Grassmannian sparse representations
NASA Astrophysics Data System (ADS)
Azary, Sherif; Savakis, Andreas
2015-05-01
We present Grassmannian sparse representations (GSR), a sparse representation Grassmann learning framework for efficient classification. Sparse representation classification offers a powerful approach for recognition in a variety of contexts. However, a major drawback of sparse representation methods is their computational performance and memory utilization for high-dimensional data. A Grassmann manifold is a space that promotes smooth surfaces where points represent subspaces and the relationship between points is defined by the mapping of an orthogonal matrix. Grassmann manifolds are well suited for computer vision problems because they promote high between-class discrimination and within-class clustering, while offering computational advantages by mapping each subspace onto a single point. The GSR framework combines Grassmannian kernels and sparse representations, including regularized least squares and least angle regression, to improve high accuracy recognition while overcoming the drawbacks of performance and dependencies on high dimensional data distributions. The effectiveness of GSR is demonstrated on computationally intensive multiview action sequences, three-dimensional action sequences, and face recognition datasets.
Multiple Sparse Representations Classification
Plenge, Esben; Klein, Stefan S.; Niessen, Wiro J.; Meijering, Erik
2015-01-01
Sparse representations classification (SRC) is a powerful technique for pixelwise classification of images and it is increasingly being used for a wide variety of image analysis tasks. The method uses sparse representation and learned redundant dictionaries to classify image pixels. In this empirical study we propose to further leverage the redundancy of the learned dictionaries to achieve a more accurate classifier. In conventional SRC, each image pixel is associated with a small patch surrounding it. Using these patches, a dictionary is trained for each class in a supervised fashion. Commonly, redundant/overcomplete dictionaries are trained and image patches are sparsely represented by a linear combination of only a few of the dictionary elements. Given a set of trained dictionaries, a new patch is sparse coded using each of them, and subsequently assigned to the class whose dictionary yields the minimum residual energy. We propose a generalization of this scheme. The method, which we call multiple sparse representations classification (mSRC), is based on the observation that an overcomplete, class specific dictionary is capable of generating multiple accurate and independent estimates of a patch belonging to the class. So instead of finding a single sparse representation of a patch for each dictionary, we find multiple, and the corresponding residual energies provides an enhanced statistic which is used to improve classification. We demonstrate the efficacy of mSRC for three example applications: pixelwise classification of texture images, lumen segmentation in carotid artery magnetic resonance imaging (MRI), and bifurcation point detection in carotid artery MRI. We compare our method with conventional SRC, K-nearest neighbor, and support vector machine classifiers. The results show that mSRC outperforms SRC and the other reference methods. In addition, we present an extensive evaluation of the effect of the main mSRC parameters: patch size, dictionary size, and
NASA Astrophysics Data System (ADS)
Yang, Yongchao; Nagarajaiah, Satish
2014-03-01
This paper addresses two problems in structural damage identification: locating damage and assessing damage severity, which are incorporated into the classification framework based on the theory of sparse representation (SR) and compressed sensing (CS). The sparsity nature implied in the classification problem itself is exploited, establishing a sparse representation framework for damage identification. Specifically, the proposed method consists of two steps: feature extraction and classification. In the feature extraction step, the modal features of both the test structure and the reference structure model are first blindly extracted by the unsupervised complexity pursuit (CP) algorithm. Then in the classification step, expressing the test modal feature as a linear combination of the bases of the over-complete reference feature dictionary—constructed by concatenating all modal features of all candidate damage classes—builds a highly underdetermined linear system of equations with an underlying sparse representation, which can be correctly recovered by ℓ1-minimization; the non-zero entry in the recovered sparse representation directly assigns the damage class which the test structure (feature) belongs to. The two-step CP-SR damage identification method alleviates the training process required by traditional pattern recognition based methods. In addition, the reference feature dictionary can be of small size by formulating the issues of locating damage and assessing damage extent as a two-stage procedure and by taking advantage of the robustness of the SR framework. Numerical simulations and experimental study are conducted to verify the developed CP-SR method. The problems of identifying multiple damage, using limited sensors and partial features, and the performance under heavy noise and random excitation are investigated, and promising results are obtained.
Jiang, Lu-Lu; Luo, Mei-Fu; Zhang, Yu; Yu, Xin-Jie; Kong, Wen-Wen; Liu, Fei
2014-01-01
An identification method based on sparse representation (SR) combined with autoencoder network (AN) manifold learning was proposed for discriminating the varieties of transmission fluid by using near infrared (NIR) spectroscopy technology. NIR transmittance spectra from 600 to 1 800 nm were collected from 300 transmission fluid samples of five varieties (each variety consists of 60 samples). For each variety, 30 samples were randomly selected as training set (totally 150 samples), and the rest 30 ones as testing set (totally 150 samples). Autoencoder network manifold learning was applied to obtain the characteristic information in the 600-1800 nm spectra and the number of characteristics was reduced to 10. Principal component analysis (PCA) was applied to extract several relevant variables to represent the useful information of spectral variables. All of the training samples made up a data dictionary of the sparse representation (SR). Then the transmission fluid variety identification problem was reduced to the problem as how to represent the testing samples from the data dictionary (training samples data). The identification result thus could be achieved by solving the L-1 norm-based optimization problem. We compared the effectiveness of the proposed method with that of linear discriminant analysis (LDA), least squares support vector machine (LS-SVM) and sparse representation (SR) using the relevant variables selected by principal component analysis (PCA) and AN. Experimental results demonstrated that the overall identification accuracy of the proposed method for the five transmission fluid varieties was 97.33% by AN-SR, which was significantly higher than that of LDA or LS-SVM. Therefore, the proposed method can provide a new effective method for identification of transmission fluid variety.
Fingerprint Compression Based on Sparse Representation.
Shao, Guangqi; Wu, Yanping; A, Yong; Liu, Xiao; Guo, Tiande
2014-02-01
A new fingerprint compression algorithm based on sparse representation is introduced. Obtaining an overcomplete dictionary from a set of fingerprint patches allows us to represent them as a sparse linear combination of dictionary atoms. In the algorithm, we first construct a dictionary for predefined fingerprint image patches. For a new given fingerprint images, represent its patches according to the dictionary by computing l(0)-minimization and then quantize and encode the representation. In this paper, we consider the effect of various factors on compression results. Three groups of fingerprint images are tested. The experiments demonstrate that our algorithm is efficient compared with several competing compression techniques (JPEG, JPEG 2000, and WSQ), especially at high compression ratios. The experiments also illustrate that the proposed algorithm is robust to extract minutiae.
SAR Image Despeckling Via Structural Sparse Representation
NASA Astrophysics Data System (ADS)
Lu, Ting; Li, Shutao; Fang, Leyuan; Benediktsson, Jón Atli
2016-12-01
A novel synthetic aperture radar (SAR) image despeckling method based on structural sparse representation is introduced. The proposed method utilizes the fact that different regions in SAR images correspond to varying terrain reflectivity. Therefore, SAR images can be split into a heterogeneous class (with a varied terrain reflectivity) and a homogeneous class (with a constant terrain reflectivity). In the proposed method, different sparse representation based despeckling schemes are designed by combining the different region characteristics in SAR images. For heterogeneous regions with rich structure and texture information, structural dictionaries are learned to appropriately represent varied structural characteristics. Specifically, each patch in these regions is sparsely coded with the best fitted structural dictionary, thus good structure preservation can be obtained. For homogenous regions without rich structure and texture information, the highly redundant photometric self-similarity is exploited to suppress speckle noise without introducing artifacts. That is achieved by firstly learning the sub-dictionary, then simultaneously sparsely coding for each group of photometrically similar image patches. Visual and objective experimental results demonstrate the superiority of the proposed method over the-state-of-the-art methods.
Sparse representation for the ISAR image reconstruction
NASA Astrophysics Data System (ADS)
Hu, Mengqi; Montalbo, John; Li, Shuxia; Sun, Ligang; Qiao, Zhijun G.
2016-05-01
In this paper, a sparse representation of the data for an inverse synthetic aperture radar (ISAR) system is provided in two dimensions. The proposed sparse representation motivates the use a of a Convex Optimization that recovers the image with far less samples, which is required by Nyquist-Shannon sampling theorem to increases the efficiency and decrease the cost of calculation in radar imaging.
Discriminative Sparse Representations in Hyperspectral Imagery
2010-03-01
classification , and unsupervised labeling (clustering) [2, 3, 4, 5, 6, 7, 8]. Recently, a non-parametric (Bayesian) approach to sparse modeling and com...DISCRIMINATIVE SPARSE REPRESENTATIONS IN HYPERSPECTRAL IMAGERY By Alexey Castrodad, Zhengming Xing John Greer, Edward Bosch Lawrence Carin and...00-00-2010 to 00-00-2010 4. TITLE AND SUBTITLE Discriminative Sparse Representations in Hyperspectral Imagery 5a. CONTRACT NUMBER 5b. GRANT
Visual Tracking Based on Extreme Learning Machine and Sparse Representation
Wang, Baoxian; Tang, Linbo; Yang, Jinglin; Zhao, Baojun; Wang, Shuigen
2015-01-01
The existing sparse representation-based visual trackers mostly suffer from both being time consuming and having poor robustness problems. To address these issues, a novel tracking method is presented via combining sparse representation and an emerging learning technique, namely extreme learning machine (ELM). Specifically, visual tracking can be divided into two consecutive processes. Firstly, ELM is utilized to find the optimal separate hyperplane between the target observations and background ones. Thus, the trained ELM classification function is able to remove most of the candidate samples related to background contents efficiently, thereby reducing the total computational cost of the following sparse representation. Secondly, to further combine ELM and sparse representation, the resultant confidence values (i.e., probabilities to be a target) of samples on the ELM classification function are used to construct a new manifold learning constraint term of the sparse representation framework, which tends to achieve robuster results. Moreover, the accelerated proximal gradient method is used for deriving the optimal solution (in matrix form) of the constrained sparse tracking model. Additionally, the matrix form solution allows the candidate samples to be calculated in parallel, thereby leading to a higher efficiency. Experiments demonstrate the effectiveness of the proposed tracker. PMID:26506359
Visual tracking based on extreme learning machine and sparse representation.
Wang, Baoxian; Tang, Linbo; Yang, Jinglin; Zhao, Baojun; Wang, Shuigen
2015-10-22
The existing sparse representation-based visual trackers mostly suffer from both being time consuming and having poor robustness problems. To address these issues, a novel tracking method is presented via combining sparse representation and an emerging learning technique, namely extreme learning machine (ELM). Specifically, visual tracking can be divided into two consecutive processes. Firstly, ELM is utilized to find the optimal separate hyperplane between the target observations and background ones. Thus, the trained ELM classification function is able to remove most of the candidate samples related to background contents efficiently, thereby reducing the total computational cost of the following sparse representation. Secondly, to further combine ELM and sparse representation, the resultant confidence values (i.e., probabilities to be a target) of samples on the ELM classification function are used to construct a new manifold learning constraint term of the sparse representation framework, which tends to achieve robuster results. Moreover, the accelerated proximal gradient method is used for deriving the optimal solution (in matrix form) of the constrained sparse tracking model. Additionally, the matrix form solution allows the candidate samples to be calculated in parallel, thereby leading to a higher efficiency. Experiments demonstrate the effectiveness of the proposed tracker.
Ensemble polarimetric SAR image classification based on contextual sparse representation
NASA Astrophysics Data System (ADS)
Zhang, Lamei; Wang, Xiao; Zou, Bin; Qiao, Zhijun
2016-05-01
Polarimetric SAR image interpretation has become one of the most interesting topics, in which the construction of the reasonable and effective technique of image classification is of key importance. Sparse representation represents the data using the most succinct sparse atoms of the over-complete dictionary and the advantages of sparse representation also have been confirmed in the field of PolSAR classification. However, it is not perfect, like the ordinary classifier, at different aspects. So ensemble learning is introduced to improve the issue, which makes a plurality of different learners training and obtained the integrated results by combining the individual learner to get more accurate and ideal learning results. Therefore, this paper presents a polarimetric SAR image classification method based on the ensemble learning of sparse representation to achieve the optimal classification.
Maxdenominator Reweighted Sparse Representation for Tumor Classification
Li, Weibiao; Liao, Bo; Zhu, Wen; Chen, Min; Peng, Li; Wei, Xiaohui; Gu, Changlong; Li, Keqin
2017-01-01
The classification of tumors is crucial for the proper treatment of cancer. Sparse representation-based classifier (SRC) exhibits good classification performance and has been successfully used to classify tumors using gene expression profile data. In this study, we propose a three-step maxdenominator reweighted sparse representation classification (MRSRC) method to classify tumors. First, we extract a set of metagenes from the training samples. These metagenes can capture the structures inherent to the data and are more effective for classification than the original gene expression data. Second, we use a reweighted regularization method to obtain the sparse representation coefficients. Reweighted regularization can enhance sparsity and obtain better sparse representation coefficients. Third, we classify the data by utilizing a maxdenominator residual error function. Maxdenominator strategy can reduce the residual error and improve the accuracy of the final classification. Extensive experiments using publicly available gene expression profile data sets show that the performance of MRSRC is comparable with or better than many existing representative methods. PMID:28393883
Automatic landslide and mudflow detection method via multichannel sparse representation
NASA Astrophysics Data System (ADS)
Chao, Chen; Zhou, Jianjun; Hao, Zhuo; Sun, Bo; He, Jun; Ge, Fengxiang
2015-10-01
Landslide and mudflow detection is an important application of aerial images and high resolution remote sensing images, which is crucial for national security and disaster relief. Since the high resolution images are often large in size, it's necessary to develop an efficient algorithm for landslide and mudflow detection. Based on the theory of sparse representation and, we propose a novel automatic landslide and mudflow detection method in this paper, which combines multi-channel sparse representation and eight neighbor judgment methods. The whole process of the detection is totally automatic. We make the experiment on a high resolution image of ZhouQu district of Gansu province in China on August, 2010 and get a promising result which proved the effective of using sparse representation on landslide and mudflow detection.
Feature Selection and Pedestrian Detection Based on Sparse Representation
Yao, Shihong; Wang, Tao; Shen, Weiming; Pan, Shaoming; Chong, Yanwen; Ding, Fei
2015-01-01
Pedestrian detection have been currently devoted to the extraction of effective pedestrian features, which has become one of the obstacles in pedestrian detection application according to the variety of pedestrian features and their large dimension. Based on the theoretical analysis of six frequently-used features, SIFT, SURF, Haar, HOG, LBP and LSS, and their comparison with experimental results, this paper screens out the sparse feature subsets via sparse representation to investigate whether the sparse subsets have the same description abilities and the most stable features. When any two of the six features are fused, the fusion feature is sparsely represented to obtain its important components. Sparse subsets of the fusion features can be rapidly generated by avoiding calculation of the corresponding index of dimension numbers of these feature descriptors; thus, the calculation speed of the feature dimension reduction is improved and the pedestrian detection time is reduced. Experimental results show that sparse feature subsets are capable of keeping the important components of these six feature descriptors. The sparse features of HOG and LSS possess the same description ability and consume less time compared with their full features. The ratios of the sparse feature subsets of HOG and LSS to their full sets are the highest among the six, and thus these two features can be used to best describe the characteristics of the pedestrian and the sparse feature subsets of the combination of HOG-LSS show better distinguishing ability and parsimony. PMID:26295480
Remote sensing image fusion via wavelet transform and sparse representation
NASA Astrophysics Data System (ADS)
Cheng, Jian; Liu, Haijun; Liu, Ting; Wang, Feng; Li, Hongsheng
2015-06-01
In this paper, we propose a remote sensing image fusion method which combines the wavelet transform and sparse representation to obtain fusion images with high spectral resolution and high spatial resolution. Firstly, intensity-hue-saturation (IHS) transform is applied to Multi-Spectral (MS) images. Then, wavelet transform is used to the intensity component of MS images and the Panchromatic (Pan) image to construct the multi-scale representation respectively. With the multi-scale representation, different fusion strategies are taken on the low-frequency and the high-frequency sub-images. Sparse representation with training dictionary is introduced into the low-frequency sub-image fusion. The fusion rule for the sparse representation coefficients of the low-frequency sub-images is defined by the spatial frequency maximum. For high-frequency sub-images with prolific detail information, the fusion rule is established by the images information fusion measurement indicator. Finally, the fused results are obtained through inverse wavelet transform and inverse IHS transform. The wavelet transform has the ability to extract the spectral information and the global spatial details from the original pairwise images, while sparse representation can extract the local structures of images effectively. Therefore, our proposed fusion method can well preserve the spectral information and the spatial detail information of the original images. The experimental results on the remote sensing images have demonstrated that our proposed method could well maintain the spectral characteristics of fusion images with a high spatial resolution.
Learning Stable Multilevel Dictionaries for Sparse Representations.
Thiagarajan, Jayaraman J; Ramamurthy, Karthikeyan Natesan; Spanias, Andreas
2015-09-01
Sparse representations using learned dictionaries are being increasingly used with success in several data processing and machine learning applications. The increasing need for learning sparse models in large-scale applications motivates the development of efficient, robust, and provably good dictionary learning algorithms. Algorithmic stability and generalizability are desirable characteristics for dictionary learning algorithms that aim to build global dictionaries, which can efficiently model any test data similar to the training samples. In this paper, we propose an algorithm to learn dictionaries for sparse representations from large scale data, and prove that the proposed learning algorithm is stable and generalizable asymptotically. The algorithm employs a 1-D subspace clustering procedure, the K-hyperline clustering, to learn a hierarchical dictionary with multiple levels. We also propose an information-theoretic scheme to estimate the number of atoms needed in each level of learning and develop an ensemble approach to learn robust dictionaries. Using the proposed dictionaries, the sparse code for novel test data can be computed using a low-complexity pursuit procedure. We demonstrate the stability and generalization characteristics of the proposed algorithm using simulations. We also evaluate the utility of the multilevel dictionaries in compressed recovery and subspace learning applications.
Hyperspectral Image Classification via Kernel Sparse Representation
2013-01-01
and sparse representations, image processing, wavelets, multirate systems , and filter banks . Nasser M. Nasrabadi (S’80–M’84–SM’92–F’01) received the...ery, sampling, multirate systems , filter banks , transforms, wavelets, and their applications in signal analysis, compression, processing, and...University of Pavia and the Center of Pavia images, are urban images acquired by the Reflective Optics System Imaging Spectrom- eter (ROSIS). The ROSIS
Sparse Representation for Time-Series Classification
2015-02-08
February 8, 2015 16:49 World Scientific Review Volume - 9in x 6in ” time - series classification” page 1 Chapter 1 Sparse Representation for Time - Series ...studies the problem of time - series classification and presents an overview of recent developments in the area of feature extraction and information...problem of target classification, and more generally time - series classification, in two main directions, feature extraction and information fusion. 1
Aerial Scene Recognition using Efficient Sparse Representation
Cheriyadat, Anil M
2012-01-01
Advanced scene recognition systems for processing large volumes of high-resolution aerial image data are in great demand today. However, automated scene recognition remains a challenging problem. Efficient encoding and representation of spatial and structural patterns in the imagery are key in developing automated scene recognition algorithms. We describe an image representation approach that uses simple and computationally efficient sparse code computation to generate accurate features capable of producing excellent classification performance using linear SVM kernels. Our method exploits unlabeled low-level image feature measurements to learn a set of basis vectors. We project the low-level features onto the basis vectors and use simple soft threshold activation function to derive the sparse features. The proposed technique generates sparse features at a significantly lower computational cost than other methods~\\cite{Yang10, newsam11}, yet it produces comparable or better classification accuracy. We apply our technique to high-resolution aerial image datasets to quantify the aerial scene classification performance. We demonstrate that the dense feature extraction and representation methods are highly effective for automatic large-facility detection on wide area high-resolution aerial imagery.
Efficient visual tracking via low-complexity sparse representation
NASA Astrophysics Data System (ADS)
Lu, Weizhi; Zhang, Jinglin; Kpalma, Kidiyo; Ronsin, Joseph
2015-12-01
Thanks to its good performance on object recognition, sparse representation has recently been widely studied in the area of visual object tracking. Up to now, little attention has been paid to the complexity of sparse representation, while most works are focused on the performance improvement. By reducing the computation load related to sparse representation hundreds of times, this paper proposes by far the most computationally efficient tracking approach based on sparse representation. The proposal simply consists of two stages of sparse representation, one is for object detection and the other for object validation. Experimentally, it achieves better performance than some state-of-the-art methods in both accuracy and speed.
Joint sparse representation for robust multimodal biometrics recognition.
Shekhar, Sumit; Patel, Vishal M; Nasrabadi, Nasser M; Chellappa, Rama
2014-01-01
Traditional biometric recognition systems rely on a single biometric signature for authentication. While the advantage of using multiple sources of information for establishing the identity has been widely recognized, computational models for multimodal biometrics recognition have only recently received attention. We propose a multimodal sparse representation method, which represents the test data by a sparse linear combination of training data, while constraining the observations from different modalities of the test subject to share their sparse representations. Thus, we simultaneously take into account correlations as well as coupling information among biometric modalities. A multimodal quality measure is also proposed to weigh each modality as it gets fused. Furthermore, we also kernelize the algorithm to handle nonlinearity in data. The optimization problem is solved using an efficient alternative direction method. Various experiments show that the proposed method compares favorably with competing fusion-based methods.
Sparse representation for color image restoration.
Mairal, Julien; Elad, Michael; Sapiro, Guillermo
2008-01-01
Sparse representations of signals have drawn considerable interest in recent years. The assumption that natural signals, such as images, admit a sparse decomposition over a redundant dictionary leads to efficient algorithms for handling such sources of data. In particular, the design of well adapted dictionaries for images has been a major challenge. The K-SVD has been recently proposed for this task and shown to perform very well for various grayscale image processing tasks. In this paper, we address the problem of learning dictionaries for color images and extend the K-SVD-based grayscale image denoising algorithm that appears in. This work puts forward ways for handling nonhomogeneous noise and missing information, paving the way to state-of-the-art results in applications such as color image denoising, demosaicing, and inpainting, as demonstrated in this paper.
Neonatal atlas construction using sparse representation.
Shi, Feng; Wang, Li; Wu, Guorong; Li, Gang; Gilmore, John H; Lin, Weili; Shen, Dinggang
2014-09-01
Atlas construction generally includes first an image registration step to normalize all images into a common space and then an atlas building step to fuse the information from all the aligned images. Although numerous atlas construction studies have been performed to improve the accuracy of the image registration step, unweighted or simply weighted average is often used in the atlas building step. In this article, we propose a novel patch-based sparse representation method for atlas construction after all images have been registered into the common space. By taking advantage of local sparse representation, more anatomical details can be recovered in the built atlas. To make the anatomical structures spatially smooth in the atlas, the anatomical feature constraints on group structure of representations and also the overlapping of neighboring patches are imposed to ensure the anatomical consistency between neighboring patches. The proposed method has been applied to 73 neonatal MR images with poor spatial resolution and low tissue contrast, for constructing a neonatal brain atlas with sharp anatomical details. Experimental results demonstrate that the proposed method can significantly enhance the quality of the constructed atlas by discovering more anatomical details especially in the highly convoluted cortical regions. The resulting atlas demonstrates superior performance of our atlas when applied to spatially normalizing three different neonatal datasets, compared with other start-of-the-art neonatal brain atlases.
