN architecture diminished the overall score on space-based data, but achieved a higher no-contamination recall. We found that using committees of five CNNs produced the best recall at zero contamination and consistently scored better AUC than a single CNN. Conclusions: We found that for every variation of our CNN lensfinder, we achieved AUC scores close to 1 within 6%. A deeper network did not outperform simpler CNN models either. This indicates that more complex networks are not needed to model the simulated lenses. To verify this, more

Automated Detection of Obstructive Sleep Apnea Events from a Single-Lead Electrocardiogram Using a Convolutional Neural Network.

PubMed

Urtnasan, Erdenebayar; Park, Jong-Uk; Joo, Eun-Yeon; Lee, Kyoung-Joung

2018-04-23

In this study, we propose a method for the automated detection of obstructive sleep apnea (OSA) from a single-lead electrocardiogram (ECG) using a convolutional neural network (CNN). A CNN model was designed with six optimized convolution layers including activation, pooling, and dropout layers. One-dimensional (1D) convolution, rectified linear units (ReLU), and max pooling were applied to the convolution, activation, and pooling layers, respectively. For training and evaluation of the CNN model, a single-lead ECG dataset was collected from 82 subjects with OSA and was divided into training (including data from 63 patients with 34,281 events) and testing (including data from 19 patients with 8571 events) datasets. Using this CNN model, a precision of 0.99%, a recall of 0.99%, and an F 1 -score of 0.99% were attained with the training dataset; these values were all 0.96% when the CNN was applied to the testing dataset. These results show that the proposed CNN model can be used to detect OSA accurately on the basis of a single-lead ECG. Ultimately, this CNN model may be used as a screening tool for those suspected to suffer from OSA.

Two projects in theoretical neuroscience: A convolution-based metric for neural membrane potentials and a combinatorial connectionist semantic network method

NASA Astrophysics Data System (ADS)

Evans, Garrett Nolan

In this work, I present two projects that both contribute to the aim of discovering how intelligence manifests in the brain. The first project is a method for analyzing recorded neural signals, which takes the form of a convolution-based metric on neural membrane potential recordings. Relying only on integral and algebraic operations, the metric compares the timing and number of spikes within recordings as well as the recordings' subthreshold features: summarizing differences in these with a single "distance" between the recordings. Like van Rossum's (2001) metric for spike trains, the metric is based on a convolution operation that it performs on the input data. The kernel used for the convolution is carefully chosen such that it produces a desirable frequency space response and, unlike van Rossum's kernel, causes the metric to be first order both in differences between nearby spike times and in differences between same-time membrane potential values: an important trait. The second project is a combinatorial syntax method for connectionist semantic network encoding. Combinatorial syntax has been a point on which those who support a symbol-processing view of intelligent processing and those who favor a connectionist view have had difficulty seeing eye-to-eye. Symbol-processing theorists have persuasively argued that combinatorial syntax is necessary for certain intelligent mental operations, such as reasoning by analogy. Connectionists have focused on the versatility and adaptability offered by self-organizing networks of simple processing units. With this project, I show that there is a way to reconcile the two perspectives and to ascribe a combinatorial syntax to a connectionist network. The critical principle is to interpret nodes, or units, in the connectionist network as bound integrations of the interpretations for nodes that they share links with. Nodes need not correspond exactly to neurons and may correspond instead to distributed sets, or assemblies, of

Compact SPAD-Based Pixel Architectures for Time-Resolved Image Sensors

PubMed Central

Perenzoni, Matteo; Pancheri, Lucio; Stoppa, David

2016-01-01

This paper reviews the state of the art of single-photon avalanche diode (SPAD) image sensors for time-resolved imaging. The focus of the paper is on pixel architectures featuring small pixel size (<25 μm) and high fill factor (>20%) as a key enabling technology for the successful implementation of high spatial resolution SPAD-based image sensors. A summary of the main CMOS SPAD implementations, their characteristics and integration challenges, is provided from the perspective of targeting large pixel arrays, where one of the key drivers is the spatial uniformity. The main analog techniques aimed at time-gated photon counting and photon timestamping suitable for compact and low-power pixels are critically discussed. The main features of these solutions are the adoption of analog counting techniques and time-to-analog conversion, in NMOS-only pixels. Reliable quantum-limited single-photon counting, self-referenced analog-to-digital conversion, time gating down to 0.75 ns and timestamping with 368 ps jitter are achieved. PMID:27223284

The pixel tracking telescope at the Fermilab Test Beam Facility

DOE PAGES

Kwan, Simon; Lei, CM; Menasce, Dario; ...