Automatic target recognition via sparse representations
NASA Astrophysics Data System (ADS)
Estabridis, Katia
2010-04-01
Automatic target recognition (ATR) based on the emerging technology of Compressed Sensing (CS) can considerably improve accuracy, speed and cost associated with these types of systems. An image based ATR algorithm has been built upon this new theory, which can perform target detection and recognition in a low dimensional space. Compressed dictionaries (A) are formed to include rotational information for a scale of interest. The algorithm seeks to identify y(test sample) as a linear combination of the dictionary elements : y=Ax, where A ∈ Rnxm(n<
Improving sparse representation algorithms for maritime video processing
NASA Astrophysics Data System (ADS)
Smith, L. N.; Nichols, J. M.; Waterman, J. R.; Olson, C. C.; Judd, K. P.
2012-06-01
We present several improvements to published algorithms for sparse image modeling with the goal of improving processing of imagery of small watercraft in littoral environments. The first improvement is to the K-SVD algorithm for training over-complete dictionaries, which are used in sparse representations. It is shown that the training converges significantly faster by incorporating multiple dictionary (i.e., codebook) update stages in each training iteration. The paper also provides several useful and practical lessons learned from our experience with sparse representations. Results of three applications of sparse representation are presented and compared to the state-of-the-art methods; image compression, image denoising, and super-resolution.
Prostate segmentation by sparse representation based classification
Gao, Yaozong; Liao, Shu; Shen, Dinggang
2012-01-01
Purpose: The segmentation of prostate in CT images is of essential importance to external beam radiotherapy, which is one of the major treatments for prostate cancer nowadays. During the radiotherapy, the prostate is radiated by high-energy x rays from different directions. In order to maximize the dose to the cancer and minimize the dose to the surrounding healthy tissues (e.g., bladder and rectum), the prostate in the new treatment image needs to be accurately localized. Therefore, the effectiveness and efficiency of external beam radiotherapy highly depend on the accurate localization of the prostate. However, due to the low contrast of the prostate with its surrounding tissues (e.g., bladder), the unpredicted prostate motion, and the large appearance variations across different treatment days, it is challenging to segment the prostate in CT images. In this paper, the authors present a novel classification based segmentation method to address these problems. Methods: To segment the prostate, the proposed method first uses sparse representation based classification (SRC) to enhance the prostate in CT images by pixel-wise classification, in order to overcome the limitation of poor contrast of the prostate images. Then, based on the classification results, previous segmented prostates of the same patient are used as patient-specific atlases to align onto the current treatment image and the majority voting strategy is finally adopted to segment the prostate. In order to address the limitations of the traditional SRC in pixel-wise classification, especially for the purpose of segmentation, the authors extend SRC from the following four aspects: (1) A discriminant subdictionary learning method is proposed to learn a discriminant and compact representation of training samples for each class so that the discriminant power of SRC can be increased and also SRC can be applied to the large-scale pixel-wise classification. (2) The L1 regularized sparse coding is replaced by
Sparse representations for online-learning-based hyperspectral image compression.
Ülkü, İrem; Töreyin, Behçet Uğur
2015-10-10
Sparse models provide data representations in the fewest possible number of nonzero elements. This inherent characteristic enables sparse models to be utilized for data compression purposes. Hyperspectral data is large in size. In this paper, a framework for sparsity-based hyperspectral image compression methods using online learning is proposed. There are various sparse optimization models. A comparative analysis of sparse representations in terms of their hyperspectral image compression performance is presented. For this purpose, online-learning-based hyperspectral image compression methods are proposed using four different sparse representations. Results indicate that, independent of the sparsity models, online-learning-based hyperspectral data compression schemes yield the best compression performances for data rates of 0.1 and 0.3 bits per sample, compared to other state-of-the-art hyperspectral data compression techniques, in terms of image quality measured as average peak signal-to-noise ratio.
Extracting pure endmembers using symmetric sparse representation for hyperspectral imagery
NASA Astrophysics Data System (ADS)
Sun, Weiwei; Liu, Chun; Sun, Yanwei; Li, Weiyue; Li, Jialin
2016-10-01
This article proposes a symmetric sparse representation (SSR) method to extract pure endmembers from hyperspectral imagery (HSI). The SSR combines the features of the linear unmixing model and the sparse subspace clustering model of endmembers, and it assumes that the desired endmembers and all the HSI pixel points can be sparsely represented by each other. It formulates the endmember extraction problem into a famous program of archetypal analysis, and accordingly, extracting pure endmembers can be transformed as finding the archetypes in the minimal convex hull containing all the HSI pixel points. The vector quantization scheme is adopted to help in carefully choosing the initial pure endmembers, and the archetypal analysis program is solved using the simple projected gradient algorithm. Seven state-of-the-art methods are implemented to make comparisons with the SSR on both synthetic and real hyperspectral images. Experimental results show that the SSR outperforms all the seven methods in spectral angle distance and root-mean-square error, and it can be a good alternative choice for extracting pure endmembers from HSI data.
Sparse coding based feature representation method for remote sensing images
NASA Astrophysics Data System (ADS)
Oguslu, Ender
In this dissertation, we study sparse coding based feature representation method for the classification of multispectral and hyperspectral images (HSI). The existing feature representation systems based on the sparse signal model are computationally expensive, requiring to solve a convex optimization problem to learn a dictionary. A sparse coding feature representation framework for the classification of HSI is presented that alleviates the complexity of sparse coding through sub-band construction, dictionary learning, and encoding steps. In the framework, we construct the dictionary based upon the extracted sub-bands from the spectral representation of a pixel. In the encoding step, we utilize a soft threshold function to obtain sparse feature representations for HSI. Experimental results showed that a randomly selected dictionary could be as effective as a dictionary learned from optimization. The new representation usually has a very high dimensionality requiring a lot of computational resources. In addition, the spatial information of the HSI data has not been included in the representation. Thus, we modify the framework by incorporating the spatial information of the HSI pixels and reducing the dimension of the new sparse representations. The enhanced model, called sparse coding based dense feature representation (SC-DFR), is integrated with a linear support vector machine (SVM) and a composite kernels SVM (CKSVM) classifiers to discriminate different types of land cover. We evaluated the proposed algorithm on three well known HSI datasets and compared our method to four recently developed classification methods: SVM, CKSVM, simultaneous orthogonal matching pursuit (SOMP) and image fusion and recursive filtering (IFRF). The results from the experiments showed that the proposed method can achieve better overall and average classification accuracies with a much more compact representation leading to more efficient sparse models for HSI classification. To further
Optical double-image encryption and authentication by sparse representation.
Mohammed, Emad A; Saadon, H L
2016-12-10
An optical double-image encryption and authentication method by sparse representation is proposed. The information from double-image encryption can be integrated into a sparse representation. Unlike the traditional double-image encryption technique, only sparse (partial) data from the encrypted data is adopted for the authentication process. Simulation results demonstrate that the correct authentication results are achieved even with partial information from the encrypted data. The randomly selected sparse encrypted information will be used as an effective key for a security system. Therefore, the proposed method is feasible, effective, and can provide an additional security layer for optical security systems. In addition, the method also achieved the general requirements of storage and transmission due to a high reduction of the encrypted information.
Pseudo spectral Chebyshev representation of few-group cross sections on sparse grids
Bokov, P. M.; Botes, D.; Zimin, V. G.
2012-07-01
This paper presents a pseudo spectral method for representing few-group homogenised cross sections, based on hierarchical polynomial interpolation. The interpolation is performed on a multi-dimensional sparse grid built from Chebyshev nodes. The representation is assembled directly from the samples using basis functions that are constructed as tensor products of the classical one-dimensional Lagrangian interpolation functions. The advantage of this representation is that it combines the accuracy of Chebyshev interpolation with the efficiency of sparse grid methods. As an initial test, this interpolation method was used to construct a representation for the two-group macroscopic cross sections of a VVER pin cell. (authors)
Yao, Jincao; Yu, Huimin; Hu, Roland
2017-01-01
This paper introduces a new implicit-kernel-sparse-shape-representation-based object segmentation framework. Given an input object whose shape is similar to some of the elements in the training set, the proposed model can automatically find a cluster of implicit kernel sparse neighbors to approximately represent the input shape and guide the segmentation. A distance-constrained probabilistic definition together with a dualization energy term is developed to connect high-level shape representation and low-level image information. We theoretically prove that our model not only derives from two projected convex sets but is also equivalent to a sparse-reconstruction-error-based representation in the Hilbert space. Finally, a "wake-sleep"-based segmentation framework is applied to drive the evolutionary curve to recover the original shape of the object. We test our model on two public datasets. Numerical experiments on both synthetic images and real applications show the superior capabilities of the proposed framework.
Medical Image Fusion Based on Feature Extraction and Sparse Representation
Wei, Gao; Zongxi, Song
2017-01-01
As a novel multiscale geometric analysis tool, sparse representation has shown many advantages over the conventional image representation methods. However, the standard sparse representation does not take intrinsic structure and its time complexity into consideration. In this paper, a new fusion mechanism for multimodal medical images based on sparse representation and decision map is proposed to deal with these problems simultaneously. Three decision maps are designed including structure information map (SM) and energy information map (EM) as well as structure and energy map (SEM) to make the results reserve more energy and edge information. SM contains the local structure feature captured by the Laplacian of a Gaussian (LOG) and EM contains the energy and energy distribution feature detected by the mean square deviation. The decision map is added to the normal sparse representation based method to improve the speed of the algorithm. Proposed approach also improves the quality of the fused results by enhancing the contrast and reserving more structure and energy information from the source images. The experiment results of 36 groups of CT/MR, MR-T1/MR-T2, and CT/PET images demonstrate that the method based on SR and SEM outperforms five state-of-the-art methods. PMID:28321246
Online Hierarchical Sparse Representation of Multifeature for Robust Object Tracking
Qu, Shiru
2016-01-01
Object tracking based on sparse representation has given promising tracking results in recent years. However, the trackers under the framework of sparse representation always overemphasize the sparse representation and ignore the correlation of visual information. In addition, the sparse coding methods only encode the local region independently and ignore the spatial neighborhood information of the image. In this paper, we propose a robust tracking algorithm. Firstly, multiple complementary features are used to describe the object appearance; the appearance model of the tracked target is modeled by instantaneous and stable appearance features simultaneously. A two-stage sparse-coded method which takes the spatial neighborhood information of the image patch and the computation burden into consideration is used to compute the reconstructed object appearance. Then, the reliability of each tracker is measured by the tracking likelihood function of transient and reconstructed appearance models. Finally, the most reliable tracker is obtained by a well established particle filter framework; the training set and the template library are incrementally updated based on the current tracking results. Experiment results on different challenging video sequences show that the proposed algorithm performs well with superior tracking accuracy and robustness. PMID:27630710
Representation-Independent Iteration of Sparse Data Arrays
NASA Technical Reports Server (NTRS)
James, Mark
2007-01-01
An approach is defined that describes a method of iterating over massively large arrays containing sparse data using an approach that is implementation independent of how the contents of the sparse arrays are laid out in memory. What is unique and important here is the decoupling of the iteration over the sparse set of array elements from how they are internally represented in memory. This enables this approach to be backward compatible with existing schemes for representing sparse arrays as well as new approaches. What is novel here is a new approach for efficiently iterating over sparse arrays that is independent of the underlying memory layout representation of the array. A functional interface is defined for implementing sparse arrays in any modern programming language with a particular focus for the Chapel programming language. Examples are provided that show the translation of a loop that computes a matrix vector product into this representation for both the distributed and not-distributed cases. This work is directly applicable to NASA and its High Productivity Computing Systems (HPCS) program that JPL and our current program are engaged in. The goal of this program is to create powerful, scalable, and economically viable high-powered computer systems suitable for use in national security and industry by 2010. This is important to NASA for its computationally intensive requirements for analyzing and understanding the volumes of science data from our returned missions.
Sparse representation-based image restoration via nonlocal supervised coding
NASA Astrophysics Data System (ADS)
Li, Ao; Chen, Deyun; Sun, Guanglu; Lin, Kezheng
2016-10-01
Sparse representation (SR) and nonlocal technique (NLT) have shown great potential in low-level image processing. However, due to the degradation of the observed image, SR and NLT may not be accurate enough to obtain a faithful restoration results when they are used independently. To improve the performance, in this paper, a nonlocal supervised coding strategy-based NLT for image restoration is proposed. The novel method has three main contributions. First, to exploit the useful nonlocal patches, a nonnegative sparse representation is introduced, whose coefficients can be utilized as the supervised weights among patches. Second, a novel objective function is proposed, which integrated the supervised weights learning and the nonlocal sparse coding to guarantee a more promising solution. Finally, to make the minimization tractable and convergence, a numerical scheme based on iterative shrinkage thresholding is developed to solve the above underdetermined inverse problem. The extensive experiments validate the effectiveness of the proposed method.
Optimized Color Filter Arrays for Sparse Representation Based Demosaicking.
Li, Jia; Bai, Chenyan; Lin, Zhouchen; Yu, Jian
2017-03-08
Demosaicking is the problem of reconstructing a color image from the raw image captured by a digital color camera that covers its only imaging sensor with a color filter array (CFA). Sparse representation based demosaicking has been shown to produce superior reconstruction quality. However, almost all existing algorithms in this category use the CFAs which are not specifically optimized for the algorithms. In this paper, we consider optimally designing CFAs for sparse representation based demosaicking, where the dictionary is well-chosen. The fact that CFAs correspond to the projection matrices used in compressed sensing inspires us to optimize CFAs via minimizing the mutual coherence. This is more challenging than that for traditional projection matrices because CFAs have physical realizability constraints. However, most of the existing methods for minimizing the mutual coherence require that the projection matrices should be unconstrained, making them inapplicable for designing CFAs. We consider directly minimizing the mutual coherence with the CFA's physical realizability constraints as a generalized fractional programming problem, which needs to find sufficiently accurate solutions to a sequence of nonconvex nonsmooth minimization problems. We adapt the redistributed proximal bundle method to address this issue. Experiments on benchmark images testify to the superiority of the proposed method. In particular, we show that a simple sparse representation based demosaicking algorithm with our specifically optimized CFA can outperform LSSC [1]. To the best of our knowledge, it is the first sparse representation based demosaicking algorithm that beats LSSC in terms of CPSNR.
Feature selection and multi-kernel learning for sparse representation on a manifold.
Wang, Jim Jing-Yan; Bensmail, Halima; Gao, Xin
2014-03-01
Sparse representation has been widely studied as a part-based data representation method and applied in many scientific and engineering fields, such as bioinformatics and medical imaging. It seeks to represent a data sample as a sparse linear combination of some basic items in a dictionary. Gao et al. (2013) recently proposed Laplacian sparse coding by regularizing the sparse codes with an affinity graph. However, due to the noisy features and nonlinear distribution of the data samples, the affinity graph constructed directly from the original feature space is not necessarily a reliable reflection of the intrinsic manifold of the data samples. To overcome this problem, we integrate feature selection and multiple kernel learning into the sparse coding on the manifold. To this end, unified objectives are defined for feature selection, multiple kernel learning, sparse coding, and graph regularization. By optimizing the objective functions iteratively, we develop novel data representation algorithms with feature selection and multiple kernel learning respectively. Experimental results on two challenging tasks, N-linked glycosylation prediction and mammogram retrieval, demonstrate that the proposed algorithms outperform the traditional sparse coding methods.
Image denoising via group Sparse representation over learned dictionary
NASA Astrophysics Data System (ADS)
Cheng, Pan; Deng, Chengzhi; Wang, Shengqian; Zhang, Chunfeng
2013-10-01
Images are one of vital ways to get information for us. However, in the practical application, images are often subject to a variety of noise, so that solving the problem of image denoising becomes particularly important. The K-SVD algorithm can improve the denoising effect by sparse coding atoms instead of the traditional method of sparse coding dictionary. In order to further improve the effect of denoising, we propose to extended the K-SVD algorithm via group sparse representation. The key point of this method is dividing the sparse coefficients into groups, so that adjusts the correlation among the elements by controlling the size of the groups. This new approach can improve the local constraints between adjacent atoms, thereby it is very important to increase the correlation between the atoms. The experimental results show that our method has a better effect on image recovery, which is efficient to prevent the block effect and can get smoother images.
Group-sparse representation with dictionary learning for medical image denoising and fusion.
Li, Shutao; Yin, Haitao; Fang, Leyuan
2012-12-01
Recently, sparse representation has attracted a lot of interest in various areas. However, the standard sparse representation does not consider the intrinsic structure, i.e., the nonzero elements occur in clusters, called group sparsity. Furthermore, there is no dictionary learning method for group sparse representation considering the geometrical structure of space spanned by atoms. In this paper, we propose a novel dictionary learning method, called Dictionary Learning with Group Sparsity and Graph Regularization (DL-GSGR). First, the geometrical structure of atoms is modeled as the graph regularization. Then, combining group sparsity and graph regularization, the DL-GSGR is presented, which is solved by alternating the group sparse coding and dictionary updating. In this way, the group coherence of learned dictionary can be enforced small enough such that any signal can be group sparse coded effectively. Finally, group sparse representation with DL-GSGR is applied to 3-D medical image denoising and image fusion. Specifically, in 3-D medical image denoising, a 3-D processing mechanism (using the similarity among nearby slices) and temporal regularization (to perverse the correlations across nearby slices) are exploited. The experimental results on 3-D image denoising and image fusion demonstrate the superiority of our proposed denoising and fusion approaches.
Orthogonal Procrustes Analysis for Dictionary Learning in Sparse Linear Representation
Grossi, Giuliano; Lin, Jianyi
2017-01-01
In the sparse representation model, the design of overcomplete dictionaries plays a key role for the effectiveness and applicability in different domains. Recent research has produced several dictionary learning approaches, being proven that dictionaries learnt by data examples significantly outperform structured ones, e.g. wavelet transforms. In this context, learning consists in adapting the dictionary atoms to a set of training signals in order to promote a sparse representation that minimizes the reconstruction error. Finding the best fitting dictionary remains a very difficult task, leaving the question still open. A well-established heuristic method for tackling this problem is an iterative alternating scheme, adopted for instance in the well-known K-SVD algorithm. Essentially, it consists in repeating two stages; the former promotes sparse coding of the training set and the latter adapts the dictionary to reduce the error. In this paper we present R-SVD, a new method that, while maintaining the alternating scheme, adopts the Orthogonal Procrustes analysis to update the dictionary atoms suitably arranged into groups. Comparative experiments on synthetic data prove the effectiveness of R-SVD with respect to well known dictionary learning algorithms such as K-SVD, ILS-DLA and the online method OSDL. Moreover, experiments on natural data such as ECG compression, EEG sparse representation, and image modeling confirm R-SVD’s robustness and wide applicability. PMID:28103283
Multiscale Sparse Image Representation with Learned Dictionaries (PREPRINT)
2007-01-01
age processing, e.g., image denoising [5]. In [1] the K- SVD is proposed for learning a single-scale dic- tionary for sparse representation of image...performance we obtain. 2. THE SINGLE-SCALE K- SVD DENOISING ALGORITHM In this section, we briefly review the main ideas of the K- SVD frame- work for sparse...weighted average: x̂ = “ λI + X ij R T ijRij ”−1“ λy + X ij R T ijD̂α̂ij ” . (4) Fig. 1. The single-scale K- SVD -based image denoising algorithm. Fig
Sparse Representation for Prediction of HIV-1 Protease Drug Resistance.
Yu, Xiaxia; Weber, Irene T; Harrison, Robert W
2013-01-01
HIV rapidly evolves drug resistance in response to antiviral drugs used in AIDS therapy. Estimating the specific resistance of a given strain of HIV to individual drugs from sequence data has important benefits for both the therapy of individual patients and the development of novel drugs. We have developed an accurate classification method based on the sparse representation theory, and demonstrate that this method is highly effective with HIV-1 protease. The protease structure is represented using our newly proposed encoding method based on Delaunay triangulation, and combined with the mutated amino acid sequences of known drug-resistant strains to train a machine-learning algorithm both for classification and regression of drug-resistant mutations. An overall cross-validated classification accuracy of 97% is obtained when trained on a publically available data base of approximately 1.5×10(4) known sequences (Stanford HIV database http://hivdb.stanford.edu/cgi-bin/GenoPhenoDS.cgi). Resistance to four FDA approved drugs is computed and comparisons with other algorithms demonstrate that our method shows significant improvements in classification accuracy.
Learning Multiscale Sparse Representations for Image and Video Restoration (PREPRINT)
2007-07-01
video denoising [35]. In this paper, we extend the basic K- SVD work, providing a framework for learning multiscale and sparse image representation. In... denoising algorithm [1], the extensions to color image denoising , non-homogeneous noise, and inpainting [25], and the K- SVD for denoising videos [35]. Section...improvements to the original single-scale K- SVD . Section 6 presents some applications of the multiscale K- SVD , covering grayscale and color image denoising
Spatial, Temporal and Spectral Satellite Image Fusion via Sparse Representation
NASA Astrophysics Data System (ADS)
Song, Huihui
Remote sensing provides good measurements for monitoring and further analyzing the climate change, dynamics of ecosystem, and human activities in global or regional scales. Over the past two decades, the number of launched satellite sensors has been increasing with the development of aerospace technologies and the growing requirements on remote sensing data in a vast amount of application fields. However, a key technological challenge confronting these sensors is that they tradeoff between spatial resolution and other properties, including temporal resolution, spectral resolution, swath width, etc., due to the limitations of hardware technology and budget constraints. To increase the spatial resolution of data with other good properties, one possible cost-effective solution is to explore data integration methods that can fuse multi-resolution data from multiple sensors, thereby enhancing the application capabilities of available remote sensing data. In this thesis, we propose to fuse the spatial resolution with temporal resolution and spectral resolution, respectively, based on sparse representation theory. Taking the study case of Landsat ETM+ (with spatial resolution of 30m and temporal resolution of 16 days) and MODIS (with spatial resolution of 250m ~ 1km and daily temporal resolution) reflectance, we propose two spatial-temporal fusion methods to combine the fine spatial information of Landsat image and the daily temporal resolution of MODIS image. Motivated by that the images from these two sensors are comparable on corresponding bands, we propose to link their spatial information on available Landsat- MODIS image pair (captured on prior date) and then predict the Landsat image from the MODIS counterpart on prediction date. To well-learn the spatial details from the prior images, we use a redundant dictionary to extract the basic representation atoms for both Landsat and MODIS images based on sparse representation. Under the scenario of two prior Landsat
Vigilance detection based on sparse representation of EEG.
Yu, Hongbin; Lu, Hongtao; Ouyang, Tian; Liu, Hongjun; Lu, Bao-Liang
2010-01-01
Electroencephalogram (EEG) based vigilance detection of those people who engage in long time attention demanding tasks such as monotonous monitoring or driving is a key field in the research of brain-computer interface (BCI). However, robust detection of human vigilance from EEG is very difficult due to the low SNR nature of EEG signals. Recently, compressive sensing and sparse representation become successful tools in the fields of signal reconstruction and machine learning. In this paper, we propose to use the sparse representation of EEG to the vigilance detection problem. We first use continuous wavelet transform to extract the rhythm features of EEG data, and then employ the sparse representation method to the wavelet transform coefficients. We collect five subjects' EEG recordings in a simulation driving environment and apply the proposed method to detect the vigilance of the subjects. The experimental results show that the algorithm framework proposed in this paper can successfully estimate driver's vigilance with the average accuracy about 94.22 %. We also compare our algorithm framework with other vigilance estimation methods using different feature extraction and classifier selection approaches, the result shows that the proposed method has obvious advantages in the classification accuracy.