2016-03-01

An all silicon pixel telescope has been assembled and used at the Fermilab Test Beam Facility (FTBF) since 2009 to provide precise tracking information for different test beam experiments with a wide range of Detectors Under Test (DUTs) requiring high resolution measurement of the track impact point. The telescope is based on CMS pixel modules left over from the CMS forward pixel production. Eight planes are arranged to achieve a resolution of less than 8 μm on the 120 GeV proton beam transverse coordinate at the DUT position. In order to achieve such resolution with 100 × 150 μm 2more » pixel cells, the planes were tilted to 25 degrees to maximize charge sharing between pixels. Crucial for obtaining this performance is the alignment software, called Monicelli, specifically designed and optimized for this system. This paper will describe the telescope hardware, the data acquisition system and the alignment software constituting this particle tracking system for test beam users.« less

Segmentation and classification of colon glands with deep convolutional neural networks and total variation regularization.

PubMed

Kainz, Philipp; Pfeiffer, Michael; Urschler, Martin

2017-01-01

Segmentation of histopathology sections is a necessary preprocessing step for digital pathology. Due to the large variability of biological tissue, machine learning techniques have shown superior performance over conventional image processing methods. Here we present our deep neural network-based approach for segmentation and classification of glands in tissue of benign and malignant colorectal cancer, which was developed to participate in the GlaS@MICCAI2015 colon gland segmentation challenge. We use two distinct deep convolutional neural networks (CNN) for pixel-wise classification of Hematoxylin-Eosin stained images. While the first classifier separates glands from background, the second classifier identifies gland-separating structures. In a subsequent step, a figure-ground segmentation based on weighted total variation produces the final segmentation result by regularizing the CNN predictions. We present both quantitative and qualitative segmentation results on the recently released and publicly available Warwick-QU colon adenocarcinoma dataset associated with the GlaS@MICCAI2015 challenge and compare our approach to the simultaneously developed other approaches that participated in the same challenge. On two test sets, we demonstrate our segmentation performance and show that we achieve a tissue classification accuracy of 98% and 95%, making use of the inherent capability of our system to distinguish between benign and malignant tissue. Our results show that deep learning approaches can yield highly accurate and reproducible results for biomedical image analysis, with the potential to significantly improve the quality and speed of medical diagnoses.

Segmentation and classification of colon glands with deep convolutional neural networks and total variation regularization

PubMed Central

Kainz, Philipp; Pfeiffer, Michael

2017-01-01

Segmentation of histopathology sections is a necessary preprocessing step for digital pathology. Due to the large variability of biological tissue, machine learning techniques have shown superior performance over conventional image processing methods. Here we present our deep neural network-based approach for segmentation and classification of glands in tissue of benign and malignant colorectal cancer, which was developed to participate in the GlaS@MICCAI2015 colon gland segmentation challenge. We use two distinct deep convolutional neural networks (CNN) for pixel-wise classification of Hematoxylin-Eosin stained images. While the first classifier separates glands from background, the second classifier identifies gland-separating structures. In a subsequent step, a figure-ground segmentation based on weighted total variation produces the final segmentation result by regularizing the CNN predictions. We present both quantitative and qualitative segmentation results on the recently released and publicly available Warwick-QU colon adenocarcinoma dataset associated with the GlaS@MICCAI2015 challenge and compare our approach to the simultaneously developed other approaches that participated in the same challenge. On two test sets, we demonstrate our segmentation performance and show that we achieve a tissue classification accuracy of 98% and 95%, making use of the inherent capability of our system to distinguish between benign and malignant tissue. Our results show that deep learning approaches can yield highly accurate and reproducible results for biomedical image analysis, with the potential to significantly improve the quality and speed of medical diagnoses. PMID:29018612

Event-based Sensing for Space Situational Awareness

NASA Astrophysics Data System (ADS)

Cohen, G.; Afshar, S.; van Schaik, A.; Wabnitz, A.; Bessell, T.; Rutten, M.; Morreale, B.

A revolutionary type of imaging device, known as a silicon retina or event-based sensor, has recently been developed and is gaining in popularity in the field of artificial vision systems. These devices are inspired by a biological retina and operate in a significantly different way to traditional CCD-based imaging sensors. While a CCD produces frames of pixel intensities, an event-based sensor produces a continuous stream of events, each of which is generated when a pixel detects a change in log light intensity. These pixels operate asynchronously and independently, producing an event-based output with high temporal resolution. There are also no fixed exposure times, allowing these devices to offer a very high dynamic range independently for each pixel. Additionally, these devices offer high-speed, low power operation and a sparse spatiotemporal output. As a consequence, the data from these sensors must be interpreted in a significantly different way to traditional imaging sensors and this paper explores the advantages this technology provides for space imaging. The applicability and capabilities of event-based sensors for SSA applications are demonstrated through telescope field trials. Trial results have confirmed that the devices are capable of observing resident space objects from LEO through to GEO orbital regimes. Significantly, observations of RSOs were made during both day-time and nighttime (terminator) conditions without modification to the camera or optics. The event based sensor’s ability to image stars and satellites during day-time hours offers a dramatic capability increase for terrestrial optical sensors. This paper shows the field testing and validation of two different architectures of event-based imaging sensors. An eventbased sensor’s asynchronous output has an intrinsically low data-rate. In addition to low-bandwidth communications requirements, the low weight, low-power and high-speed make them ideally suitable to meeting the demanding

Study of cluster shapes in a monolithic active pixel detector

NASA Astrophysics Data System (ADS)

Maçzewski, ł.; Adamus, M.; Ciborowski, J.; Grzelak, G.; łużniak, P.; Nieżurawski, P.; Żarnecki, A. F.