Qiu, Jun-Wei; Zao, John K; Wang, Peng-Hua; Chou, Yu-Hsiang
2010-01-01
A randomized search algorithm for sparse representations of EEG event-related potentials (ERPs) and their statistically independent components is presented. This algorithm combines greedy matching pursuit (MP) technique with covariance matrix adaptation evolution strategy (CMA-ES) to select small number of signal atoms from over-complete wavelet and chirplet dictionaries that offer best approximations of quasi-sparse ERP signals. During the search process, adaptive pruning of signal parameters was used to eliminate redundant or degenerative atoms. As a result, the CMA-ES/MP algorithm is capable of producing accurate efficient and consistent sparse representations of ERP signals and their ICA components. This paper explains the working principles of the algorithm and presents the preliminary results of its use.
Finger vein verification system based on sparse representation.
Xin, Yang; Liu, Zhi; Zhang, Haixia; Zhang, Hong
2012-09-01
Finger vein verification is a promising biometric pattern for personal identification in terms of security and convenience. The recognition performance of this technology heavily relies on the quality of finger vein images and on the recognition algorithm. To achieve efficient recognition performance, a special finger vein imaging device is developed, and a finger vein recognition method based on sparse representation is proposed. The motivation for the proposed method is that finger vein images exhibit a sparse property. In the proposed system, the regions of interest (ROIs) in the finger vein images are segmented and enhanced. Sparse representation and sparsity preserving projection on ROIs are performed to obtain the features. Finally, the features are measured for recognition. An equal error rate of 0.017% was achieved based on the finger vein image database, which contains images that were captured by using the near-IR imaging device that was developed in this study. The experimental results demonstrate that the proposed method is faster and more robust than previous methods.
Sparse representation-based spectral clustering for SAR image segmentation
NASA Astrophysics Data System (ADS)
Zhang, Xiangrong; Wei, Zhengli; Feng, Jie; Jiao, Licheng
2011-12-01
A new method, sparse representation based spectral clustering (SC) with Nyström method, is proposed for synthetic aperture radar (SAR) image segmentation. Different from the conventional SC, this proposed technique is developed by using the sparse coefficients which obtained by solving l1 minimization problem to construct the affinity matrix and the Nyström method is applied to alleviate the segmentation process. The advantage of our proposed method is that we do not need to select the scaling parameter in the Gaussian kernel function artificially. We apply the proposed method, k-means and the classic spectral clustering algorithm with Nyström method to SAR image segmentation. The results show that compared with the other two methods, the proposed method can obtain much better segmentation results.
Sparse representation based image interpolation with nonlocal autoregressive modeling.
Dong, Weisheng; Zhang, Lei; Lukac, Rastislav; Shi, Guangming
2013-04-01
Sparse representation is proven to be a promising approach to image super-resolution, where the low-resolution (LR) image is usually modeled as the down-sampled version of its high-resolution (HR) counterpart after blurring. When the blurring kernel is the Dirac delta function, i.e., the LR image is directly down-sampled from its HR counterpart without blurring, the super-resolution problem becomes an image interpolation problem. In such cases, however, the conventional sparse representation models (SRM) become less effective, because the data fidelity term fails to constrain the image local structures. In natural images, fortunately, many nonlocal similar patches to a given patch could provide nonlocal constraint to the local structure. In this paper, we incorporate the image nonlocal self-similarity into SRM for image interpolation. More specifically, a nonlocal autoregressive model (NARM) is proposed and taken as the data fidelity term in SRM. We show that the NARM-induced sampling matrix is less coherent with the representation dictionary, and consequently makes SRM more effective for image interpolation. Our extensive experimental results demonstrate that the proposed NARM-based image interpolation method can effectively reconstruct the edge structures and suppress the jaggy/ringing artifacts, achieving the best image interpolation results so far in terms of PSNR as well as perceptual quality metrics such as SSIM and FSIM.
3D ear identification based on sparse representation.
Zhang, Lin; Ding, Zhixuan; Li, Hongyu; Shen, Ying
2014-01-01
Biometrics based personal authentication is an effective way for automatically recognizing, with a high confidence, a person's identity. Recently, 3D ear shape has attracted tremendous interests in research field due to its richness of feature and ease of acquisition. However, the existing ICP (Iterative Closet Point)-based 3D ear matching methods prevalent in the literature are not quite efficient to cope with the one-to-many identification case. In this paper, we aim to fill this gap by proposing a novel effective fully automatic 3D ear identification system. We at first propose an accurate and efficient template-based ear detection method. By utilizing such a method, the extracted ear regions are represented in a common canonical coordinate system determined by the ear contour template, which facilitates much the following stages of feature extraction and classification. For each extracted 3D ear, a feature vector is generated as its representation by making use of a PCA-based local feature descriptor. At the stage of classification, we resort to the sparse representation based classification approach, which actually solves an l1-minimization problem. To the best of our knowledge, this is the first work introducing the sparse representation framework into the field of 3D ear identification. Extensive experiments conducted on a benchmark dataset corroborate the effectiveness and efficiency of the proposed approach. The associated Matlab source code and the evaluation results have been made publicly online available at http://sse.tongji.edu.cn/linzhang/ear/srcear/srcear.htm.
Sparse signal representation and its applications in ultrasonic NDE.
Zhang, Guang-Ming; Zhang, Cheng-Zhong; Harvey, David M
2012-03-01
Many sparse signal representation (SSR) algorithms have been developed in the past decade. The advantages of SSR such as compact representations and super resolution lead to the state of the art performance of SSR for processing ultrasonic non-destructive evaluation (NDE) signals. Choosing a suitable SSR algorithm and designing an appropriate overcomplete dictionary is a key for success. After a brief review of sparse signal representation methods and the design of overcomplete dictionaries, this paper addresses the recent accomplishments of SSR for processing ultrasonic NDE signals. The advantages and limitations of SSR algorithms and various overcomplete dictionaries widely-used in ultrasonic NDE applications are explored in depth. Their performance improvement compared to conventional signal processing methods in many applications such as ultrasonic flaw detection and noise suppression, echo separation and echo estimation, and ultrasonic imaging is investigated. The challenging issues met in practical ultrasonic NDE applications for example the design of a good dictionary are discussed. Representative experimental results are presented for demonstration.
An enhanced sparse representation strategy for signal classification
NASA Astrophysics Data System (ADS)
Zhou, Yin; Gao, Jinglun; Barner, Kenneth E.
2012-06-01
Sparse representation based classification (SRC) has achieved state-of-the-art results on face recognition. It is hence desired to extend its power to a broader range of classification tasks in pattern recognition. SRC first encodes a query sample as a linear combination of a few atoms from a predefined dictionary. It then identifies the label by evaluating which class results in the minimum reconstruction error. The effectiveness of SRC is limited by an important assumption that data points from different classes are not distributed along the same radius direction. Otherwise, this approach will lose their discrimination ability, even though data from different classes are actually well-separated in terms of Euclidean distance. This assumption is reasonable for face recognition as images of the same subject under different intensity levels are still considered to be of same-class. However, the assumption is not always satisfied when dealing with many other real-world data, e.g., the Iris dataset, where classes are stratified along the radius direction. In this paper, we propose a new coding strategy, called Nearest- Farthest Neighbors based SRC (NF-SRC), to effectively overcome the limitation within SRC. The dictionary is composed of both the Nearest Neighbors and the Farthest Neighbors. While the Nearest Neighbors are used to narrow the selection of candidate samples, the Farthest Neighbors are employed to make the dictionary more redundant. NF-SRC encodes each query signal in a greedy way similar to OMP. The proposed approach is evaluated over extensive experiments. The encouraging results demonstrate the feasibility of the proposed method.
Supervised Discriminative Group Sparse Representation for Mild Cognitive Impairment Diagnosis
Suk, Heung-Il; Wee, Chong-Yaw; Lee, Seong-Whan; Shen, Dinggang
2014-01-01
Research on an early detection of Mild Cognitive Impairment (MCI), a prodromal stage of Alzheimer’s Disease (AD), with resting-state functional Magnetic Resonance Imaging (rs-fMRI) has been of great interest for the last decade. Witnessed by recent studies, functional connectivity is a useful concept in extracting brain network features and finding biomarkers for brain disease diagnosis. However, it still remains challenging for the estimation of functional connectivity from rs-fMRI due to the inevitable high dimensional problem. In order to tackle this problem, we utilize a group sparse representation along with a structural equation model. Unlike the conventional group sparse representation method that does not explicitly consider class-label information, which can help enhance the diagnostic performance, in this paper, we propose a novel supervised discriminative group sparse representation method by penalizing a large within-class variance and a small between-class variance of connectivity coefficients. Thanks to the newly devised penalization terms, we can learn connectivity coefficients that are similar within the same class and distinct between classes, thus helping enhance the diagnostic accuracy. The proposed method also allows the learned common network structure to preserve the network specific and label-related characteristics. In our experiments on the rs-fMRI data of 37 subjects (12 MCI; 25 healthy normal control) with a cross-validation technique, we demonstrated the validity and effectiveness of the proposed method, showing the diagnostic accuracy of 89.19% and the sensitivity of 0.9167. PMID:25501275
A MRI-CT prostate registration using sparse representation technique
NASA Astrophysics Data System (ADS)
Yang, Xiaofeng; Jani, Ashesh B.; Rossi, Peter J.; Mao, Hui; Curran, Walter J.; Liu, Tian
2016-03-01
Purpose: To develop a new MRI-CT prostate registration using patch-based deformation prediction framework to improve MRI-guided prostate radiotherapy by incorporating multiparametric MRI into planning CT images. Methods: The main contribution is to estimate the deformation between prostate MRI and CT images in a patch-wise fashion by using the sparse representation technique. We assume that two image patches should follow the same deformation if their patch-wise appearance patterns are similar. Specifically, there are two stages in our proposed framework, i.e., the training stage and the application stage. In the training stage, each prostate MR images are carefully registered to the corresponding CT images and all training MR and CT images are carefully registered to a selected CT template. Thus, we obtain the dense deformation field for each training MR and CT image. In the application stage, for registering a new subject MR image with the same subject CT image, we first select a small number of key points at the distinctive regions of this subject CT image. Then, for each key point in the subject CT image, we extract the image patch, centered at the underlying key point. Then, we adaptively construct the coupled dictionary for the underlying point where each atom in the dictionary consists of image patches and the respective deformations obtained from training pair-wise MRI-CT images. Next, the subject image patch can be sparsely represented by a linear combination of training image patches in the dictionary, where we apply the same sparse coefficients to the respective deformations in the dictionary to predict the deformation for the subject MR image patch. After we repeat the same procedure for each subject CT key point, we use B-splines to interpolate a dense deformation field, which is used as the initialization to allow the registration algorithm estimating the remaining small segment of deformations from MRI to CT image
Inpainting with sparse linear combinations of exemplars
Wohlberg, Brendt
2008-01-01
We introduce a new exemplar-based inpainting algorithm based on representing the region to be inpainted as a sparse linear combination of blocks extracted from similar parts of the image being inpainted. This method is conceptually simple, being computed by functional minimization, and avoids the complexity of correctly ordering the filling in of missing regions of other exemplar-based methods. Initial performance comparisons on small inpainting regions indicate that this method provides similar or better performance than other recent methods.
Learning feature representations with a cost-relevant sparse autoencoder.
Längkvist, Martin; Loutfi, Amy
2015-02-01
There is an increasing interest in the machine learning community to automatically learn feature representations directly from the (unlabeled) data instead of using hand-designed features. The autoencoder is one method that can be used for this purpose. However, for data sets with a high degree of noise, a large amount of the representational capacity in the autoencoder is used to minimize the reconstruction error for these noisy inputs. This paper proposes a method that improves the feature learning process by focusing on the task relevant information in the data. This selective attention is achieved by weighting the reconstruction error and reducing the influence of noisy inputs during the learning process. The proposed model is trained on a number of publicly available image data sets and the test error rate is compared to a standard sparse autoencoder and other methods, such as the denoising autoencoder and contractive autoencoder.
Dictionary learning method for joint sparse representation-based image fusion
NASA Astrophysics Data System (ADS)
Zhang, Qiheng; Fu, Yuli; Li, Haifeng; Zou, Jian
2013-05-01
Recently, sparse representation (SR) and joint sparse representation (JSR) have attracted a lot of interest in image fusion. The SR models signals by sparse linear combinations of prototype signal atoms that make a dictionary. The JSR indicates that different signals from the various sensors of the same scene form an ensemble. These signals have a common sparse component and each individual signal owns an innovation sparse component. The JSR offers lower computational complexity compared with SR. First, for JSR-based image fusion, we give a new fusion rule. Then, motivated by the method of optimal directions (MOD), for JSR, we propose a novel dictionary learning method (MODJSR) whose dictionary updating procedure is derived by employing the JSR structure one time with singular value decomposition (SVD). MODJSR has lower complexity than the K-SVD algorithm which is often used in previous JSR-based fusion algorithms. To capture the image details more efficiently, we proposed the generalized JSR in which the signals ensemble depends on two dictionaries. MODJSR is extended to MODGJSR in this case. MODJSR/MODGJSR can simultaneously carry out dictionary learning, denoising, and fusion of noisy source images. Some experiments are given to demonstrate the validity of the MODJSR/MODGJSR for image fusion.
Blind deconvolution using an improved L0 sparse representation
NASA Astrophysics Data System (ADS)
Ye, Pengzhao; Feng, Huajun; Li, Qi; Xu, Zhihai; Chen, Yueting
2014-09-01
In this paper, we present a method for single image blind deconvolution. Many common forms of blind deconvolution methods need to previously generate a salient image, while the paper presents a novel L0 sparse expression to directly solve the ill-positioned problem. It has no need to filter the blurred image as a restoration step and can use the gradient information as a fidelity term during optimization. The key to blind deconvolution problem is to estimate an accurate kernel. First, based on L2 sparse expression using gradient operator as a prior, the kernel can be estimated roughly and efficiently in the frequency domain. We adopt the multi-scale scheme which can estimate blur kernel from coarser level to finer level. After the estimation of this level's kernel, L0 sparse representation is employed as the fidelity term during restoration. After derivation, L0 norm can be approximately converted to a sum term and L1 norm term which can be addressed by the Split-Bregman method. By using the estimated blur kernel and the TV deconvolution model, the final restoration image is obtained. Experimental results show that the proposed method is fast and can accurately reconstruct the kernel, especially when the blur is motion blur, defocus blur or the superposition of the two. The restored image is of higher quality than that of some of the art algorithms.
Magnetic resonance brain tissue segmentation based on sparse representations
NASA Astrophysics Data System (ADS)
Rueda, Andrea
2015-12-01
Segmentation or delineation of specific organs and structures in medical images is an important task in the clinical diagnosis and treatment, since it allows to characterize pathologies through imaging measures (biomarkers). In brain imaging, segmentation of main tissues or specific structures is challenging, due to the anatomic variability and complexity, and the presence of image artifacts (noise, intensity inhomogeneities, partial volume effect). In this paper, an automatic segmentation strategy is proposed, based on sparse representations and coupled dictionaries. Image intensity patterns are singly related to tissue labels at the level of small patches, gathering this information in coupled intensity/segmentation dictionaries. This dictionaries are used within a sparse representation framework to find the projection of a new intensity image onto the intensity dictionary, and the same projection can be used with the segmentation dictionary to estimate the corresponding segmentation. Preliminary results obtained with two publicly available datasets suggest that the proposal is capable of estimating adequate segmentations for gray matter (GM) and white matter (WM) tissues, with an average overlapping of 0:79 for GM and 0:71 for WM (with respect to original segmentations).
Robust Pedestrian Classification Based on Hierarchical Kernel Sparse Representation
Sun, Rui; Zhang, Guanghai; Yan, Xiaoxing; Gao, Jun
2016-01-01
Vision-based pedestrian detection has become an active topic in computer vision and autonomous vehicles. It aims at detecting pedestrians appearing ahead of the vehicle using a camera so that autonomous vehicles can assess the danger and take action. Due to varied illumination and appearance, complex background and occlusion pedestrian detection in outdoor environments is a difficult problem. In this paper, we propose a novel hierarchical feature extraction and weighted kernel sparse representation model for pedestrian classification. Initially, hierarchical feature extraction based on a CENTRIST descriptor is used to capture discriminative structures. A max pooling operation is used to enhance the invariance of varying appearance. Then, a kernel sparse representation model is proposed to fully exploit the discrimination information embedded in the hierarchical local features, and a Gaussian weight function as the measure to effectively handle the occlusion in pedestrian images. Extensive experiments are conducted on benchmark databases, including INRIA, Daimler, an artificially generated dataset and a real occluded dataset, demonstrating the more robust performance of the proposed method compared to state-of-the-art pedestrian classification methods. PMID:27537888
3D Ear Identification Based on Sparse Representation
Zhang, Lin; Ding, Zhixuan; Li, Hongyu; Shen, Ying
2014-01-01
Biometrics based personal authentication is an effective way for automatically recognizing, with a high confidence, a person’s identity. Recently, 3D ear shape has attracted tremendous interests in research field due to its richness of feature and ease of acquisition. However, the existing ICP (Iterative Closet Point)-based 3D ear matching methods prevalent in the literature are not quite efficient to cope with the one-to-many identification case. In this paper, we aim to fill this gap by proposing a novel effective fully automatic 3D ear identification system. We at first propose an accurate and efficient template-based ear detection method. By utilizing such a method, the extracted ear regions are represented in a common canonical coordinate system determined by the ear contour template, which facilitates much the following stages of feature extraction and classification. For each extracted 3D ear, a feature vector is generated as its representation by making use of a PCA-based local feature descriptor. At the stage of classification, we resort to the sparse representation based classification approach, which actually solves an l1-minimization problem. To the best of our knowledge, this is the first work introducing the sparse representation framework into the field of 3D ear identification. Extensive experiments conducted on a benchmark dataset corroborate the effectiveness and efficiency of the proposed approach. The associated Matlab source code and the evaluation results have been made publicly online available at http://sse.tongji.edu.cn/linzhang/ear/srcear/srcear.htm. PMID:24740247
Auditory Sketches: Very Sparse Representations of Sounds Are Still Recognizable
Isnard, Vincent; Taffou, Marine; Viaud-Delmon, Isabelle; Suied, Clara
2016-01-01
Sounds in our environment like voices, animal calls or musical instruments are easily recognized by human listeners. Understanding the key features underlying this robust sound recognition is an important question in auditory science. Here, we studied the recognition by human listeners of new classes of sounds: acoustic and auditory sketches, sounds that are severely impoverished but still recognizable. Starting from a time-frequency representation, a sketch is obtained by keeping only sparse elements of the original signal, here, by means of a simple peak-picking algorithm. Two time-frequency representations were compared: a biologically grounded one, the auditory spectrogram, which simulates peripheral auditory filtering, and a simple acoustic spectrogram, based on a Fourier transform. Three degrees of sparsity were also investigated. Listeners were asked to recognize the category to which a sketch sound belongs: singing voices, bird calls, musical instruments, and vehicle engine noises. Results showed that, with the exception of voice sounds, very sparse representations of sounds (10 features, or energy peaks, per second) could be recognized above chance. No clear differences could be observed between the acoustic and the auditory sketches. For the voice sounds, however, a completely different pattern of results emerged, with at-chance or even below-chance recognition performances, suggesting that the important features of the voice, whatever they are, were removed by the sketch process. Overall, these perceptual results were well correlated with a model of auditory distances, based on spectro-temporal excitation patterns (STEPs). This study confirms the potential of these new classes of sounds, acoustic and auditory sketches, to study sound recognition. PMID:26950589
Image denoising via sparse and redundant representations over learned dictionaries.
Elad, Michael; Aharon, Michal
2006-12-01
We address the image denoising problem, where zero-mean white and homogeneous Gaussian additive noise is to be removed from a given image. The approach taken is based on sparse and redundant representations over trained dictionaries. Using the K-SVD algorithm, we obtain a dictionary that describes the image content effectively. Two training options are considered: using the corrupted image itself, or training on a corpus of high-quality image database. Since the K-SVD is limited in handling small image patches, we extend its deployment to arbitrary image sizes by defining a global image prior that forces sparsity over patches in every location in the image. We show how such Bayesian treatment leads to a simple and effective denoising algorithm. This leads to a state-of-the-art denoising performance, equivalent and sometimes surpassing recently published leading alternative denoising methods.
Learned dictionaries for sparse image representation: properties and results
NASA Astrophysics Data System (ADS)
Skretting, Karl; Engan, Kjersti
2011-09-01
Sparse representation of images using learned dictionaries have been shown to work well for applications like image denoising, impainting, image compression, etc. In this paper dictionary properties are reviewed from a theoretical approach, and experimental results for learned dictionaries are presented. The main dictionary properties are the upper and lower frame (dictionary) bounds, and (mutual) coherence properties based on the angle between dictionary atoms. Both l0 sparsity and l1 sparsity are considered by using a matching pursuit method, order recursive matching Pursuit (ORMP), and a basis pursuit method, i.e. LARS or Lasso. For dictionary learning the following methods are considered: Iterative least squares (ILS-DLA or MOD), recursive least squares (RLS-DLA), K-SVD and online dictionary learning (ODL). Finally, it is shown how these properties relate to an image compression example.
Face recognition under variable illumination via sparse representation of patches
NASA Astrophysics Data System (ADS)
Fan, Shouke; Liu, Rui; Feng, Weiguo; Zhu, Ming
2013-10-01
The objective of this work is to recognize faces under variations in illumination. Previous works have indicated that the variations in illumination can dramatically reduce the performance of face recognition. To this end - ;an efficient method for face recognition which is robust under variable illumination is proposed in this paper. First of all, a discrete cosine transform(DCT) in the logarithm domain is employed to preprocess the images, removing the illumination variations by discarding an appropriate number of low-frequency DCT coefficients. Then, a face image is partitioned into several patches, and we classify the patches using Sparse Representation-based Classification, respectively. At last, the identity of a test image can be determined by the classification results of its patches. Experimental results on the Yale B database and the CMU PIE database show that excellent recognition rates can be achieved by the proposed method.
Sparse Representation of Electrodermal Activity With Knowledge-Driven Dictionaries
Tsiartas, Andreas; Stein, Leah I.; Cermak, Sharon A.; Narayanan, Shrikanth S.
2015-01-01
Biometric sensors and portable devices are being increasingly embedded into our everyday life, creating the need for robust physiological models that efficiently represent, analyze, and interpret the acquired signals. We propose a knowledge-driven method to represent electrodermal activity (EDA), a psychophysiological signal linked to stress, affect, and cognitive processing. We build EDA-specific dictionaries that accurately model both the slow varying tonic part and the signal fluctuations, called skin conductance responses (SCR), and use greedy sparse representation techniques to decompose the signal into a small number of atoms from the dictionary. Quantitative evaluation of our method considers signal reconstruction, compression rate, and information retrieval measures, that capture the ability of the model to incorporate the main signal characteristics, such as SCR occurrences. Compared to previous studies fitting a predetermined structure to the signal, results indicate that our approach provides benefits across all aforementioned criteria. This paper demonstrates the ability of appropriate dictionaries along with sparse decomposition methods to reliably represent EDA signals and provides a foundation for automatic measurement of SCR characteristics and the extraction of meaningful EDA features. PMID:25494494
Pedestrian detection from thermal images: A sparse representation based approach
NASA Astrophysics Data System (ADS)
Qi, Bin; John, Vijay; Liu, Zheng; Mita, Seiichi
2016-05-01
Pedestrian detection, a key technology in computer vision, plays a paramount role in the applications of advanced driver assistant systems (ADASs) and autonomous vehicles. The objective of pedestrian detection is to identify and locate people in a dynamic environment so that accidents can be avoided. With significant variations introduced by illumination, occlusion, articulated pose, and complex background, pedestrian detection is a challenging task for visual perception. Different from visible images, thermal images are captured and presented with intensity maps based objects' emissivity, and thus have an enhanced spectral range to make human beings perceptible from the cool background. In this study, a sparse representation based approach is proposed for pedestrian detection from thermal images. We first adopted the histogram of sparse code to represent image features and then detect pedestrian with the extracted features in an unimodal and a multimodal framework respectively. In the unimodal framework, two types of dictionaries, i.e. joint dictionary and individual dictionary, are built by learning from prepared training samples. In the multimodal framework, a weighted fusion scheme is proposed to further highlight the contributions from features with higher separability. To validate the proposed approach, experiments were conducted to compare with three widely used features: Haar wavelets (HWs), histogram of oriented gradients (HOG), and histogram of phase congruency (HPC) as well as two classification methods, i.e. AdaBoost and support vector machine (SVM). Experimental results on a publicly available data set demonstrate the superiority of the proposed approach.
Sinogram denoising via simultaneous sparse representation in learned dictionaries.
Karimi, Davood; Ward, Rabab K
2016-05-07
Reducing the radiation dose in computed tomography (CT) is highly desirable but it leads to excessive noise in the projection measurements. This can significantly reduce the diagnostic value of the reconstructed images. Removing the noise in the projection measurements is, therefore, essential for reconstructing high-quality images, especially in low-dose CT. In recent years, two new classes of patch-based denoising algorithms proved superior to other methods in various denoising applications. The first class is based on sparse representation of image patches in a learned dictionary. The second class is based on the non-local means method. Here, the image is searched for similar patches and the patches are processed together to find their denoised estimates. In this paper, we propose a novel denoising algorithm for cone-beam CT projections. The proposed method has similarities to both these algorithmic classes but is more effective and much faster. In order to exploit both the correlation between neighboring pixels within a projection and the correlation between pixels in neighboring projections, the proposed algorithm stacks noisy cone-beam projections together to form a 3D image and extracts small overlapping 3D blocks from this 3D image for processing. We propose a fast algorithm for clustering all extracted blocks. The central assumption in the proposed algorithm is that all blocks in a cluster have a joint-sparse representation in a well-designed dictionary. We describe algorithms for learning such a dictionary and for denoising a set of projections using this dictionary. We apply the proposed algorithm on simulated and real data and compare it with three other algorithms. Our results show that the proposed algorithm outperforms some of the best denoising algorithms, while also being much faster.
Sinogram denoising via simultaneous sparse representation in learned dictionaries
NASA Astrophysics Data System (ADS)
Karimi, Davood; Ward, Rabab K.
2016-05-01
Reducing the radiation dose in computed tomography (CT) is highly desirable but it leads to excessive noise in the projection measurements. This can significantly reduce the diagnostic value of the reconstructed images. Removing the noise in the projection measurements is, therefore, essential for reconstructing high-quality images, especially in low-dose CT. In recent years, two new classes of patch-based denoising algorithms proved superior to other methods in various denoising applications. The first class is based on sparse representation of image patches in a learned dictionary. The second class is based on the non-local means method. Here, the image is searched for similar patches and the patches are processed together to find their denoised estimates. In this paper, we propose a novel denoising algorithm for cone-beam CT projections. The proposed method has similarities to both these algorithmic classes but is more effective and much faster. In order to exploit both the correlation between neighboring pixels within a projection and the correlation between pixels in neighboring projections, the proposed algorithm stacks noisy cone-beam projections together to form a 3D image and extracts small overlapping 3D blocks from this 3D image for processing. We propose a fast algorithm for clustering all extracted blocks. The central assumption in the proposed algorithm is that all blocks in a cluster have a joint-sparse representation in a well-designed dictionary. We describe algorithms for learning such a dictionary and for denoising a set of projections using this dictionary. We apply the proposed algorithm on simulated and real data and compare it with three other algorithms. Our results show that the proposed algorithm outperforms some of the best denoising algorithms, while also being much faster.
Flutter signal extracting technique based on FOG and self-adaptive sparse representation algorithm
NASA Astrophysics Data System (ADS)
Lei, Jian; Meng, Xiangtao; Xiang, Zheng
2016-10-01
Due to various moving parts inside, when a spacecraft runs in orbits, its structure could get a minor angular vibration, which results in vague image formation of space camera. Thus, image compensation technique is required to eliminate or alleviate the effect of movement on image formation and it is necessary to realize precise measuring of flutter angle. Due to the advantages such as high sensitivity, broad bandwidth, simple structure and no inner mechanical moving parts, FOG (fiber optical gyro) is adopted in this study to measure minor angular vibration. Then, movement leading to image degeneration is achieved by calculation. The idea of the movement information extracting algorithm based on self-adaptive sparse representation is to use arctangent function approximating L0 norm to construct unconstrained noisy-signal-aimed sparse reconstruction model and then solve the model by a method based on steepest descent algorithm and BFGS algorithm to estimate sparse signal. Then taking the advantage of the principle of random noises not able to be represented by linear combination of elements, useful signal and random noised are separated effectively. Because the main interference of minor angular vibration to image formation of space camera is random noises, sparse representation algorithm could extract useful information to a large extent and acts as a fitting pre-process method of image restoration. The self-adaptive sparse representation algorithm presented in this paper is used to process the measured minor-angle-vibration signal of FOG used by some certain spacecraft. By component analysis of the processing results, we can find out that the algorithm could extract micro angular vibration signal of FOG precisely and effectively, and can achieve the precision degree of 0.1".
Classification of Clouds in Satellite Imagery Using Adaptive Fuzzy Sparse Representation.
Jin, Wei; Gong, Fei; Zeng, Xingbin; Fu, Randi
2016-12-16
Automatic cloud detection and classification using satellite cloud imagery have various meteorological applications such as weather forecasting and climate monitoring. Cloud pattern analysis is one of the research hotspots recently. Since satellites sense the clouds remotely from space, and different cloud types often overlap and convert into each other, there must be some fuzziness and uncertainty in satellite cloud imagery. Satellite observation is susceptible to noises, while traditional cloud classification methods are sensitive to noises and outliers; it is hard for traditional cloud classification methods to achieve reliable results. To deal with these problems, a satellite cloud classification method using adaptive fuzzy sparse representation-based classification (AFSRC) is proposed. Firstly, by defining adaptive parameters related to attenuation rate and critical membership, an improved fuzzy membership is introduced to accommodate the fuzziness and uncertainty of satellite cloud imagery; secondly, by effective combination of the improved fuzzy membership function and sparse representation-based classification (SRC), atoms in training dictionary are optimized; finally, an adaptive fuzzy sparse representation classifier for cloud classification is proposed. Experiment results on FY-2G satellite cloud image show that, the proposed method not only improves the accuracy of cloud classification, but also has strong stability and adaptability with high computational efficiency.
Classification of Clouds in Satellite Imagery Using Adaptive Fuzzy Sparse Representation
Jin, Wei; Gong, Fei; Zeng, Xingbin; Fu, Randi
2016-01-01
Automatic cloud detection and classification using satellite cloud imagery have various meteorological applications such as weather forecasting and climate monitoring. Cloud pattern analysis is one of the research hotspots recently. Since satellites sense the clouds remotely from space, and different cloud types often overlap and convert into each other, there must be some fuzziness and uncertainty in satellite cloud imagery. Satellite observation is susceptible to noises, while traditional cloud classification methods are sensitive to noises and outliers; it is hard for traditional cloud classification methods to achieve reliable results. To deal with these problems, a satellite cloud classification method using adaptive fuzzy sparse representation-based classification (AFSRC) is proposed. Firstly, by defining adaptive parameters related to attenuation rate and critical membership, an improved fuzzy membership is introduced to accommodate the fuzziness and uncertainty of satellite cloud imagery; secondly, by effective combination of the improved fuzzy membership function and sparse representation-based classification (SRC), atoms in training dictionary are optimized; finally, an adaptive fuzzy sparse representation classifier for cloud classification is proposed. Experiment results on FY-2G satellite cloud image show that, the proposed method not only improves the accuracy of cloud classification, but also has strong stability and adaptability with high computational efficiency. PMID:27999261
Locality-preserving sparse representation-based classification in hyperspectral imagery
NASA Astrophysics Data System (ADS)
Gao, Lianru; Yu, Haoyang; Zhang, Bing; Li, Qingting
2016-10-01
This paper proposes to combine locality-preserving projections (LPP) and sparse representation (SR) for hyperspectral image classification. The LPP is first used to reduce the dimensionality of all the training and testing data by finding the optimal linear approximations to the eigenfunctions of the Laplace Beltrami operator on the manifold, where the high-dimensional data lies. Then, SR codes the projected testing pixels as sparse linear combinations of all the training samples to classify the testing pixels by evaluating which class leads to the minimum approximation error. The integration of LPP and SR represents an innovative contribution to the literature. The proposed approach, called locality-preserving SR-based classification, addresses the imbalance between high dimensionality of hyperspectral data and the limited number of training samples. Experimental results on three real hyperspectral data sets demonstrate that the proposed approach outperforms the original counterpart, i.e., SR-based classification.
Enhanced low-rank representation via sparse manifold adaption for semi-supervised learning.
Peng, Yong; Lu, Bao-Liang; Wang, Suhang
2015-05-01
Constructing an informative and discriminative graph plays an important role in various pattern recognition tasks such as clustering and classification. Among the existing graph-based learning models, low-rank representation (LRR) is a very competitive one, which has been extensively employed in spectral clustering and semi-supervised learning (SSL). In SSL, the graph is composed of both labeled and unlabeled samples, where the edge weights are calculated based on the LRR coefficients. However, most of existing LRR related approaches fail to consider the geometrical structure of data, which has been shown beneficial for discriminative tasks. In this paper, we propose an enhanced LRR via sparse manifold adaption, termed manifold low-rank representation (MLRR), to learn low-rank data representation. MLRR can explicitly take the data local manifold structure into consideration, which can be identified by the geometric sparsity idea; specifically, the local tangent space of each data point was sought by solving a sparse representation objective. Therefore, the graph to depict the relationship of data points can be built once the manifold information is obtained. We incorporate a regularizer into LRR to make the learned coefficients preserve the geometric constraints revealed in the data space. As a result, MLRR combines both the global information emphasized by low-rank property and the local information emphasized by the identified manifold structure. Extensive experimental results on semi-supervised classification tasks demonstrate that MLRR is an excellent method in comparison with several state-of-the-art graph construction approaches.
Face sketch synthesis via sparse representation-based greedy search.
Shengchuan Zhang; Xinbo Gao; Nannan Wang; Jie Li; Mingjin Zhang
2015-08-01
Face sketch synthesis has wide applications in digital entertainment and law enforcement. Although there is much research on face sketch synthesis, most existing algorithms cannot handle some nonfacial factors, such as hair style, hairpins, and glasses if these factors are excluded in the training set. In addition, previous methods only work on well controlled conditions and fail on images with different backgrounds and sizes as the training set. To this end, this paper presents a novel method that combines both the similarity between different image patches and prior knowledge to synthesize face sketches. Given training photo-sketch pairs, the proposed method learns a photo patch feature dictionary from the training photo patches and replaces the photo patches with their sparse coefficients during the searching process. For a test photo patch, we first obtain its sparse coefficient via the learnt dictionary and then search its nearest neighbors (candidate patches) in the whole training photo patches with sparse coefficients. After purifying the nearest neighbors with prior knowledge, the final sketch corresponding to the test photo can be obtained by Bayesian inference. The contributions of this paper are as follows: 1) we relax the nearest neighbor search area from local region to the whole image without too much time consuming and 2) our method can produce nonfacial factors that are not contained in the training set and is robust against image backgrounds and can even ignore the alignment and image size aspects of test photos. Our experimental results show that the proposed method outperforms several state-of-the-arts in terms of perceptual and objective metrics.
Learning Sparse Feature Representations using Probabilistic Quadtrees and Deep Belief Nets
2015-04-24
Feature Representations usingProbabilistic Quadtrees and Deep Belief Nets Learning sparse feature representations is a useful instru- ment for solving an...novel framework for the classifi cation of handwritten digits that learns sparse representations using probabilistic quadtrees and Deep Belief Nets...S) AND ADDRESS (ES) U.S. Army Research Office P.O. Box 12211 Research Triangle Park, NC 27709-2211 Deep Belief Networks; MNIST REPORT
Pavement crack characteristic detection based on sparse representation
NASA Astrophysics Data System (ADS)
Sun, Xiaoming; Huang, Jianping; Liu, Wanyu; Xu, Mantao
2012-12-01
Pavement crack detection plays an important role in pavement maintaining and management. The three-dimensional (3D) pavement crack detection technique based on laser is a recent trend due to its ability of discriminating dark areas, which are not caused by pavement distress such as tire marks, oil spills and shadows. In the field of 3D pavement crack detection, the most important thing is the accurate extraction of cracks in individual pavement profile without destroying pavement profile. So after analyzing the pavement profile signal characteristics and the changeability of pavement crack characteristics, a new method based on the sparse representation is developed to decompose pavement profile signal into a summation of the mainly pavement profile and cracks. Based on the characteristics of the pavement profile signal and crack, the mixed dictionary is constructed with an over-complete exponential function and an over-complete trapezoidal membership function, and the signal is separated by learning in this mixed dictionary with a matching pursuit algorithm. Some experiments were conducted and promising results were obtained, showing that we can detect the pavement crack efficiently and achieve a good separation of crack from pavement profile without destroying pavement profile.
Extreme learning machine and adaptive sparse representation for image classification.
Cao, Jiuwen; Zhang, Kai; Luo, Minxia; Yin, Chun; Lai, Xiaoping
2016-09-01
Recent research has shown the speed advantage of extreme learning machine (ELM) and the accuracy advantage of sparse representation classification (SRC) in the area of image classification. Those two methods, however, have their respective drawbacks, e.g., in general, ELM is known to be less robust to noise while SRC is known to be time-consuming. Consequently, ELM and SRC complement each other in computational complexity and classification accuracy. In order to unify such mutual complementarity and thus further enhance the classification performance, we propose an efficient hybrid classifier to exploit the advantages of ELM and SRC in this paper. More precisely, the proposed classifier consists of two stages: first, an ELM network is trained by supervised learning. Second, a discriminative criterion about the reliability of the obtained ELM output is adopted to decide whether the query image can be correctly classified or not. If the output is reliable, the classification will be performed by ELM; otherwise the query image will be fed to SRC. Meanwhile, in the stage of SRC, a sub-dictionary that is adaptive to the query image instead of the entire dictionary is extracted via the ELM output. The computational burden of SRC thus can be reduced. Extensive experiments on handwritten digit classification, landmark recognition and face recognition demonstrate that the proposed hybrid classifier outperforms ELM and SRC in classification accuracy with outstanding computational efficiency.
Image sequence denoising via sparse and redundant representations.
Protter, Matan; Elad, Michael
2009-01-01
In this paper, we consider denoising of image sequences that are corrupted by zero-mean additive white Gaussian noise. Relative to single image denoising techniques, denoising of sequences aims to also utilize the temporal dimension. This assists in getting both faster algorithms and better output quality. This paper focuses on utilizing sparse and redundant representations for image sequence denoising, extending the work reported in. In the single image setting, the K-SVD algorithm is used to train a sparsifying dictionary for the corrupted image. This paper generalizes the above algorithm by offering several extensions: i) the atoms used are 3-D; ii) the dictionary is propagated from one frame to the next, reducing the number of required iterations; and iii) averaging is done on patches in both spatial and temporal neighboring locations. These modifications lead to substantial benefits in complexity and denoising performance, compared to simply running the single image algorithm sequentially. The algorithm's performance is experimentally compared to several state-of-the-art algorithms, demonstrating comparable or favorable results.
Sparse representation utilizing tight frame for phase retrieval
NASA Astrophysics Data System (ADS)
Shi, Baoshun; Lian, Qiusheng; Chen, Shuzhen
2015-12-01
We treat the phase retrieval (PR) problem of reconstructing the interest signal from its Fourier magnitude. Since the Fourier phase information is lost, the problem is ill-posed. Several techniques have been used to address this problem by utilizing various priors such as non-negative, support, and Fourier magnitude constraints. Recent methods exploiting sparsity are developed to improve the reconstruction quality. However, the previous algorithms of utilizing sparsity prior suffer from either the low reconstruction quality at low oversampled factors or being sensitive to noise. To address these issues, we propose a framework that exploits sparsity of the signal in the translation invariant Haar pyramid (TIHP) tight frame. Based on this sparsity prior, we formulate the sparse representation regularization term and incorporate it into the PR optimization problem. We propose the alternating iterative algorithm for solving the corresponding non-convex problem by dividing the problem into several subproblems. We give the optimal solution to each subproblem, and experimental simulations under the noise-free and noisy scenario indicate that our proposed algorithm can obtain a better reconstruction quality compared to the conventional alternative projection methods, even outperform the recent sparsity-based algorithms in terms of reconstruction quality.
Learning sparse discriminative representations for land cover classification in the Arctic
NASA Astrophysics Data System (ADS)
Moody, Daniela I.; Brumby, Steven P.; Rowland, Joel C.; Gangodagamage, Chandana
2012-10-01
Neuroscience-inspired machine vision algorithms are of current interest in the areas of detection and monitoring of climate change impacts, and general Land Use/Land Cover classification using satellite image data. We describe an approach for automatic classification of land cover in multispectral satellite imagery of the Arctic using sparse representations over learned dictionaries. We demonstrate our method using DigitalGlobe Worldview-2 8-band visible/near infrared high spatial resolution imagery of the MacKenzie River basin. We use an on-line batch Hebbian learning rule to build spectral-textural dictionaries that are adapted to this multispectral data. We learn our dictionaries from millions of overlapping image patches and then use a pursuit search to generate sparse classification features. We explore unsupervised clustering in the sparse representation space to produce land-cover category labels. This approach combines spectral and spatial textural characteristics to detect geologic, vegetative, and hydrologic features. We compare our technique to standard remote sensing algorithms. Our results suggest that neuroscience-based models are a promising approach to practical pattern recognition problems in remote sensing, even for datasets using spectral bands not found in natural visual systems.
Robust ear recognition via nonnegative sparse representation of Gabor orientation information.
Zhang, Baoqing; Mu, Zhichun; Zeng, Hui; Luo, Shuang
2014-01-01
Orientation information is critical to the accuracy of ear recognition systems. In this paper, a new feature extraction approach is investigated for ear recognition by using orientation information of Gabor wavelets. The proposed Gabor orientation feature can not only avoid too much redundancy in conventional Gabor feature but also tend to extract more precise orientation information of the ear shape contours. Then, Gabor orientation feature based nonnegative sparse representation classification (Gabor orientation + NSRC) is proposed for ear recognition. Compared with SRC in which the sparse coding coefficients can be negative, the nonnegativity of NSRC conforms to the intuitive notion of combining parts to form a whole and therefore is more consistent with the biological modeling of visual data. Additionally, the use of Gabor orientation features increases the discriminative power of NSRC. Extensive experimental results show that the proposed Gabor orientation feature based nonnegative sparse representation classification paradigm achieves much better recognition performance and is found to be more robust to challenging problems such as pose changes, illumination variations, and ear partial occlusion in real-world applications.
Temporal Super Resolution Enhancement of Echocardiographic Images Based on Sparse Representation.
Gifani, Parisa; Behnam, Hamid; Haddadi, Farzan; Sani, Zahra Alizadeh; Shojaeifard, Maryam
2016-01-01
A challenging issue for echocardiographic image interpretation is the accurate analysis of small transient motions of myocardium and valves during real-time visualization. A higher frame rate video may reduce this difficulty, and temporal super resolution (TSR) is useful for illustrating the fast-moving structures. In this paper, we introduce a novel framework that optimizes TSR enhancement of echocardiographic images by utilizing temporal information and sparse representation. The goal of this method is to increase the frame rate of echocardiographic videos, and therefore enable more accurate analyses of moving structures. For the proposed method, we first derived temporal information by extracting intensity variation time curves (IVTCs) assessed for each pixel. We then designed both low-resolution and high-resolution overcomplete dictionaries based on prior knowledge of the temporal signals and a set of prespecified known functions. The IVTCs can then be described as linear combinations of a few prototype atoms in the low-resolution dictionary. We used the Bayesian compressive sensing (BCS) sparse recovery algorithm to find the sparse coefficients of the signals. We extracted the sparse coefficients and the corresponding active atoms in the low-resolution dictionary to construct new sparse coefficients corresponding to the high-resolution dictionary. Using the estimated atoms and the high-resolution dictionary, a new IVTC with more samples was constructed. Finally, by placing the new IVTC signals in the original IVTC positions, we were able to reconstruct the original echocardiography video with more frames. The proposed method does not require training of low-resolution and high-resolution dictionaries, nor does it require motion estimation; it does not blur fast-moving objects, and does not have blocking artifacts.
Sparse representation based multi-threshold segmentation for hyperspectral target detection
NASA Astrophysics Data System (ADS)
Feng, Wei-yi; Chen, Qian; Miao, Zhuang; He, Wei-ji; Gu, Guo-hua; Zhuang, Jia-yan
2013-08-01
A sparse representation based multi-threshold segmentation (SRMTS) algorithm for target detection in hyperspectral images is proposed. Benefiting from the sparse representation, the high-dimensional spectral data can be characterized into a series of sparse feature vectors which has only a few nonzero coefficients. Through setting an appropriate threshold, the noise removed sparse spectral vectors are divided into two subspaces in the sparse domain consistent with the sample spectrum to separate the target from the background. Then a correlation and a vector 1-norm are calculated respectively in the subspaces. The sparse characteristic of the target is used to ext ract the target with a multi -threshold method. Unlike the conventional hyperspectral dimensionality reduction methods used in target detection algorithms, like Principal Components Analysis (PCA) and Maximum Noise Fraction (MNF), this algorithm maintains the spectral characteristics while removing the noise due to the sparse representation. In the experiments, an orthogonal wavelet sparse base is used to sparse the spectral information and a best contraction threshold to remove the hyperspectral image noise according to the noise estimation of the test images. Compared with co mmon algorithms, such as Adaptive Cosine Estimator (ACE), Constrained Energy Minimizat ion (CEM) and the noise removed MNF-CEM algorithm, the proposed algorithm demonstrates higher detection rates and robustness via the ROC curves.
Weighted sparse representation for human ear recognition based on local descriptor
NASA Astrophysics Data System (ADS)
Mawloud, Guermoui; Djamel, Melaab
2016-01-01
A two-stage ear recognition framework is presented where two local descriptors and a sparse representation algorithm are combined. In a first stage, the algorithm proceeds by deducing a subset of the closest training neighbors to the test ear sample. The selection is based on the K-nearest neighbors classifier in the pattern of oriented edge magnitude feature space. In a second phase, the co-occurrence of adjacent local binary pattern features are extracted from the preselected subset and combined to form a dictionary. Afterward, sparse representation classifier is employed on the developed dictionary in order to infer the closest element to the test sample. Thus, by splitting up the ear image into a number of segments and applying the described recognition routine on each of them, the algorithm finalizes by attributing a final class label based on majority voting over the individual labels pointed out by each segment. Experimental results demonstrate the effectiveness as well as the robustness of the proposed scheme over leading state-of-the-art methods. Especially when the ear image is occluded, the proposed algorithm exhibits a great robustness and reaches the recognition performances outlined in the state of the art.