2009-11-01

Beamstrahlung will constitute an important source of background in a pixel vertex detector at the future International Linear Collider. Electron and positron tracks of this origin impact the pixel planes at angles generally larger than those of secondary hadrons and the corresponding clusters are elongated. We report studies of cluster characteristics using test beam electron tracks incident at various angles on a MIMOSA-5 monolithic active pixel sensor matrix.

Convolutional coding combined with continuous phase modulation

NASA Technical Reports Server (NTRS)

Pizzi, S. V.; Wilson, S. G.

1985-01-01

Background theory and specific coding designs for combined coding/modulation schemes utilizing convolutional codes and continuous-phase modulation (CPM) are presented. In this paper the case of r = 1/2 coding onto a 4-ary CPM is emphasized, with short-constraint length codes presented for continuous-phase FSK, double-raised-cosine, and triple-raised-cosine modulation. Coding buys several decibels of coding gain over the Gaussian channel, with an attendant increase of bandwidth. Performance comparisons in the power-bandwidth tradeoff with other approaches are made.

Convolutional Neural Network for Histopathological Analysis of Osteosarcoma.

PubMed

Mishra, Rashika; Daescu, Ovidiu; Leavey, Patrick; Rakheja, Dinesh; Sengupta, Anita

2018-03-01

Pathologists often deal with high complexity and sometimes disagreement over osteosarcoma tumor classification due to cellular heterogeneity in the dataset. Segmentation and classification of histology tissue in H&E stained tumor image datasets is a challenging task because of intra-class variations, inter-class similarity, crowded context, and noisy data. In recent years, deep learning approaches have led to encouraging results in breast cancer and prostate cancer analysis. In this article, we propose convolutional neural network (CNN) as a tool to improve efficiency and accuracy of osteosarcoma tumor classification into tumor classes (viable tumor, necrosis) versus nontumor. The proposed CNN architecture contains eight learned layers: three sets of stacked two convolutional layers interspersed with max pooling layers for feature extraction and two fully connected layers with data augmentation strategies to boost performance. The use of a neural network results in higher accuracy of average 92% for the classification. We compare the proposed architecture with three existing and proven CNN architectures for image classification: AlexNet, LeNet, and VGGNet. We also provide a pipeline to calculate percentage necrosis in a given whole slide image. We conclude that the use of neural networks can assure both high accuracy and efficiency in osteosarcoma classification.

Metric learning with spectral graph convolutions on brain connectivity networks.

PubMed

Ktena, Sofia Ira; Parisot, Sarah; Ferrante, Enzo; Rajchl, Martin; Lee, Matthew; Glocker, Ben; Rueckert, Daniel

2018-04-01

Graph representations are often used to model structured data at an individual or population level and have numerous applications in pattern recognition problems. In the field of neuroscience, where such representations are commonly used to model structural or functional connectivity between a set of brain regions, graphs have proven to be of great importance. This is mainly due to the capability of revealing patterns related to brain development and disease, which were previously unknown. Evaluating similarity between these brain connectivity networks in a manner that accounts for the graph structure and is tailored for a particular application is, however, non-trivial. Most existing methods fail to accommodate the graph structure, discarding information that could be beneficial for further classification or regression analyses based on these similarities. We propose to learn a graph similarity metric using a siamese graph convolutional neural network (s-GCN) in a supervised setting. The proposed framework takes into consideration the graph structure for the evaluation of similarity between a pair of graphs, by employing spectral graph convolutions that allow the generalisation of traditional convolutions to irregular graphs and operates in the graph spectral domain. We apply the proposed model on two datasets: the challenging ABIDE database, which comprises functional MRI data of 403 patients with autism spectrum disorder (ASD) and 468 healthy controls aggregated from multiple acquisition sites, and a set of 2500 subjects from UK Biobank. We demonstrate the performance of the method for the tasks of classification between matching and non-matching graphs, as well as individual subject classification and manifold learning, showing that it leads to significantly improved results compared to traditional methods. Copyright © 2017 Elsevier Inc. All rights reserved.

Limits in point to point resolution of MOS based pixels detector arrays

NASA Astrophysics Data System (ADS)