A Max-Margin Perspective on Sparse Representation-Based Classification
2013-11-30
ABSTRACT 16. SECURITY CLASSIFICATION OF: 1. REPORT DATE (DD-MM-YYYY) 4. TITLE AND SUBTITLE 13. SUPPLEMENTARY NOTES 12. DISTRIBUTION AVAILIBILITY...Perspective on Sparse Representation-Based Classification Sparse Representation-based Classification (SRC) is a powerful tool in distinguishing signal...a reconstructive perspective, which neither offer- s any guarantee on its classification performance nor pro- The views, opinions and/or findings
Micro-Expression Recognition based on 2D Gabor Filter and Sparse Representation
NASA Astrophysics Data System (ADS)
Zheng, Hao
2017-01-01
Micro-expression recognition is always a challenging problem for its quick facial expression. This paper proposed a novel method named 2D Gabor filter and Sparse Representation (2DGSR) to deal with the recognition of micro-expression. In our method, 2D Gabor filter is used for enhancing the robustness of the variations due to increasing the discrimination power. While the sparse representation is applied to deal with the subtlety, and cast recognition as a sparse approximation problem. We compare our method to other popular methods in three spontaneous micro-expression recognition databases. The results show that our method has more excellent performance than other methods.
Adaptive Nonlocal Sparse Representation for Dual-Camera Compressive Hyperspectral Imaging.
Wang, Lizhi; Xiong, Zhiwei; Shi, Guangming; Wu, Feng; Zeng, Wenjun
2016-10-25
Leveraging the compressive sensing (CS) theory, coded aperture snapshot spectral imaging (CASSI) provides an efficient solution to recover 3D hyperspectral data from a 2D measurement. The dual-camera design of CASSI, by adding an uncoded panchromatic measurement, enhances the reconstruction fidelity while maintaining the snapshot advantage. In this paper, we propose an adaptive nonlocal sparse representation (ANSR) model to boost the performance of dualcamera compressive hyperspectral imaging (DCCHI). Specifically, the CS reconstruction problem is formulated as a 3D cube based sparse representation to make full use of the nonlocal similarity in both the spatial and spectral domains. Our key observation is that, the panchromatic image, besides playing the role of direct measurement, can be further exploited to help the nonlocal similarity estimation. Therefore, we design a joint similarity metric by adaptively combining the internal similarity within the reconstructed hyperspectral image and the external similarity within the panchromatic image. In this way, the fidelity of CS reconstruction is greatly enhanced. Both simulation and hardware experimental results show significant improvement of the proposed method over the state-of-the-art.
Hao, Dong-mei; Zhou, Ya-nan; Wang, Yu; Zhang, Song; Yang, Yi-min; Lin, Ling; Li, Gang; Wang, Xiu-li
2015-01-01
The present paper proposed a new nondestructive method based on visible/near infrared spectrum (Vis/NIRS) and sparse representation to rapidly and accurately discriminate between raw meat and water-injected meat. Water-injected meat model was built by injecting water into non-destructed meat samples comprising pigskin, fat layer and muscle layer. Vis/NIRS data were collected from raw meat and six scales of water-injected meat with spectrometers. To reduce the redundant information in the spectrum and improve the difference between the samples,. some preprocessing steps were performed for the spectral data, including light modulation and normalization. Effective spectral bands were extracted from the preprocessed spectral data. The meat samples were classified as raw meat and water-injected meat, and further, water-injected meat with different water injection rates. All the training samples were used to compose an atom dictionary, and test samples were represented by the sparsest linear combinations of these atoms via l1-minimization. Projection errors of test samples with respect to each category were calculated. A test sample was classified to the category with the minimum projection error, and leave-one-out cross-validation was conducted. The recognition performance from sparse representation was compared with that from support vector machine (SVM).. Experimental results showed that the overall recognition accuracy of sparse representation for raw meat and water-injected meat was more than 90%, which was higher than that of SVM. For water-injected meat samples with different water injection rates, the recognition accuracy presented a positive correlation with the water injection rate difference. Spare representation-based classifier eliminates the need for the training and feature extraction steps required by conventional pattern recognition models, and is suitable for processing data of high dimensionality and small sample size. Furthermore, it has a low
Deformable segmentation via sparse representation and dictionary learning.
Zhang, Shaoting; Zhan, Yiqiang; Metaxas, Dimitris N
2012-10-01
"Shape" and "appearance", the two pillars of a deformable model, complement each other in object segmentation. In many medical imaging applications, while the low-level appearance information is weak or mis-leading, shape priors play a more important role to guide a correct segmentation, thanks to the strong shape characteristics of biological structures. Recently a novel shape prior modeling method has been proposed based on sparse learning theory. Instead of learning a generative shape model, shape priors are incorporated on-the-fly through the sparse shape composition (SSC). SSC is robust to non-Gaussian errors and still preserves individual shape characteristics even when such characteristics is not statistically significant. Although it seems straightforward to incorporate SSC into a deformable segmentation framework as shape priors, the large-scale sparse optimization of SSC has low runtime efficiency, which cannot satisfy clinical requirements. In this paper, we design two strategies to decrease the computational complexity of SSC, making a robust, accurate and efficient deformable segmentation system. (1) When the shape repository contains a large number of instances, which is often the case in 2D problems, K-SVD is used to learn a more compact but still informative shape dictionary. (2) If the derived shape instance has a large number of vertices, which often appears in 3D problems, an affinity propagation method is used to partition the surface into small sub-regions, on which the sparse shape composition is performed locally. Both strategies dramatically decrease the scale of the sparse optimization problem and hence speed up the algorithm. Our method is applied on a diverse set of biomedical image analysis problems. Compared to the original SSC, these two newly-proposed modules not only significant reduce the computational complexity, but also improve the overall accuracy.
Optimized sparse-particle aerosol representations for modeling cloud-aerosol interactions
NASA Astrophysics Data System (ADS)
Fierce, Laura; McGraw, Robert
2016-04-01
Sparse representations of atmospheric aerosols are needed for efficient regional- and global-scale chemical transport models. Here we introduce a new framework for representing aerosol distributions, based on the method of moments. Given a set of moment constraints, we show how linear programming can be used to identify collections of sparse particles that approximately maximize distributional entropy. The collections of sparse particles derived from this approach reproduce CCN activity of the exact model aerosol distributions with high accuracy. Additionally, the linear programming techniques described in this study can be used to bound key aerosol properties, such as the number concentration of CCN. Unlike the commonly used sparse representations, such as modal and sectional schemes, the maximum-entropy moment-based approach is not constrained to pre-determined size bins or assumed distribution shapes. This study is a first step toward a new aerosol simulation scheme that will track multivariate aerosol distributions with sufficient computational efficiency for large-scale simulations.
Manifold Kernel Sparse Representation of Symmetric Positive-Definite Matrices and Its Applications.
Wu, Yuwei; Jia, Yunde; Li, Peihua; Zhang, Jian; Yuan, Junsong
2015-11-01
The symmetric positive-definite (SPD) matrix, as a connected Riemannian manifold, has become increasingly popular for encoding image information. Most existing sparse models are still primarily developed in the Euclidean space. They do not consider the non-linear geometrical structure of the data space, and thus are not directly applicable to the Riemannian manifold. In this paper, we propose a novel sparse representation method of SPD matrices in the data-dependent manifold kernel space. The graph Laplacian is incorporated into the kernel space to better reflect the underlying geometry of SPD matrices. Under the proposed framework, we design two different positive definite kernel functions that can be readily transformed to the corresponding manifold kernels. The sparse representation obtained has more discriminating power. Extensive experimental results demonstrate good performance of manifold kernel sparse codes in image classification, face recognition, and visual tracking.
Low-dose computed tomography image denoising based on joint wavelet and sparse representation.
Ghadrdan, Samira; Alirezaie, Javad; Dillenseger, Jean-Louis; Babyn, Paul
2014-01-01
Image denoising and signal enhancement are the most challenging issues in low dose computed tomography (CT) imaging. Sparse representational methods have shown initial promise for these applications. In this work we present a wavelet based sparse representation denoising technique utilizing dictionary learning and clustering. By using wavelets we extract the most suitable features in the images to obtain accurate dictionary atoms for the denoising algorithm. To achieve improved results we also lower the number of clusters which reduces computational complexity. In addition, a single image noise level estimation is developed to update the cluster centers in higher PSNRs. Our results along with the computational efficiency of the proposed algorithm clearly demonstrates the improvement of the proposed algorithm over other clustering based sparse representation (CSR) and K-SVD methods.
Sparse Representation-Based Image Quality Index With Adaptive Sub-Dictionaries.
Li, Leida; Cai, Hao; Zhang, Yabin; Lin, Weisi; Kot, Alex C; Sun, Xingming
2016-08-01
Distortions cause structural changes in digital images, leading to degraded visual quality. Dictionary-based sparse representation has been widely studied recently due to its ability to extract inherent image structures. Meantime, it can extract image features with slightly higher level semantics. Intuitively, sparse representation can be used for image quality assessment, because visible distortions can cause significant changes to the sparse features. In this paper, a new sparse representation-based image quality assessment model is proposed based on the construction of adaptive sub-dictionaries. An overcomplete dictionary trained from natural images is employed to capture the structure changes between the reference and distorted images by sparse feature extraction via adaptive sub-dictionary selection. Based on the observation that image sparse features are invariant to weak degradations and the perceived image quality is generally influenced by diverse issues, three auxiliary quality features are added, including gradient, color, and luminance information. The proposed method is not sensitive to training images, so a universal dictionary can be adopted for quality evaluation. Extensive experiments on five public image quality databases demonstrate that the proposed method produces the state-of-the-art results, and it delivers consistently well performances when tested in different image quality databases.
Tan, Lee N; Alwan, Abeer; Kossan, George; Cody, Martin L; Taylor, Charles E
2015-03-01
Annotation of phrases in birdsongs can be helpful to behavioral and population studies. To reduce the need for manual annotation, an automated birdsong phrase classification algorithm for limited data is developed. Limited data occur because of limited recordings or the existence of rare phrases. In this paper, classification of up to 81 phrase classes of Cassin's Vireo is performed using one to five training samples per class. The algorithm involves dynamic time warping (DTW) and two passes of sparse representation (SR) classification. DTW improves the similarity between training and test phrases from the same class in the presence of individual bird differences and phrase segmentation inconsistencies. The SR classifier works by finding a sparse linear combination of training feature vectors from all classes that best approximates the test feature vector. When the class decisions from DTW and the first pass SR classification are different, SR classification is repeated using training samples from these two conflicting classes. Compared to DTW, support vector machines, and an SR classifier without DTW, the proposed classifier achieves the highest classification accuracies of 94% and 89% on manually segmented and automatically segmented phrases, respectively, from unseen Cassin's Vireo individuals, using five training samples per class.
Sparse Representation for Computer Vision and Pattern Recognition
2009-05-01
with 40% occlusion. Figure 2 (right) shows the validation perfor - mance of the various methods, under 30% contiguous occlu- sion, plotted as a...performance. B. Sparse Modeling for Image Reconstruction Let X ∈ Rm×N be a set of N column data vectors xj ∈ Rm (e.g., image patches ), D ∈ Rm×K be a dictionary...K > m, and the patch sizes vary from 7× 7, m = 49, to 20× 20, m = 400 (in the multiscale case), with a sparsity of about 1/10th of the signal
NASA Astrophysics Data System (ADS)
He, Qingbo; Ding, Xiaoxi
2016-05-01
The transients caused by the localized fault are important measurement information for bearing fault diagnosis. Thus it is crucial to extract the transients from the bearing vibration or acoustic signals that are always corrupted by a large amount of background noise. In this paper, an iterative transient feature extraction approach is proposed based on time-frequency (TF) domain sparse representation. The approach is realized by presenting a new method, called local TF template matching. In this method, the TF atoms are constructed based on the TF distribution (TFD) of the Morlet wavelet bases and local TF templates are formulated from the TF atoms for the matching process. The instantaneous frequency (IF) ridge calculated from the TFD of an analyzed signal provides the frequency parameter values for the TF atoms as well as an effective template matching path on the TF plane. In each iteration, local TF templates are employed to do correlation with the TFD of the analyzed signal along the IF ridge tube for identifying the optimum parameters of transient wavelet model. With this iterative procedure, transients can be extracted in the TF domain from measured signals one by one. The final signal can be synthesized by combining the extracted TF atoms and the phase of the raw signal. The local TF template matching builds an effective TF matching-based sparse representation approach with the merit of satisfying the native pulse waveform structure of transients. The effectiveness of the proposed method is verified by practical defective bearing signals. Comparison results also show that the proposed method is superior to traditional methods in transient feature extraction.
Automated identification of crystallographic ligands using sparse-density representations
Carolan, C. G.; Lamzin, V. S.
2014-01-01
A novel procedure for the automatic identification of ligands in macromolecular crystallographic electron-density maps is introduced. It is based on the sparse parameterization of density clusters and the matching of the pseudo-atomic grids thus created to conformationally variant ligands using mathematical descriptors of molecular shape, size and topology. In large-scale tests on experimental data derived from the Protein Data Bank, the procedure could quickly identify the deposited ligand within the top-ranked compounds from a database of candidates. This indicates the suitability of the method for the identification of binding entities in fragment-based drug screening and in model completion in macromolecular structure determination. PMID:25004962
Compressive Fresnel digital holography using Fresnelet based sparse representation
NASA Astrophysics Data System (ADS)
Ramachandran, Prakash; Alex, Zachariah C.; Nelleri, Anith
2015-04-01
Compressive sensing (CS) in digital holography requires only very less number of pixel level detections in hologram plane for accurate image reconstruction and this is achieved by exploiting the sparsity of the object wave. When the input object fields are non-sparse in spatial domain, CS demands a suitable sparsification method like wavelet decomposition. The Fresnelet, a suitable wavelet basis for processing Fresnel digital holograms is an efficient sparsifier for the complex Fresnel field obtained by the Fresnel transform of the object field and minimizes the mutual coherence between sensing and sparsifying matrices involved in CS. The paper demonstrates the merits of Fresnelet based sparsification in compressive digital Fresnel holography over conventional method of sparsifying the input object field. The phase shifting digital Fresnel holography (PSDH) is used to retrieve the complex Fresnel field for the chosen problem. The results are presented from a numerical experiment to show the proof of the concept.
Single-Trial Sparse Representation-Based Approach for VEP Extraction
Yu, Nannan; Hu, Funian; Zou, Dexuan; Ding, Qisheng
2016-01-01
Sparse representation is a powerful tool in signal denoising, and visual evoked potentials (VEPs) have been proven to have strong sparsity over an appropriate dictionary. Inspired by this idea, we present in this paper a novel sparse representation-based approach to solving the VEP extraction problem. The extraction process is performed in three stages. First, instead of using the mixed signals containing the electroencephalogram (EEG) and VEPs, we utilise an EEG from a previous trial, which did not contain VEPs, to identify the parameters of the EEG autoregressive (AR) model. Second, instead of the moving average (MA) model, sparse representation is used to model the VEPs in the autoregressive-moving average (ARMA) model. Finally, we calculate the sparse coefficients and derive VEPs by using the AR model. Next, we tested the performance of the proposed algorithm with synthetic and real data, after which we compared the results with that of an AR model with exogenous input modelling and a mixed overcomplete dictionary-based sparse component decomposition method. Utilising the synthetic data, the algorithms are then employed to estimate the latencies of P100 of the VEPs corrupted by added simulated EEG at different signal-to-noise ratio (SNR) values. The validations demonstrate that our method can well preserve the details of the VEPs for latency estimation, even in low SNR environments. PMID:27807541
A joint sparse representation-based method for double-trial evoked potentials estimation.
Yu, Nannan; Liu, Haikuan; Wang, Xiaoyan; Lu, Hanbing
2013-12-01
In this paper, we present a novel approach to solving an evoked potentials estimating problem. Generally, the evoked potentials in two consecutive trials obtained by repeated identical stimuli of the nerves are extremely similar. In order to trace evoked potentials, we propose a joint sparse representation-based double-trial evoked potentials estimation method, taking full advantage of this similarity. The estimation process is performed in three stages: first, according to the similarity of evoked potentials and the randomness of a spontaneous electroencephalogram, the two consecutive observations of evoked potentials are considered as superpositions of the common component and the unique components; second, making use of their characteristics, the two sparse dictionaries are constructed; and finally, we apply the joint sparse representation method in order to extract the common component of double-trial observations, instead of the evoked potential in each trial. A series of experiments carried out on simulated and human test responses confirmed the superior performance of our method.
Robust multi-atlas label propagation by deep sparse representation.
Zu, Chen; Wang, Zhengxia; Zhang, Daoqiang; Liang, Peipeng; Shi, Yonghong; Shen, Dinggang; Wu, Guorong
2017-03-01
Recently, multi-atlas patch-based label fusion has achieved many successes in medical imaging area. The basic assumption in the current state-of-the-art approaches is that the image patch at the target image point can be represented by a patch dictionary consisting of atlas patches from registered atlas images. Therefore, the label at the target image point can be determined by fusing labels of atlas image patches with similar anatomical structures. However, such assumption on image patch representation does not always hold in label fusion since (1) the image content within the patch may be corrupted due to noise and artifact; and (2) the distribution of morphometric patterns among atlas patches might be unbalanced such that the majority patterns can dominate label fusion result over other minority patterns. The violation of the above basic assumptions could significantly undermine the label fusion accuracy. To overcome these issues, we first consider forming label-specific group for the atlas patches with the same label. Then, we alter the conventional flat and shallow dictionary to a deep multi-layer structure, where the top layer (label-specific dictionaries) consists of groups of representative atlas patches and the subsequent layers (residual dictionaries) hierarchically encode the patchwise residual information in different scales. Thus, the label fusion follows the representation consensus across representative dictionaries. However, the representation of target patch in each group is iteratively optimized by using the representative atlas patches in each label-specific dictionary exclusively to match the principal patterns and also using all residual patterns across groups collaboratively to overcome the issue that some groups might be absent of certain variation patterns presented in the target image patch. Promising segmentation results have been achieved in labeling hippocampus on ADNI dataset, as well as basal ganglia and brainstem structures, compared
Robust multi-atlas label propagation by deep sparse representation
Zu, Chen; Wang, Zhengxia; Zhang, Daoqiang; Liang, Peipeng; Shi, Yonghong; Shen, Dinggang; Wu, Guorong
2016-01-01
Recently, multi-atlas patch-based label fusion has achieved many successes in medical imaging area. The basic assumption in the current state-of-the-art approaches is that the image patch at the target image point can be represented by a patch dictionary consisting of atlas patches from registered atlas images. Therefore, the label at the target image point can be determined by fusing labels of atlas image patches with similar anatomical structures. However, such assumption on image patch representation does not always hold in label fusion since (1) the image content within the patch may be corrupted due to noise and artifact; and (2) the distribution of morphometric patterns among atlas patches might be unbalanced such that the majority patterns can dominate label fusion result over other minority patterns. The violation of the above basic assumptions could significantly undermine the label fusion accuracy. To overcome these issues, we first consider forming label-specific group for the atlas patches with the same label. Then, we alter the conventional flat and shallow dictionary to a deep multi-layer structure, where the top layer (label-specific dictionaries) consists of groups of representative atlas patches and the subsequent layers (residual dictionaries) hierarchically encode the patchwise residual information in different scales. Thus, the label fusion follows the representation consensus across representative dictionaries. However, the representation of target patch in each group is iteratively optimized by using the representative atlas patches in each label-specific dictionary exclusively to match the principal patterns and also using all residual patterns across groups collaboratively to overcome the issue that some groups might be absent of certain variation patterns presented in the target image patch. Promising segmentation results have been achieved in labeling hippocampus on ADNI dataset, as well as basal ganglia and brainstem structures, compared
NASA Astrophysics Data System (ADS)
Jóźwiak, Grzegorz
2017-03-01
Scanning probe microscopy (SPM) is a well known tool used for the investigation of phenomena in objects in the nanometer size range. However, quantitative results are limited by the size and the shape of the nanoprobe used in experiments. Blind tip reconstruction (BTR) is a very popular method used to reconstruct the upper boundary on the shape of the probe. This method is known to be very sensitive to all kinds of interference in the atomic force microscopy (AFM) image. Due to mathematical morphology calculus, the interference makes the BTR results biased rather than randomly disrupted. For this reason, the careful choice of methods used for image enhancement and denoising, as well as the shape of a calibration sample are very important. In the paper, the results of thorough investigations on the shape of a calibration standard are shown. A novel shape is proposed and a tool for the simulation of AFM images of this calibration standard was designed. It was shown that careful choice of the initial tip allows us to use images of hole structures to blindly reconstruct the shape of a probe. The simulator was used to test the impact of modern filtration algorithms on the BTR process. These techniques are based on sparse approximation with function dictionaries learned on the basis of an image itself. Various learning algorithms and parameters were tested to determine the optimal combination for sparse representation. It was observed that the strong reduction of noise does not guarantee strong reduction in reconstruction errors. It seems that further improvements will be possible by the combination of BTR and a noise reduction procedure.
Low-rank and eigenface based sparse representation for face recognition.
Hou, Yi-Fu; Sun, Zhan-Li; Chong, Yan-Wen; Zheng, Chun-Hou
2014-01-01
In this paper, based on low-rank representation and eigenface extraction, we present an improvement to the well known Sparse Representation based Classification (SRC). Firstly, the low-rank images of the face images of each individual in training subset are extracted by the Robust Principal Component Analysis (Robust PCA) to alleviate the influence of noises (e.g., illumination difference and occlusions). Secondly, Singular Value Decomposition (SVD) is applied to extract the eigenfaces from these low-rank and approximate images. Finally, we utilize these eigenfaces to construct a compact and discriminative dictionary for sparse representation. We evaluate our method on five popular databases. Experimental results demonstrate the effectiveness and robustness of our method.
Low-Rank and Eigenface Based Sparse Representation for Face Recognition
Hou, Yi-Fu; Sun, Zhan-Li; Chong, Yan-Wen; Zheng, Chun-Hou
2014-01-01
In this paper, based on low-rank representation and eigenface extraction, we present an improvement to the well known Sparse Representation based Classification (SRC). Firstly, the low-rank images of the face images of each individual in training subset are extracted by the Robust Principal Component Analysis (Robust PCA) to alleviate the influence of noises (e.g., illumination difference and occlusions). Secondly, Singular Value Decomposition (SVD) is applied to extract the eigenfaces from these low-rank and approximate images. Finally, we utilize these eigenfaces to construct a compact and discriminative dictionary for sparse representation. We evaluate our method on five popular databases. Experimental results demonstrate the effectiveness and robustness of our method. PMID:25334027
NASA Astrophysics Data System (ADS)
Yu, Caixia; Zhao, Jingtao; Wang, Yanfei
2017-02-01
Studying small-scale geologic discontinuities, such as faults, cavities and fractures, plays a vital role in analyzing the inner conditions of reservoirs, as these geologic structures and elements can provide storage spaces and migration pathways for petroleum. However, these geologic discontinuities have weak energy and are easily contaminated with noises, and therefore effectively extracting them from seismic data becomes a challenging problem. In this paper, a method for detecting small-scale discontinuities using dictionary learning and sparse representation is proposed that can dig up high-resolution information by sparse coding. A K-SVD (K-means clustering via Singular Value Decomposition) sparse representation model that contains two stage of iteration procedure: sparse coding and dictionary updating, is suggested for mathematically expressing these seismic small-scale discontinuities. Generally, the orthogonal matching pursuit (OMP) algorithm is employed for sparse coding. However, the method can only update one dictionary atom at one time. In order to improve calculation efficiency, a regularized version of OMP algorithm is presented for simultaneously updating a number of atoms at one time. Two numerical experiments demonstrate the validity of the developed method for clarifying and enhancing small-scale discontinuities. The field example of carbonate reservoirs further demonstrates its effectiveness in revealing masked tiny faults and small-scale cavities.
Zhang, Xiaodong; Jing, Shasha; Gao, Peiyi; Xue, Jing; Su, Lu; Li, Weiping; Ren, Lijie
2016-01-01
Segmentation of infarcts at hyperacute stage is challenging as they exhibit substantial variability which may even be hard for experts to delineate manually. In this paper, a sparse representation based classification method is explored. For each patient, four volumetric data items including three volumes of diffusion weighted imaging and a computed asymmetry map are employed to extract patch features which are then fed to dictionary learning and classification based on sparse representation. Elastic net is adopted to replace the traditional L0-norm/L1-norm constraints on sparse representation to stabilize sparse code. To decrease computation cost and to reduce false positives, regions-of-interest are determined to confine candidate infarct voxels. The proposed method has been validated on 98 consecutive patients recruited within 6 hours from onset. It is shown that the proposed method could handle well infarcts with intensity variability and ill-defined edges to yield significantly higher Dice coefficient (0.755 ± 0.118) than the other two methods and their enhanced versions by confining their segmentations within the regions-of-interest (average Dice coefficient less than 0.610). The proposed method could provide a potential tool to quantify infarcts from diffusion weighted imaging at hyperacute stage with accuracy and speed to assist the decision making especially for thrombolytic therapy. PMID:27746825
Detection of dual-band infrared small target based on joint dynamic sparse representation
NASA Astrophysics Data System (ADS)
Zhou, Jinwei; Li, Jicheng; Shi, Zhiguang; Lu, Xiaowei; Ren, Dongwei
2015-10-01
Infrared small target detection is a crucial and yet still is a difficult issue in aeronautic and astronautic applications. Sparse representation is an important mathematic tool and has been used extensively in image processing in recent years. Joint sparse representation is applied in dual-band infrared dim target detection in this paper. Firstly, according to the characters of dim targets in dual-band infrared images, 2-dimension Gaussian intensity model was used to construct target dictionary, then the dictionary was classified into different sub-classes according to different positions of Gaussian function's center point in image block; The fact that dual-band small targets detection can use the same dictionary and the sparsity doesn't lie in atom-level but in sub-class level was utilized, hence the detection of targets in dual-band infrared images was converted to be a joint dynamic sparse representation problem. And the dynamic active sets were used to describe the sparse constraint of coefficients. Two modified sparsity concentration index (SCI) criteria was proposed to evaluate whether targets exist in the images. In experiments, it shows that the proposed algorithm can achieve better detecting performance and dual-band detection is much more robust to noise compared with single-band detection. Moreover, the proposed method can be expanded to multi-spectrum small target detection.
Signal denoising and ultrasonic flaw detection via overcomplete and sparse representations.
Zhang, Guang-Ming; Harvey, David M; Braden, Derek R
2008-11-01
Sparse signal representations from overcomplete dictionaries are the most recent technique in the signal processing community. Applications of this technique extend into many fields. In this paper, this technique is utilized to cope with ultrasonic flaw detection and noise suppression problem. In particular, a noisy ultrasonic signal is decomposed into sparse representations using a sparse Bayesian learning algorithm and an overcomplete dictionary customized from a Gabor dictionary by incorporating some a priori information of the transducer used. Nonlinear postprocessing including thresholding and pruning is then applied to the decomposed coefficients to reduce the noise contribution and extract the flaw information. Because of the high compact essence of sparse representations, flaw echoes are packed into a few significant coefficients, and noise energy is likely scattered all over the dictionary atoms, generating insignificant coefficients. This property greatly increases the efficiency of the pruning and thresholding operations and is extremely useful for detecting flaw echoes embedded in background noise. The performance of the proposed approach is verified experimentally and compared with the wavelet transform signal processor. Experimental results to detect ultrasonic flaw echoes contaminated by white Gaussian additive noise or correlated noise are presented in the paper.
NASA Astrophysics Data System (ADS)
Chen, Jun-xin; Zhu, Zhi-liang; Fu, Chong; Yu, Hai; Zhang, Li-bo
2015-07-01
Optical information security systems have drawn long-term concerns. In this paper, an optical information authentication approach using gyrator transform based double random phase encoding with sparse representation is proposed. Different from traditional optical encryption schemes, only sparse version of the ciphertext is preserved, and hence the decrypted result is completely unrecognizable and shows no similarity to the plaintext. However, we demonstrate that the noise-like decipher result can be effectively authenticated by means of optical correlation approach. Simulations prove that the proposed method is feasible and effective, and can provide additional protection for optical security systems.
NASA Astrophysics Data System (ADS)
Song, Xiaoning; Feng, Zhen-Hua; Hu, Guosheng; Yang, Xibei; Yang, Jingyu; Qi, Yunsong
2015-09-01
This paper proposes a progressive sparse representation-based classification algorithm using local discrete cosine transform (DCT) evaluation to perform face recognition. Specifically, the sum of the contributions of all training samples of each subject is first taken as the contribution of this subject, then the redundant subject with the smallest contribution to the test sample is iteratively eliminated. Second, the progressive method aims at representing the test sample as a linear combination of all the remaining training samples, by which the representation capability of each training sample is exploited to determine the optimal "nearest neighbors" for the test sample. Third, the transformed DCT evaluation is constructed to measure the similarity between the test sample and each local training sample using cosine distance metrics in the DCT domain. The final goal of the proposed method is to determine an optimal weighted sum of nearest neighbors that are obtained under the local correlative degree evaluation, which is approximately equal to the test sample, and we can use this weighted linear combination to perform robust classification. Experimental results conducted on the ORL database of faces (created by the Olivetti Research Laboratory in Cambridge), the FERET face database (managed by the Defense Advanced Research Projects Agency and the National Institute of Standards and Technology), AR face database (created by Aleix Martinez and Robert Benavente in the Computer Vision Center at U.A.B), and USPS handwritten digit database (gathered at the Center of Excellence in Document Analysis and Recognition at SUNY Buffalo) demonstrate the effectiveness of the proposed method.
A dedicated greedy pursuit algorithm for sparse spectral representation of music sound.
Rebollo-Neira, Laura; Aggarwal, Gagan
2016-10-01
A dedicated algorithm for sparse spectral representation of music sound is presented. The goal is to enable the representation of a piece of music signal as a linear superposition of as few spectral components as possible, without affecting the quality of the reproduction. A representation of this nature is said to be sparse. In the present context sparsity is accomplished by greedy selection of the spectral components, from an overcomplete set called a dictionary. The proposed algorithm is tailored to be applied with trigonometric dictionaries. Its distinctive feature being that it avoids the need for the actual construction of the whole dictionary, by implementing the required operations via the fast Fourier transform. The achieved sparsity is theoretically equivalent to that rendered by the orthogonal matching pursuit (OMP) method. The contribution of the proposed dedicated implementation is to extend the applicability of the standard OMP algorithm, by reducing its storage and computational demands. The suitability of the approach for producing sparse spectral representation is illustrated by comparison with the traditional method, in the line of the short time Fourier transform, involving only the corresponding orthonormal trigonometric basis.
Embedded sparse representation of fMRI data via group-wise dictionary optimization
NASA Astrophysics Data System (ADS)
Zhu, Dajiang; Lin, Binbin; Faskowitz, Joshua; Ye, Jieping; Thompson, Paul M.
2016-03-01
Sparse learning enables dimension reduction and efficient modeling of high dimensional signals and images, but it may need to be tailored to best suit specific applications and datasets. Here we used sparse learning to efficiently represent functional magnetic resonance imaging (fMRI) data from the human brain. We propose a novel embedded sparse representation (ESR), to identify the most consistent dictionary atoms across different brain datasets via an iterative group-wise dictionary optimization procedure. In this framework, we introduced additional criteria to make the learned dictionary atoms more consistent across different subjects. We successfully identified four common dictionary atoms that follow the external task stimuli with very high accuracy. After projecting the corresponding coefficient vectors back into the 3-D brain volume space, the spatial patterns are also consistent with traditional fMRI analysis results. Our framework reveals common features of brain activation in a population, as a new, efficient fMRI analysis method.
Enhancement of snow cover change detection with sparse representation and dictionary learning
NASA Astrophysics Data System (ADS)
Varade, D.; Dikshit, O.
2014-11-01
Sparse representation and decoding is often used for denoising images and compression of images with respect to inherent features. In this paper, we adopt a methodology incorporating sparse representation of a snow cover change map using the K-SVD trained dictionary and sparse decoding to enhance the change map. The pixels often falsely characterized as "changes" are eliminated using this approach. The preliminary change map was generated using differenced NDSI or S3 maps in case of Resourcesat-2 and Landsat 8 OLI imagery respectively. These maps are extracted into patches for compressed sensing using Discrete Cosine Transform (DCT) to generate an initial dictionary which is trained by the K-SVD approach. The trained dictionary is used for sparse coding of the change map using the Orthogonal Matching Pursuit (OMP) algorithm. The reconstructed change map incorporates a greater degree of smoothing and represents the features (snow cover changes) with better accuracy. The enhanced change map is segmented using kmeans to discriminate between the changed and non-changed pixels. The segmented enhanced change map is compared, firstly with the difference of Support Vector Machine (SVM) classified NDSI maps and secondly with a reference data generated as a mask by visual interpretation of the two input images. The methodology is evaluated using multi-spectral datasets from Resourcesat-2 and Landsat-8. The k-hat statistic is computed to determine the accuracy of the proposed approach.
High resolution OCT image generation using super resolution via sparse representation
NASA Astrophysics Data System (ADS)
Asif, Muhammad; Akram, Muhammad Usman; Hassan, Taimur; Shaukat, Arslan; Waqar, Razi
2017-02-01
In this paper we propose a technique for obtaining a high resolution (HR) image from a single low resolution (LR) image -using joint learning dictionary - on the basis of image statistic research. It suggests that with an appropriate choice of an over-complete dictionary, image patches can be well represented as a sparse linear combination. Medical imaging for clinical analysis and medical intervention is being used for creating visual representations of the interior of a body, as well as visual representation of the function of some organs or tissues (physiology). A number of medical imaging techniques are in use like MRI, CT scan, X-rays and Optical Coherence Tomography (OCT). OCT is one of the new technologies in medical imaging and one of its uses is in ophthalmology where it is being used for analysis of the choroidal thickness in the eyes in healthy and disease states such as age-related macular degeneration, central serous chorioretinopathy, diabetic retinopathy and inherited retinal dystrophies. We have proposed a technique for enhancing the OCT images which can be used for clearly identifying and analyzing the particular diseases. Our method uses dictionary learning technique for generating a high resolution image from a single input LR image. We train two joint dictionaries, one with OCT images and the second with multiple different natural images, and compare the results with previous SR technique. Proposed method for both dictionaries produces HR images which are comparatively superior in quality with the other proposed method of SR. Proposed technique is very effective for noisy OCT images and produces up-sampled and enhanced OCT images.
Super-resolution of hyperspectral images using sparse representation and Gabor prior
NASA Astrophysics Data System (ADS)
Patel, Rakesh C.; Joshi, Manjunath V.
2016-04-01
Super-resolution (SR) as a postprocessing technique is quite useful in enhancing the spatial resolution of hyperspectral (HS) images without affecting its spectral resolution. We present an approach to increase the spatial resolution of HS images by making use of sparse representation and Gabor prior. The low-resolution HS observations consisting of large number of bands are represented as a linear combination of a small number of basis images using principal component analysis (PCA), and the significant components are used in our work. We first obtain initial estimates of SR on this reduced dimension by using compressive sensing-based method. Since SR is an ill-posed problem, the final solution is obtained by using a regularization framework. The novelty of our approach lies in: (1) estimation of optimal point spread function in the form of decimation matrix, and (2) using a new prior called "Gabor prior" to super-resolve the significant PCA components. Experiments are conducted on two different HS datasets namely, 31-band natural HS image set collected under controlled laboratory environment and a set of 224-band real HS images collected by airborne visible/infrared imaging spectrometer remote sensing sensor. Visual inspections and quantitative comparison confirm that our method enhances spatial information without introducing significant spectral distortion. Our conclusions include: (1) incorporate the sensor characteristics in the form of estimated decimation matrix for SR, and (2) preserve various frequencies in super-resolved image by making use of Gabor prior.
Zhang, Guoqing; Sun, Huaijiang; Xia, Guiyu; Sun, Quansen
2016-07-07
Sparse representation based classification (SRC) has been developed and shown great potential for real-world application. Based on SRC, Yang et al. [10] devised a SRC steered discriminative projection (SRC-DP) method. However, as a linear algorithm, SRC-DP cannot handle the data with highly nonlinear distribution. Kernel sparse representation-based classifier (KSRC) is a non-linear extension of SRC and can remedy the drawback of SRC. KSRC requires the use of a predetermined kernel function and selection of the kernel function and its parameters is difficult. Recently, multiple kernel learning for SRC (MKL-SRC) [22] has been proposed to learn a kernel from a set of base kernels. However, MKL-SRC only considers the within-class reconstruction residual while ignoring the between-class relationship, when learning the kernel weights. In this paper, we propose a novel multiple kernel sparse representation-based classifier (MKSRC), and then we use it as a criterion to design a multiple kernel sparse representation based orthogonal discriminative projection method (MK-SR-ODP). The proposed algorithm aims at learning a projection matrix and a corresponding kernel from the given base kernels such that in the low dimension subspace the between-class reconstruction residual is maximized and the within-class reconstruction residual is minimized. Furthermore, to achieve a minimum overall loss by performing recognition in the learned low-dimensional subspace, we introduce cost information into the dimensionality reduction method. The solutions for the proposed method can be efficiently found based on trace ratio optimization method [33]. Extensive experimental results demonstrate the superiority of the proposed algorithm when compared with the state-of-the-art methods.
Sparse coding based dense feature representation model for hyperspectral image classification
NASA Astrophysics Data System (ADS)
Oguslu, Ender; Zhou, Guoqing; Zheng, Zezhong; Iftekharuddin, Khan; Li, Jiang
2015-11-01
We present a sparse coding based dense feature representation model (a preliminary version of the paper was presented at the SPIE Remote Sensing Conference, Dresden, Germany, 2013) for hyperspectral image (HSI) classification. The proposed method learns a new representation for each pixel in HSI through the following four steps: sub-band construction, dictionary learning, encoding, and feature selection. The new representation usually has a very high dimensionality requiring a large amount of computational resources. We applied the l1/lq regularized multiclass logistic regression technique to reduce the size of the new representation. We integrated the method with a linear support vector machine (SVM) and a composite kernels SVM (CKSVM) to discriminate different types of land cover. We evaluated the proposed algorithm on three well-known HSI datasets and compared our method to four recently developed classification methods: SVM, CKSVM, simultaneous orthogonal matching pursuit, and image fusion and recursive filtering. Experimental results show that the proposed method can achieve better overall and average classification accuracies with a much more compact representation leading to more efficient sparse models for HSI classification.
A dedicated greedy pursuit algorithm for sparse spectral representation of music sound
NASA Astrophysics Data System (ADS)
Rebollo-Neira, Laura; Aggarwal, Gagan
2016-10-01
A dedicated algorithm for sparse spectral representation of music sound is presented. The goal is to enable the representation of a piece of music signal, as a linear superposition of as few spectral components as possible. A representation of this nature is said to be sparse. In the present context sparsity is accomplished by greedy selection of the spectral components, from an overcomplete set called a dictionary. The proposed algorithm is tailored to be applied with trigonometric dictionaries. Its distinctive feature being that it avoids the need for the actual construction of the whole dictionary, by implementing the required operations via the Fast Fourier Transform. The achieved sparsity is theoretically equivalent to that rendered by the Orthogonal Matching Pursuit method. The contribution of the proposed dedicated implementation is to extend the applicability of the standard Orthogonal Matching Pursuit algorithm, by reducing its storage and computational demands. The suitability of the approach for producing sparse spectral models is illustrated by comparison with the traditional method, in the line of the Short Time Fourier Transform, involving only the corresponding orthonormal trigonometric basis.
Robust brain parcellation using sparse representation on resting-state fMRI.
Zhang, Yu; Caspers, Svenja; Fan, Lingzhong; Fan, Yong; Song, Ming; Liu, Cirong; Mo, Yin; Roski, Christian; Eickhoff, Simon; Amunts, Katrin; Jiang, Tianzi
2015-11-01
Resting-state fMRI (rs-fMRI) has been widely used to segregate the brain into individual modules based on the presence of distinct connectivity patterns. Many parcellation methods have been proposed for brain parcellation using rs-fMRI, but their results have been somewhat inconsistent, potentially due to various types of noise. In this study, we provide a robust parcellation method for rs-fMRI-based brain parcellation, which constructs a sparse similarity graph based on the sparse representation coefficients of each seed voxel and then uses spectral clustering to identify distinct modules. Both the local time-varying BOLD signals and whole-brain connectivity patterns may be used as features and yield similar parcellation results. The robustness of our method was tested on both simulated and real rs-fMRI datasets. In particular, on simulated rs-fMRI data, sparse representation achieved good performance across different noise levels, including high accuracy of parcellation and high robustness to noise. On real rs-fMRI data, stable parcellation of the medial frontal cortex (MFC) and parietal operculum (OP) were achieved on three different datasets, with high reproducibility within each dataset and high consistency across these results. Besides, the parcellation of MFC was little influenced by the degrees of spatial smoothing. Furthermore, the consistent parcellation of OP was also well corresponding to cytoarchitectonic subdivisions and known somatotopic organizations. Our results demonstrate a new promising approach to robust brain parcellation using resting-state fMRI by sparse representation.
Hosseini-Asl, Ehsan; Zurada, Jacek M; Nasraoui, Olfa
2016-12-01
We demonstrate a new deep learning autoencoder network, trained by a nonnegativity constraint algorithm (nonnegativity-constrained autoencoder), that learns features that show part-based representation of data. The learning algorithm is based on constraining negative weights. The performance of the algorithm is assessed based on decomposing data into parts and its prediction performance is tested on three standard image data sets and one text data set. The results indicate that the nonnegativity constraint forces the autoencoder to learn features that amount to a part-based representation of data, while improving sparsity and reconstruction quality in comparison with the traditional sparse autoencoder and nonnegative matrix factorization. It is also shown that this newly acquired representation improves the prediction performance of a deep neural network.
A Sparse Hierarchical Map Representation for Mars Science Laboratory Science Operations
NASA Astrophysics Data System (ADS)
Nefian, A. V.; Edwards, L. J.; Keely, L.; Lees, D. S.; Fluckinger, L.; Malin, M. C.; Parker, T. J.
2015-12-01
We describe a solution for multi-scale Mars terrain modeling and mapping with Digital Elevation Models (DEMs) and co-registered orthogonally projected imagery (ortho-images). High resolution DEMs and ortho-images derived from Mars Science Laboratory (MSL) rover science and navigation cameras are represented in context with lower resolution, wide coverage DEMs and ortho-images derived from Mars Reconnaissance Orbiter (MRO) HiRISE and CTX camera images and Mars Express (MEX) mission HRSC images. Merging MSL rover image derived terrain models with those from orbital images at a uniform high resolution would require super-sampling of the orbital data across a large area to maintain significant context. This solution is not practical, and would result in a mapping product of enormous size. Instead, we choose a sparse hierarchical map representation. Each level in this hierarchical representation is a map described by a set of tiles with fixed number of samples and fixed resolution. The number of samples in a tile is fixed for all levels and each level is associated with a specific resolution. In this work, the resolution ratio between two adjacent levels is set to two. The map at each level is sparse and it contains only the tiles for which data is available at the resolution of the given level. For example, at the highest resolution level only MSL science camera models are available and only a small set of tiles are generated in a sparse map. At the lowest resolution, the map contains the complete set of tiles. The reference level of the representation is chosen to be the HiRISE terrain model and CTX, HRSC and MSL data are projected onto this model before being mapped. While our terrain representation was developed for use in "Antares", a visual planning and sequencing tool for MSL science cameras developed at NASA Ames Research Center, it is general purpose and has a number of potential geo-science visualization applications.
Spectrum recovery method based on sparse representation for segmented multi-Gaussian model
NASA Astrophysics Data System (ADS)
Teng, Yidan; Zhang, Ye; Ti, Chunli; Su, Nan
2016-09-01
Hyperspectral images can realize crackajack features discriminability for supplying diagnostic characteristics with high spectral resolution. However, various degradations may generate negative influence on the spectral information, including water absorption, bands-continuous noise. On the other hand, the huge data volume and strong redundancy among spectrums produced intense demand on compressing HSIs in spectral dimension, which also leads to the loss of spectral information. The reconstruction of spectral diagnostic characteristics has irreplaceable significance for the subsequent application of HSIs. This paper introduces a spectrum restoration method for HSIs making use of segmented multi-Gaussian model (SMGM) and sparse representation. A SMGM is established to indicating the unsymmetrical spectral absorption and reflection characteristics, meanwhile, its rationality and sparse property are discussed. With the application of compressed sensing (CS) theory, we implement sparse representation to the SMGM. Then, the degraded and compressed HSIs can be reconstructed utilizing the uninjured or key bands. Finally, we take low rank matrix recovery (LRMR) algorithm for post processing to restore the spatial details. The proposed method was tested on the spectral data captured on the ground with artificial water absorption condition and an AVIRIS-HSI data set. The experimental results in terms of qualitative and quantitative assessments demonstrate that the effectiveness on recovering the spectral information from both degradations and loss compression. The spectral diagnostic characteristics and the spatial geometry feature are well preserved.
Infrared moving small target detection based on saliency extraction and image sparse representation
NASA Astrophysics Data System (ADS)
Zhang, Xiaomin; Ren, Kan; Gao, Jin; Li, Chaowei; Gu, Guohua; Wan, Minjie
2016-10-01
Moving small target detection in infrared image is a crucial technique of infrared search and tracking system. This paper present a novel small target detection technique based on frequency-domain saliency extraction and image sparse representation. First, we exploit the features of Fourier spectrum image and magnitude spectrum of Fourier transform to make a rough extract of saliency regions and use a threshold segmentation system to classify the regions which look salient from the background, which gives us a binary image as result. Second, a new patch-image model and over-complete dictionary were introduced to the detection system, then the infrared small target detection was converted into a problem solving and optimization process of patch-image information reconstruction based on sparse representation. More specifically, the test image and binary image can be decomposed into some image patches follow certain rules. We select the target potential area according to the binary patch-image which contains salient region information, then exploit the over-complete infrared small target dictionary to reconstruct the test image blocks which may contain targets. The coefficients of target image patch satisfy sparse features. Finally, for image sequence, Euclidean distance was used to reduce false alarm ratio and increase the detection accuracy of moving small targets in infrared images due to the target position correlation between frames.
NASA Astrophysics Data System (ADS)
Tian, Shu; Zhang, Ye; Yan, Yimin; Su, Nan; Zhang, Junping
2016-09-01
Latent low-rank representation (LatLRR) has been attached considerable attention in the field of remote sensing image segmentation, due to its effectiveness in exploring the multiple subspace structures of data. However, the increasingly heterogeneous texture information in the high spatial resolution remote sensing images, leads to more severe interference of pixels in local neighborhood, and the LatLRR fails to capture the local complex structure information. Therefore, we present a local sparse structure constrainted latent low-rank representation (LSSLatLRR) segmentation method, which explicitly imposes the local sparse structure constraint on LatLRR to capture the intrinsic local structure in manifold structure feature subspaces. The whole segmentation framework can be viewed as two stages in cascade. In the first stage, we use the local histogram transform to extract the texture local histogram features (LHOG) at each pixel, which can efficiently capture the complex and micro-texture pattern. In the second stage, a local sparse structure (LSS) formulation is established on LHOG, which aims to preserve the local intrinsic structure and enhance the relationship between pixels having similar local characteristics. Meanwhile, by integrating the LSS and the LatLRR, we can efficiently capture the local sparse and low-rank structure in the mixture of feature subspace, and we adopt the subspace segmentation method to improve the segmentation accuracy. Experimental results on the remote sensing images with different spatial resolution show that, compared with three state-of-the-art image segmentation methods, the proposed method achieves more accurate segmentation results.
Zhang, Zutao; Luo, Dianyuan; Rasim, Yagubov; Li, Yanjun; Meng, Guanjun; Xu, Jian; Wang, Chunbai
2016-02-19
In this paper, we present a vehicle active safety model for vehicle speed control based on driver vigilance detection using low-cost, comfortable, wearable electroencephalographic (EEG) sensors and sparse representation. The proposed system consists of three main steps, namely wireless wearable EEG collection, driver vigilance detection, and vehicle speed control strategy. First of all, a homemade low-cost comfortable wearable brain-computer interface (BCI) system with eight channels is designed for collecting the driver's EEG signal. Second, wavelet de-noising and down-sample algorithms are utilized to enhance the quality of EEG data, and Fast Fourier Transformation (FFT) is adopted to extract the EEG power spectrum density (PSD). In this step, sparse representation classification combined with k-singular value decomposition (KSVD) is firstly introduced in PSD to estimate the driver's vigilance level. Finally, a novel safety strategy of vehicle speed control, which controls the electronic throttle opening and automatic braking after driver fatigue detection using the above method, is presented to avoid serious collisions and traffic accidents. The simulation and practical testing results demonstrate the feasibility of the vehicle active safety model.
Zhang, Zutao; Luo, Dianyuan; Rasim, Yagubov; Li, Yanjun; Meng, Guanjun; Xu, Jian; Wang, Chunbai
2016-01-01
In this paper, we present a vehicle active safety model for vehicle speed control based on driver vigilance detection using low-cost, comfortable, wearable electroencephalographic (EEG) sensors and sparse representation. The proposed system consists of three main steps, namely wireless wearable EEG collection, driver vigilance detection, and vehicle speed control strategy. First of all, a homemade low-cost comfortable wearable brain-computer interface (BCI) system with eight channels is designed for collecting the driver’s EEG signal. Second, wavelet de-noising and down-sample algorithms are utilized to enhance the quality of EEG data, and Fast Fourier Transformation (FFT) is adopted to extract the EEG power spectrum density (PSD). In this step, sparse representation classification combined with k-singular value decomposition (KSVD) is firstly introduced in PSD to estimate the driver’s vigilance level . Finally, a novel safety strategy of vehicle speed control, which controls the electronic throttle opening and automatic braking after driver fatigue detection using the above method, is presented to avoid serious collisions and traffic accidents. The simulation and practical testing results demonstrate the feasibility of the vehicle active safety model. PMID:26907278
Jiang, Xi; Li, Xiang; Lv, Jinglei; Zhang, Tuo; Zhang, Shu; Guo, Lei; Liu, Tianming
2015-12-01
The recently publicly released Human Connectome Project (HCP) grayordinate-based fMRI data not only has high spatial and temporal resolution, but also offers group-corresponding fMRI signals across a large population for the first time in the brain imaging field, thus significantly facilitating mapping the functional brain architecture with much higher resolution and in a group-wise fashion. In this article, we adopt the HCP grayordinate task-based fMRI (tfMRI) data to systematically identify and characterize task-based heterogeneous functional regions (THFRs) on cortical surface, i.e., the regions that are activated during multiple tasks conditions and contribute to multiple task-evoked systems during a specific task performance, and to assess the spatial patterns of identified THFRs on cortical gyri and sulci by applying a computational framework of sparse representations of grayordinate brain tfMRI signals. Experimental results demonstrate that both consistent task-evoked networks and intrinsic connectivity networks across all subjects and tasks in HCP grayordinate data are effectively and robustly reconstructed via the proposed sparse representation framework. Moreover, it is found that there are relatively consistent THFRs locating at bilateral parietal lobe, frontal lobe, and visual association cortices across all subjects and tasks. Particularly, those identified THFRs locate significantly more on gyral regions than on sulcal regions. These results based on sparse representation of HCP grayordinate data reveal novel functional architecture of cortical gyri and sulci, and might provide a foundation to better understand functional mechanisms of the human cerebral cortex in the future.
Multiple-image encryption and authentication with sparse representation by space multiplexing.
Gong, Qiong; Liu, Xuyan; Li, Genquan; Qin, Yi
2013-11-01
A multiple-image encryption and authentication approach by space multiplexing has been proposed. The redundant spaces in the previous security systems employing sparse representation strategy are optimized. With the proposal the information of multiple images can be integrated into a synthesized ciphertext that is convenient for storage and transmission. Only when all the keys are correct can the information of the primary images be authenticated. Computer simulation results have demonstrated that the proposed method is feasible and effective. Moreover, the proposal is also proved to be robust against occlusion and noise attacks.
SPARSE REPRESENTATIONS WITH DATA FIDELITY TERM VIA AN ITERATIVELY REWEIGHTED LEAST SQUARES ALGORITHM
WOHLBERG, BRENDT; RODRIGUEZ, PAUL
2007-01-08
Basis Pursuit and Basis Pursuit Denoising, well established techniques for computing sparse representations, minimize an {ell}{sup 2} data fidelity term subject to an {ell}{sup 1} sparsity constraint or regularization term on the solution by mapping the problem to a linear or quadratic program. Basis Pursuit Denoising with an {ell}{sup 1} data fidelity term has recently been proposed, also implemented via a mapping to a linear program. They introduce an alternative approach via an iteratively Reweighted Least Squares algorithm, providing greater flexibility in the choice of data fidelity term norm, and computational advantages in certain circumstances.
NASA Astrophysics Data System (ADS)
Zhang, Hong; Hou, Rui; Yi, Lei; Meng, Juan; Pan, Zhisong; Zhou, Yuhuan
2016-07-01
The accurate identification of encrypted data stream helps to regulate illegal data, detect network attacks and protect users' information. In this paper, a novel encrypted data stream identification algorithm is introduced. The proposed method is based on randomness characteristics of encrypted data stream. We use a l1-norm regularized logistic regression to improve sparse representation of randomness features and Fuzzy Gaussian Mixture Model (FGMM) to improve identification accuracy. Experimental results demonstrate that the method can be adopted as an effective technique for encrypted data stream identification.
Yu, Renping; Zhang, Han; An, Le; Chen, Xiaobo; Wei, Zhihui; Shen, Dinggang
2017-02-02
Brain functional network analysis has shown great potential in understanding brain functions and also in identifying biomarkers for brain diseases, such as Alzheimer's disease (AD) and its early stage, mild cognitive impairment (MCI). In these applications, accurate construction of biologically meaningful brain network is critical. Sparse learning has been widely used for brain network construction; however, its l1 -norm penalty simply penalizes each edge of a brain network equally, without considering the original connectivity strength which is one of the most important inherent linkwise characters. Besides, based on the similarity of the linkwise connectivity, brain network shows prominent group structure (i.e., a set of edges sharing similar attributes). In this article, we propose a novel brain functional network modeling framework with a "connectivity strength-weighted sparse group constraint." In particular, the network modeling can be optimized by considering both raw connectivity strength and its group structure, without losing the merit of sparsity. Our proposed method is applied to MCI classification, a challenging task for early AD diagnosis. Experimental results based on the resting-state functional MRI, from 50 MCI patients and 49 healthy controls, show that our proposed method is more effective (i.e., achieving a significantly higher classification accuracy, 84.8%) than other competing methods (e.g., sparse representation, accuracy = 65.6%). Post hoc inspection of the informative features further shows more biologically meaningful brain functional connectivities obtained by our proposed method. Hum Brain Mapp, 2017. © 2017 Wiley Periodicals, Inc.
Aspect-Aided Dynamic Non-Negative Sparse Representation-Based Microwave Image Classification
Zhang, Xinzheng; Yang, Qiuyue; Liu, Miaomiao; Jia, Yunjian; Liu, Shujun; Li, Guojun
2016-01-01
Classification of target microwave images is an important application in much areas such as security, surveillance, etc. With respect to the task of microwave image classification, a recognition algorithm based on aspect-aided dynamic non-negative least square (ADNNLS) sparse representation is proposed. Firstly, an aspect sector is determined, the center of which is the estimated aspect angle of the testing sample. The training samples in the aspect sector are divided into active atoms and inactive atoms by smooth self-representative learning. Secondly, for each testing sample, the corresponding active atoms are selected dynamically, thereby establishing dynamic dictionary. Thirdly, the testing sample is represented with ℓ1-regularized non-negative sparse representation under the corresponding dynamic dictionary. Finally, the class label of the testing sample is identified by use of the minimum reconstruction error. Verification of the proposed algorithm was conducted using the Moving and Stationary Target Acquisition and Recognition (MSTAR) database which was acquired by synthetic aperture radar. Experiment results validated that the proposed approach was able to capture the local aspect characteristics of microwave images effectively, thereby improving the classification performance. PMID:27598172
NASA Astrophysics Data System (ADS)
Korez, Robert; Likar, Boštjan; Pernuš, Franjo; Vrtovec, Tomaž
2016-03-01
Automated detection and segmentation of vertebral bodies from spinal computed tomography (CT) images is usually a prerequisite step for numerous spine-related medical applications, such as diagnosis, surgical planning and follow-up assessment of spinal pathologies. However, automated detection and segmentation are challenging tasks due to a relatively high degree of anatomical complexity, presence of unclear boundaries and articulation of vertebrae with each other. In this paper, we describe a sparse representation error minimization (SEM) framework for joint detection and segmentation of vertebral bodies in CT images. By minimizing the sparse representation error of sampled intensity values, we are able to recover the oriented bounding box (OBB) and segmentation binary mask for each vertebral body in the CT image. The performance of the proposed SEM framework was evaluated on five CT images of the thoracolumbar spine. The resulting Euclidean distance of 1:75+/-1:02 mm, computed between the center points of recovered and corresponding reference OBBs, and Dice coefficient of 92:3+/-2:7%, computed between the resulting and corresponding reference segmentation binary masks, indicate that the proposed framework can successfully detect and segment vertebral bodies in CT images of the thoracolumbar spine.
Contour tracking in echocardiographic sequences via sparse representation and dictionary learning.
Huang, Xiaojie; Dione, Donald P; Compas, Colin B; Papademetris, Xenophon; Lin, Ben A; Bregasi, Alda; Sinusas, Albert J; Staib, Lawrence H; Duncan, James S
2014-02-01
This paper presents a dynamical appearance model based on sparse representation and dictionary learning for tracking both endocardial and epicardial contours of the left ventricle in echocardiographic sequences. Instead of learning offline spatiotemporal priors from databases, we exploit the inherent spatiotemporal coherence of individual data to constraint cardiac contour estimation. The contour tracker is initialized with a manual tracing of the first frame. It employs multiscale sparse representation of local image appearance and learns online multiscale appearance dictionaries in a boosting framework as the image sequence is segmented frame-by-frame sequentially. The weights of multiscale appearance dictionaries are optimized automatically. Our region-based level set segmentation integrates a spectrum of complementary multilevel information including intensity, multiscale local appearance, and dynamical shape prediction. The approach is validated on twenty-six 4D canine echocardiographic images acquired from both healthy and post-infarct canines. The segmentation results agree well with expert manual tracings. The ejection fraction estimates also show good agreement with manual results. Advantages of our approach are demonstrated by comparisons with a conventional pure intensity model, a registration-based contour tracker, and a state-of-the-art database-dependent offline dynamical shape model. We also demonstrate the feasibility of clinical application by applying the method to four 4D human data sets.
Sparse Representation with Spatio-Temporal Online Dictionary Learning for Efficient Video Coding.
Dai, Wenrui; Shen, Yangmei; Tang, Xin; Zou, Junni; Xiong, Hongkai; Chen, Chang Wen
2016-07-27
Classical dictionary learning methods for video coding suer from high computational complexity and interfered coding eciency by disregarding its underlying distribution. This paper proposes a spatio-temporal online dictionary learning (STOL) algorithm to speed up the convergence rate of dictionary learning with a guarantee of approximation error. The proposed algorithm incorporates stochastic gradient descents to form a dictionary of pairs of 3-D low-frequency and highfrequency spatio-temporal volumes. In each iteration of the learning process, it randomly selects one sample volume and updates the atoms of dictionary by minimizing the expected cost, rather than optimizes empirical cost over the complete training data like batch learning methods, e.g. K-SVD. Since the selected volumes are supposed to be i.i.d. samples from the underlying distribution, decomposition coecients attained from the trained dictionary are desirable for sparse representation. Theoretically, it is proved that the proposed STOL could achieve better approximation for sparse representation than K-SVD and maintain both structured sparsity and hierarchical sparsity. It is shown to outperform batch gradient descent methods (K-SVD) in the sense of convergence speed and computational complexity, and its upper bound for prediction error is asymptotically equal to the training error. With lower computational complexity, extensive experiments validate that the STOL based coding scheme achieves performance improvements than H.264/AVC or HEVC as well as existing super-resolution based methods in ratedistortion performance and visual quality.
Aspect-Aided Dynamic Non-Negative Sparse Representation-Based Microwave Image Classification.
Zhang, Xinzheng; Yang, Qiuyue; Liu, Miaomiao; Jia, Yunjian; Liu, Shujun; Li, Guojun
2016-09-02
Classification of target microwave images is an important application in much areas such as security, surveillance, etc. With respect to the task of microwave image classification, a recognition algorithm based on aspect-aided dynamic non-negative least square (ADNNLS) sparse representation is proposed. Firstly, an aspect sector is determined, the center of which is the estimated aspect angle of the testing sample. The training samples in the aspect sector are divided into active atoms and inactive atoms by smooth self-representative learning. Secondly, for each testing sample, the corresponding active atoms are selected dynamically, thereby establishing dynamic dictionary. Thirdly, the testing sample is represented with ℓ 1 -regularized non-negative sparse representation under the corresponding dynamic dictionary. Finally, the class label of the testing sample is identified by use of the minimum reconstruction error. Verification of the proposed algorithm was conducted using the Moving and Stationary Target Acquisition and Recognition (MSTAR) database which was acquired by synthetic aperture radar. Experiment results validated that the proposed approach was able to capture the local aspect characteristics of microwave images effectively, thereby improving the classification performance.
Wang, Li; Gao, Yaozong; Shi, Feng; Liao, Shu; Li, Gang; Chen, Ken Chung; Shen, Steve G. F.; Yan, Jin; Lee, Philip K. M.; Chow, Ben; Liu, Nancy X.; Xia, James J.; Shen, Dinggang
2014-04-15
Purpose: Cone-beam computed tomography (CBCT) is an increasingly utilized imaging modality for the diagnosis and treatment planning of the patients with craniomaxillofacial (CMF) deformities. Accurate segmentation of CBCT image is an essential step to generate three-dimensional (3D) models for the diagnosis and treatment planning of the patients with CMF deformities. However, due to the poor image quality, including very low signal-to-noise ratio and the widespread image artifacts such as noise, beam hardening, and inhomogeneity, it is challenging to segment the CBCT images. In this paper, the authors present a new automatic segmentation method to address these problems. Methods: To segment CBCT images, the authors propose a new method for fully automated CBCT segmentation by using patch-based sparse representation to (1) segment bony structures from the soft tissues and (2) further separate the mandible from the maxilla. Specifically, a region-specific registration strategy is first proposed to warp all the atlases to the current testing subject and then a sparse-based label propagation strategy is employed to estimate a patient-specific atlas from all aligned atlases. Finally, the patient-specific atlas is integrated into amaximum a posteriori probability-based convex segmentation framework for accurate segmentation. Results: The proposed method has been evaluated on a dataset with 15 CBCT images. The effectiveness of the proposed region-specific registration strategy and patient-specific atlas has been validated by comparing with the traditional registration strategy and population-based atlas. The experimental results show that the proposed method achieves the best segmentation accuracy by comparison with other state-of-the-art segmentation methods. Conclusions: The authors have proposed a new CBCT segmentation method by using patch-based sparse representation and convex optimization, which can achieve considerably accurate segmentation results in CBCT
Tight Graph Framelets for Sparse Diffusion MRI q-Space Representation.
Yap, Pew-Thian; Dong, Bin; Zhang, Yong; Shen, Dinggang
2016-10-01
In diffusion MRI, the outcome of estimation problems can often be improved by taking into account the correlation of diffusion-weighted images scanned with neighboring wavevectors in q-space. For this purpose, we propose in this paper to employ tight wavelet frames constructed on non-flat domains for multi-scale sparse representation of diffusion signals. This representation is well suited for signals sampled regularly or irregularly, such as on a grid or on multiple shells, in q-space. Using spectral graph theory, the frames are constructed based on quasi-affine systems (i.e., generalized dilations and shifts of a finite collection of wavelet functions) defined on graphs, which can be seen as a discrete representation of manifolds. The associated wavelet analysis and synthesis transforms can be computed efficiently and accurately without the need for explicit eigen-decomposition of the graph Laplacian, allowing scalability to very large problems. We demonstrate the effectiveness of this representation, generated using what we call tight graph framelets, in two specific applications: denoising and super-resolution in q-space using ℓ0 regularization. The associated optimization problem involves only thresholding and solving a trivial inverse problem in an iterative manner. The effectiveness of graph framelets is confirmed via evaluation using synthetic data with noncentral chi noise and real data with repeated scans.
Fernandes, Steven Lawrence; Bala, G Josemin
2016-09-01
gait recognitionare developed. Then a novel biomechanics based gait recognition is developed using Sparse Representation to generate what we term as "score 1." Further another novel technique for composite sketch matching is developed using Dictionary Matching to generate what we term as "score 2." Finally, score level fusion using Dempster Shafer and Proportional Conflict Distribution Rule Number 5 is performed. The proposed fusion approach is validated using a database containing biomechanics based gait sequences and biometric based composite sketches. From our analysis we find that a fusion of gait recognition and composite sketch matching provides excellent results for real-time human identification.
NASA Astrophysics Data System (ADS)
Fan, Wei; Cai, Gaigai; Zhu, Z. K.; Shen, Changqing; Huang, Weiguo; Shang, Li
2015-05-01
Vibration signals from a defective gearbox are often associated with important measurement information useful for gearbox fault diagnosis. The extraction of transient features from the vibration signals has always been a key issue for detecting the localized fault. In this paper, a new transient feature extraction technique is proposed for gearbox fault diagnosis based on sparse representation in wavelet basis. With the proposed method, both the impulse time and the period of transients can be effectively identified, and thus the transient features can be extracted. The effectiveness of the proposed method is verified by the simulated signals as well as the practical gearbox vibration signals. Comparison study shows that the proposed method outperforms empirical mode decomposition (EMD) in transient feature extraction.
Fast Acquisition and Reconstruction of Optical Coherence Tomography Images via Sparse Representation
Li, Shutao; McNabb, Ryan P.; Nie, Qing; Kuo, Anthony N.; Toth, Cynthia A.; Izatt, Joseph A.; Farsiu, Sina
2014-01-01
In this paper, we present a novel technique, based on compressive sensing principles, for reconstruction and enhancement of multi-dimensional image data. Our method is a major improvement and generalization of the multi-scale sparsity based tomographic denoising (MSBTD) algorithm we recently introduced for reducing speckle noise. Our new technique exhibits several advantages over MSBTD, including its capability to simultaneously reduce noise and interpolate missing data. Unlike MSBTD, our new method does not require an a priori high-quality image from the target imaging subject and thus offers the potential to shorten clinical imaging sessions. This novel image restoration method, which we termed sparsity based simultaneous denoising and interpolation (SBSDI), utilizes sparse representation dictionaries constructed from previously collected datasets. We tested the SBSDI algorithm on retinal spectral domain optical coherence tomography images captured in the clinic. Experiments showed that the SBSDI algorithm qualitatively and quantitatively outperforms other state-of-the-art methods. PMID:23846467
Secure and Robust Iris Recognition Using Random Projections and Sparse Representations.
Pillai, Jaishanker K; Patel, Vishal M; Chellappa, Rama; Ratha, Nalini K
2011-09-01
Noncontact biometrics such as face and iris have additional benefits over contact-based biometrics such as fingerprint and hand geometry. However, three important challenges need to be addressed in a noncontact biometrics-based authentication system: ability to handle unconstrained acquisition, robust and accurate matching, and privacy enhancement without compromising security. In this paper, we propose a unified framework based on random projections and sparse representations, that can simultaneously address all three issues mentioned above in relation to iris biometrics. Our proposed quality measure can handle segmentation errors and a wide variety of possible artifacts during iris acquisition. We demonstrate how the proposed approach can be easily extended to handle alignment variations and recognition from iris videos, resulting in a robust and accurate system. The proposed approach includes enhancements to privacy and security by providing ways to create cancelable iris templates. Results on public data sets show significant benefits of the proposed approach.
NASA Astrophysics Data System (ADS)
Li, Yung-Hui; Zheng, Bo-Ren; Ji, Dai-Yan; Tien, Chung-Hao; Liu, Po-Tsun
2014-09-01
Cross sensor iris matching may seriously degrade the recognition performance because of the sensor mis-match problem of iris images between the enrollment and test stage. In this paper, we propose two novel patch-based heterogeneous dictionary learning method to attack this problem. The first method applies the latest sparse representation theory while the second method tries to learn the correspondence relationship through PCA in heterogeneous patch space. Both methods learn the basic atoms in iris textures across different image sensors and build connections between them. After such connections are built, at test stage, it is possible to hallucinate (synthesize) iris images across different sensors. By matching training images with hallucinated images, the recognition rate can be successfully enhanced. The experimental results showed the satisfied results both visually and in terms of recognition rate. Experimenting with an iris database consisting of 3015 images, we show that the EER is decreased 39.4% relatively by the proposed method.
Piao, Xinglin; Zhang, Yong; Li, Tingshu; Hu, Yongli; Liu, Hao; Zhang, Ke; Ge, Yun
2016-11-03
The Received Signal Strength (RSS) fingerprint-based indoor localization is an important research topic in wireless network communications. Most current RSS fingerprint-based indoor localization methods do not explore and utilize the spatial or temporal correlation existing in fingerprint data and measurement data, which is helpful for improving localization accuracy. In this paper, we propose an RSS fingerprint-based indoor localization method by integrating the spatio-temporal constraints into the sparse representation model. The proposed model utilizes the inherent spatial correlation of fingerprint data in the fingerprint matching and uses the temporal continuity of the RSS measurement data in the localization phase. Experiments on the simulated data and the localization tests in the real scenes show that the proposed method improves the localization accuracy and stability effectively compared with state-of-the-art indoor localization methods.
Piao, Xinglin; Zhang, Yong; Li, Tingshu; Hu, Yongli; Liu, Hao; Zhang, Ke; Ge, Yun
2016-01-01
The Received Signal Strength (RSS) fingerprint-based indoor localization is an important research topic in wireless network communications. Most current RSS fingerprint-based indoor localization methods do not explore and utilize the spatial or temporal correlation existing in fingerprint data and measurement data, which is helpful for improving localization accuracy. In this paper, we propose an RSS fingerprint-based indoor localization method by integrating the spatio-temporal constraints into the sparse representation model. The proposed model utilizes the inherent spatial correlation of fingerprint data in the fingerprint matching and uses the temporal continuity of the RSS measurement data in the localization phase. Experiments on the simulated data and the localization tests in the real scenes show that the proposed method improves the localization accuracy and stability effectively compared with state-of-the-art indoor localization methods. PMID:27827882
NASA Astrophysics Data System (ADS)
Zhang, Zhaohui; Liu, Anran; Lei, Qian
2015-12-01
In this paper, we propose a method for single image super-resolution(SR). Given the training set produced from large amount of high-low resolution image patches, an over-complete joint dictionary is firstly learned from a pair of high-low resolution image feature space based on Restricted Boltzmann Machines (RBM). Then for each low resolution image patch densely extracted from an up-scaled low resolution input image , its high resolution image patch can be reconstructed based on sparse representation. Finally, the reconstructed image patches are overlapped to form a large image, and a high resolution image can be achieved by means of iterated residual image compensation. Experimental results verify the effectiveness of the proposed method.
NASA Astrophysics Data System (ADS)
Wang, Qi; Lian, Zhijie; Wang, Jianming; Chen, Qingliang; Sun, Yukuan; Li, Xiuyan; Duan, Xiaojie; Cui, Ziqiang; Wang, Huaxiang
2016-11-01
Electrical impedance tomography (EIT) reconstruction is a nonlinear and ill-posed problem. Exact reconstruction of an EIT image inverts a high dimensional mathematical model to calculate the conductivity field, which causes significant problems regarding that the computational complexity will reduce the achievable frame rate, which is considered as a major advantage of EIT imaging. The single-step method, state estimation method, and projection method were always used to accelerate reconstruction process. The basic principle of these methods is to reduce computational complexity. However, maintaining high resolution in space together with not much cost is still challenging, especially for complex conductivity distribution. This study proposes an idea to accelerate image reconstruction of EIT based on compressive sensing (CS) theory, namely, CSEIT method. The novel CSEIT method reduces the sampling rate through minimizing redundancy in measurements, so that detailed information of reconstruction is not lost. In order to obtain sparse solution, which is the prior condition of signal recovery required by CS theory, a novel image reconstruction algorithm based on patch-based sparse representation is proposed. By applying the new framework of CSEIT, the data acquisition time, or the sampling rate, is reduced by more than two times, while the accuracy of reconstruction is significantly improved.
A proximal iteration for deconvolving Poisson noisy images using sparse representations.
Dupé, François-Xavier; Fadili, Jalal M; Starck, Jean-Luc
2009-02-01
We propose an image deconvolution algorithm when the data is contaminated by Poisson noise. The image to restore is assumed to be sparsely represented in a dictionary of waveforms such as the wavelet or curvelet transforms. Our key contributions are as follows. First, we handle the Poisson noise properly by using the Anscombe variance stabilizing transform leading to a nonlinear degradation equation with additive Gaussian noise. Second, the deconvolution problem is formulated as the minimization of a convex functional with a data-fidelity term reflecting the noise properties, and a nonsmooth sparsity-promoting penalty over the image representation coefficients (e.g., l(1) -norm). An additional term is also included in the functional to ensure positivity of the restored image. Third, a fast iterative forward-backward splitting algorithm is proposed to solve the minimization problem. We derive existence and uniqueness conditions of the solution, and establish convergence of the iterative algorithm. Finally, a GCV-based model selection procedure is proposed to objectively select the regularization parameter. Experimental results are carried out to show the striking benefits gained from taking into account the Poisson statistics of the noise. These results also suggest that using sparse-domain regularization may be tractable in many deconvolution applications with Poisson noise such as astronomy and microscopy.
Epileptic Seizure Detection with Log-Euclidean Gaussian Kernel-Based Sparse Representation.
Yuan, Shasha; Zhou, Weidong; Wu, Qi; Zhang, Yanli
2016-05-01
Epileptic seizure detection plays an important role in the diagnosis of epilepsy and reducing the massive workload of reviewing electroencephalography (EEG) recordings. In this work, a novel algorithm is developed to detect seizures employing log-Euclidean Gaussian kernel-based sparse representation (SR) in long-term EEG recordings. Unlike the traditional SR for vector data in Euclidean space, the log-Euclidean Gaussian kernel-based SR framework is proposed for seizure detection in the space of the symmetric positive definite (SPD) matrices, which form a Riemannian manifold. Since the Riemannian manifold is nonlinear, the log-Euclidean Gaussian kernel function is applied to embed it into a reproducing kernel Hilbert space (RKHS) for performing SR. The EEG signals of all channels are divided into epochs and the SPD matrices representing EEG epochs are generated by covariance descriptors. Then, the testing samples are sparsely coded over the dictionary composed by training samples utilizing log-Euclidean Gaussian kernel-based SR. The classification of testing samples is achieved by computing the minimal reconstructed residuals. The proposed method is evaluated on the Freiburg EEG dataset of 21 patients and shows its notable performance on both epoch-based and event-based assessments. Moreover, this method handles multiple channels of EEG recordings synchronously which is more speedy and efficient than traditional seizure detection methods.
Yang, Su; Shi, Shixiong; Hu, Xiaobing; Wang, Minjie
2015-01-01
Spatial-temporal correlations among the data play an important role in traffic flow prediction. Correspondingly, traffic modeling and prediction based on big data analytics emerges due to the city-scale interactions among traffic flows. A new methodology based on sparse representation is proposed to reveal the spatial-temporal dependencies among traffic flows so as to simplify the correlations among traffic data for the prediction task at a given sensor. Three important findings are observed in the experiments: (1) Only traffic flows immediately prior to the present time affect the formation of current traffic flows, which implies the possibility to reduce the traditional high-order predictors into an 1-order model. (2) The spatial context relevant to a given prediction task is more complex than what is assumed to exist locally and can spread out to the whole city. (3) The spatial context varies with the target sensor undergoing prediction and enlarges with the increment of time lag for prediction. Because the scope of human mobility is subject to travel time, identifying the varying spatial context against time lag is crucial for prediction. Since sparse representation can capture the varying spatial context to adapt to the prediction task, it outperforms the traditional methods the inputs of which are confined as the data from a fixed number of nearby sensors. As the spatial-temporal context for any prediction task is fully detected from the traffic data in an automated manner, where no additional information regarding network topology is needed, it has good scalability to be applicable to large-scale networks. PMID:26496370
Yang, Su; Shi, Shixiong; Hu, Xiaobing; Wang, Minjie
2015-01-01
Spatial-temporal correlations among the data play an important role in traffic flow prediction. Correspondingly, traffic modeling and prediction based on big data analytics emerges due to the city-scale interactions among traffic flows. A new methodology based on sparse representation is proposed to reveal the spatial-temporal dependencies among traffic flows so as to simplify the correlations among traffic data for the prediction task at a given sensor. Three important findings are observed in the experiments: (1) Only traffic flows immediately prior to the present time affect the formation of current traffic flows, which implies the possibility to reduce the traditional high-order predictors into an 1-order model. (2) The spatial context relevant to a given prediction task is more complex than what is assumed to exist locally and can spread out to the whole city. (3) The spatial context varies with the target sensor undergoing prediction and enlarges with the increment of time lag for prediction. Because the scope of human mobility is subject to travel time, identifying the varying spatial context against time lag is crucial for prediction. Since sparse representation can capture the varying spatial context to adapt to the prediction task, it outperforms the traditional methods the inputs of which are confined as the data from a fixed number of nearby sensors. As the spatial-temporal context for any prediction task is fully detected from the traffic data in an automated manner, where no additional information regarding network topology is needed, it has good scalability to be applicable to large-scale networks.
Sparse Representation of Brain Aging: Extracting Covariance Patterns from Structural MRI
Su, Longfei; Wang, Lubin; Chen, Fanglin; Shen, Hui; Li, Baojuan; Hu, Dewen
2012-01-01
An enhanced understanding of how normal aging alters brain structure is urgently needed for the early diagnosis and treatment of age-related mental diseases. Structural magnetic resonance imaging (MRI) is a reliable technique used to detect age-related changes in the human brain. Currently, multivariate pattern analysis (MVPA) enables the exploration of subtle and distributed changes of data obtained from structural MRI images. In this study, a new MVPA approach based on sparse representation has been employed to investigate the anatomical covariance patterns of normal aging. Two groups of participants (group 1∶290 participants; group 2∶56 participants) were evaluated in this study. These two groups were scanned with two 1.5 T MRI machines. In the first group, we obtained the discriminative patterns using a t-test filter and sparse representation step. We were able to distinguish the young from old cohort with a very high accuracy using only a few voxels of the discriminative patterns (group 1∶98.4%; group 2∶96.4%). The experimental results showed that the selected voxels may be categorized into two components according to the two steps in the proposed method. The first component focuses on the precentral and postcentral gyri, and the caudate nucleus, which play an important role in sensorimotor tasks. The strongest volume reduction with age was observed in these clusters. The second component is mainly distributed over the cerebellum, thalamus, and right inferior frontal gyrus. These regions are not only critical nodes of the sensorimotor circuitry but also the cognitive circuitry although their volume shows a relative resilience against aging. Considering the voxels selection procedure, we suggest that the aging of the sensorimotor and cognitive brain regions identified in this study has a covarying relationship with each other. PMID:22590522
Improving Low-dose Cardiac CT Images based on 3D Sparse Representation
NASA Astrophysics Data System (ADS)
Shi, Luyao; Hu, Yining; Chen, Yang; Yin, Xindao; Shu, Huazhong; Luo, Limin; Coatrieux, Jean-Louis
2016-03-01
Cardiac computed tomography (CCT) is a reliable and accurate tool for diagnosis of coronary artery diseases and is also frequently used in surgery guidance. Low-dose scans should be considered in order to alleviate the harm to patients caused by X-ray radiation. However, low dose CT (LDCT) images tend to be degraded by quantum noise and streak artifacts. In order to improve the cardiac LDCT image quality, a 3D sparse representation-based processing (3D SR) is proposed by exploiting the sparsity and regularity of 3D anatomical features in CCT. The proposed method was evaluated by a clinical study of 14 patients. The performance of the proposed method was compared to the 2D spares representation-based processing (2D SR) and the state-of-the-art noise reduction algorithm BM4D. The visual assessment, quantitative assessment and qualitative assessment results show that the proposed approach can lead to effective noise/artifact suppression and detail preservation. Compared to the other two tested methods, 3D SR method can obtain results with image quality most close to the reference standard dose CT (SDCT) images.
Improving Low-dose Cardiac CT Images based on 3D Sparse Representation
Shi, Luyao; Hu, Yining; Chen, Yang; Yin, Xindao; Shu, Huazhong; Luo, Limin; Coatrieux, Jean-Louis
2016-01-01
Cardiac computed tomography (CCT) is a reliable and accurate tool for diagnosis of coronary artery diseases and is also frequently used in surgery guidance. Low-dose scans should be considered in order to alleviate the harm to patients caused by X-ray radiation. However, low dose CT (LDCT) images tend to be degraded by quantum noise and streak artifacts. In order to improve the cardiac LDCT image quality, a 3D sparse representation-based processing (3D SR) is proposed by exploiting the sparsity and regularity of 3D anatomical features in CCT. The proposed method was evaluated by a clinical study of 14 patients. The performance of the proposed method was compared to the 2D spares representation-based processing (2D SR) and the state-of-the-art noise reduction algorithm BM4D. The visual assessment, quantitative assessment and qualitative assessment results show that the proposed approach can lead to effective noise/artifact suppression and detail preservation. Compared to the other two tested methods, 3D SR method can obtain results with image quality most close to the reference standard dose CT (SDCT) images. PMID:26980176
Improving Low-dose Cardiac CT Images based on 3D Sparse Representation.
Shi, Luyao; Hu, Yining; Chen, Yang; Yin, Xindao; Shu, Huazhong; Luo, Limin; Coatrieux, Jean-Louis
2016-03-16
Cardiac computed tomography (CCT) is a reliable and accurate tool for diagnosis of coronary artery diseases and is also frequently used in surgery guidance. Low-dose scans should be considered in order to alleviate the harm to patients caused by X-ray radiation. However, low dose CT (LDCT) images tend to be degraded by quantum noise and streak artifacts. In order to improve the cardiac LDCT image quality, a 3D sparse representation-based processing (3D SR) is proposed by exploiting the sparsity and regularity of 3D anatomical features in CCT. The proposed method was evaluated by a clinical study of 14 patients. The performance of the proposed method was compared to the 2D spares representation-based processing (2D SR) and the state-of-the-art noise reduction algorithm BM4D. The visual assessment, quantitative assessment and qualitative assessment results show that the proposed approach can lead to effective noise/artifact suppression and detail preservation. Compared to the other two tested methods, 3D SR method can obtain results with image quality most close to the reference standard dose CT (SDCT) images.
Sparse Distributed Representation of Odors in a Large-scale Olfactory Bulb Circuit
Yu, Yuguo; McTavish, Thomas S.; Hines, Michael L.; Shepherd, Gordon M.; Valenti, Cesare; Migliore, Michele
2013-01-01
In the olfactory bulb, lateral inhibition mediated by granule cells has been suggested to modulate the timing of mitral cell firing, thereby shaping the representation of input odorants. Current experimental techniques, however, do not enable a clear study of how the mitral-granule cell network sculpts odor inputs to represent odor information spatially and temporally. To address this critical step in the neural basis of odor recognition, we built a biophysical network model of mitral and granule cells, corresponding to 1/100th of the real system in the rat, and used direct experimental imaging data of glomeruli activated by various odors. The model allows the systematic investigation and generation of testable hypotheses of the functional mechanisms underlying odor representation in the olfactory bulb circuit. Specifically, we demonstrate that lateral inhibition emerges within the olfactory bulb network through recurrent dendrodendritic synapses when constrained by a range of balanced excitatory and inhibitory conductances. We find that the spatio-temporal dynamics of lateral inhibition plays a critical role in building the glomerular-related cell clusters observed in experiments, through the modulation of synaptic weights during odor training. Lateral inhibition also mediates the development of sparse and synchronized spiking patterns of mitral cells related to odor inputs within the network, with the frequency of these synchronized spiking patterns also modulated by the sniff cycle. PMID:23555237
Learning to detect objects in images via a sparse, part-based representation.
Agarwal, Shivani; Awan, Aatif; Roth, Dan
2004-11-01
We study the problem of detecting objects in still, gray-scale images. Our primary focus is the development of a learning-based approach to the problem that makes use of a sparse, part-based representation. A vocabulary of distinctive object parts is automatically constructed from a set of sample images of the object class of interest; images are then represented using parts from this vocabulary, together with spatial relations observed among the parts. Based on this representation, a learning algorithm is used to automatically learn to detect instances of the object class in new images. The approach can be applied to any object with distinguishable parts in a relatively fixed spatial configuration; it is evaluated here on difficult sets of real-world images containing side views of cars, and is seen to successfully detect objects in varying conditions amidst background clutter and mild occlusion. In evaluating object detection approaches, several important methodological issues arise that have not been satisfactorily addressed in previous work. A secondary focus of this paper is to highlight these issues and to develop rigorous evaluation standards for the object detection problem. A critical evaluation of our approach under the proposed standards is presented.
NASA Astrophysics Data System (ADS)
Wozniak, Przemyslaw R.; Moody, D. I.; Ji, Z.; Brumby, S. P.; Brink, H.; Richards, J.; Bloom, J. S.
2013-01-01
Exponential growth in data streams and discovery power delivered by modern time-domain imaging surveys creates a pressing need for variability extraction algorithms that are both fully automated and highly reliable. The current state of the art methods based on image differencing are limited by the fact that for every real variable source the algorithm returns a large number of bogus "detections" caused by atmospheric effects and instrumental signatures coupled with imperfect image processing. Here we present a new approach to this problem inspired by recent advances in computer vision and train the machine directly on pixel data. The training data set comes from the Palomar Transient Factory survey and consists of small images centered around transient candidates with known real/bogus classification. This set of 441-dimensional vectors (21x21 pixel images) is then transformed to a linear representation using the so called dictionary, an overcomplete basis constructed separately for each class. The learning algorithm captures the fact that the intrinsic dimensionality of the input images is typically much lower than the size of the dictionary, and therefore the data vectors are well approximated with a small number of dictionary elements. This sparse representation can be used to construct informative features for any suitable machine learning classifier. In our preliminary analysis automatically extracted features approach the performance of features constructed by humans using subject domain knowledge.
Online sparse representation for remote sensing compressed-sensed video sampling
NASA Astrophysics Data System (ADS)
Wang, Jie; Liu, Kun; Li, Sheng-liang; Zhang, Li
2014-11-01
Most recently, an emerging Compressed Sensing (CS) theory has brought a major breakthrough for data acquisition and recovery. It asserts that a signal, which is highly compressible in a known basis, can be reconstructed with high probability through sampling frequency which is well below Nyquist Sampling Frequency. When applying CS to Remote Sensing (RS) Video imaging, it can directly and efficiently acquire compressed image data by randomly projecting original data to obtain linear and non-adaptive measurements. In this paper, with the help of distributed video coding scheme which is a low-complexity technique for resource limited sensors, the frames of a RS video sequence are divided into Key frames (K frames) and Non-Key frames (CS frames). In other words, the input video sequence consists of many groups of pictures (GOPs) and each GOP consists of one K frame followed by several CS frames. Both of them are measured based on block, but at different sampling rates. In this way, the major encoding computation burden will be shifted to the decoder. At the decoder, the Side Information (SI) is generated for the CS frames using traditional Motion-Compensated Interpolation (MCI) technique according to the reconstructed key frames. The over-complete dictionary is trained by dictionary learning methods based on SI. These learning methods include ICA-like, PCA, K-SVD, MOD, etc. Using these dictionaries, the CS frames could be reconstructed according to sparse-land model. In the numerical experiments, the reconstruction performance of ICA algorithm, which is often evaluated by Peak Signal-to-Noise Ratio (PSNR), has been made compared with other online sparse representation algorithms. The simulation results show its advantages in reducing reconstruction time and robustness in reconstruction performance when applying ICA algorithm to remote sensing video reconstruction.
Combining sparseness and smoothness improves classification accuracy and interpretability.
de Brecht, Matthew; Yamagishi, Noriko
2012-04-02
Sparse logistic regression (SLR) has been shown to be a useful method for decoding high-dimensional fMRI and MEG data by automatically selecting relevant feature dimensions. However, when applied to signals with high spatio-temporal correlations, SLR often over-prunes the feature space, which can result in overfitting and weight vectors that are difficult to interpret. To overcome this problem, we investigate a modification of ℓ₁-normed sparse logistic regression, called smooth sparse logistic regression (SSLR), which has a spatio-temporal "smoothing" prior that encourages weights that are close in time and space to have similar values. This causes the classifier to select spatio-temporally continuous groups of features, whereas SLR classifiers often select a scattered collection of independent features. We applied the method to both simulation data and real MEG data. We found that SSLR consistently increases classification accuracy, and produces weight vectors that are more meaningful from a neuroscientific perspective.
Javidi, Malihe; Pourreza, Hamid-Reza; Harati, Ahad
2017-02-01
Diabetic retinopathy (DR) is a major cause of visual impairment, and the analysis of retinal image can assist patients to take action earlier when it is more likely to be effective. The accurate segmentation of blood vessels in the retinal image can diagnose DR directly. In this paper, a novel scheme for blood vessel segmentation based on discriminative dictionary learning (DDL) and sparse representation has been proposed. The proposed system yields a strong representation which contains the semantic concept of the image. To extract blood vessel, two separate dictionaries, for vessel and non-vessel, capable of providing reconstructive and discriminative information of the retinal image are learned. In the test step, an unseen retinal image is divided into overlapping patches and classified to vessel and non-vessel patches. Then, a voting scheme is applied to generate the binary vessel map. The proposed vessel segmentation method can achieve the accuracy of 95% and a sensitivity of 75% in the same range of specificity 97% on two public datasets. The results show that the proposed method can achieve comparable results to existing methods and decrease false positive vessels in abnormal retinal images with pathological regions. Microaneurysm (MA) is the earliest sign of DR that appears as a small red dot on the surface of the retina. Despite several attempts to develop automated MA detection systems, it is still a challenging problem. In this paper, a method for MA detection, which is similar to our vessel segmentation approach, is proposed. In our method, a candidate detection algorithm based on the Morlet wavelet is applied to identify all possible MA candidates. In the next step, two discriminative dictionaries with the ability to distinguish MA from non-MA object are learned. These dictionaries are then used to classify the detected candidate objects. The evaluations indicate that the proposed MA detection method achieves higher average sensitivity about 2
NASA Astrophysics Data System (ADS)
Wu, Bo; Zhu, Yong; Huang, Xin; Li, Jiayi
2016-10-01
Sparse representation classification (SRC) is becoming a promising tool for hyperspectral image (HSI) classification, where the Euclidean spectral distance (ESD) is widely used to reflect the fidelity between the original and reconstructed signals. In this paper, a generalized model is proposed to extend SRC by characterizing the spectral fidelity with flexible similarity measures. To validate the flexibility, several typical similarity measures-the spectral angle similarity (SAS), spectral information divergence (SID), the structural similarity index measure (SSIM), and the ESD-are included in the generalized model. Furthermore, a general solution based on a gradient descent technique is used to solve the nonlinear optimization problem formulated by the flexible similarity measures. To test the generalized model, two actual HSIs were used, and the experimental results confirm the ability of the proposed model to accommodate the various spectral similarity measures. Performance comparisons with the ESD, SAS, SID, and SSIM criteria were also conducted, and the results consistently show the advantages of the generalized model for HSI classification in terms of overall accuracy and kappa coefficient.
Clustering-weighted SIFT-based classification method via sparse representation
NASA Astrophysics Data System (ADS)
Sun, Bo; Xu, Feng; He, Jun
2014-07-01
In recent years, sparse representation-based classification (SRC) has received significant attention due to its high recognition rate. However, the original SRC method requires a rigid alignment, which is crucial for its application. Therefore, features such as SIFT descriptors are introduced into the SRC method, resulting in an alignment-free method. However, a feature-based dictionary always contains considerable useful information for recognition. We explore the relationship of the similarity of the SIFT descriptors to multitask recognition and propose a clustering-weighted SIFT-based SRC method (CWS-SRC). The proposed approach is considerably more suitable for multitask recognition with sufficient samples. Using two public face databases (AR and Yale face) and a self-built car-model database, the performance of the proposed method is evaluated and compared to that of the SRC, SIFT matching, and MKD-SRC methods. Experimental results indicate that the proposed method exhibits better performance in the alignment-free scenario with sufficient samples.
3D Face Recognition Based on Multiple Keypoint Descriptors and Sparse Representation
Zhang, Lin; Ding, Zhixuan; Li, Hongyu; Shen, Ying; Lu, Jianwei
2014-01-01
Recent years have witnessed a growing interest in developing methods for 3D face recognition. However, 3D scans often suffer from the problems of missing parts, large facial expressions, and occlusions. To be useful in real-world applications, a 3D face recognition approach should be able to handle these challenges. In this paper, we propose a novel general approach to deal with the 3D face recognition problem by making use of multiple keypoint descriptors (MKD) and the sparse representation-based classification (SRC). We call the proposed method 3DMKDSRC for short. Specifically, with 3DMKDSRC, each 3D face scan is represented as a set of descriptor vectors extracted from keypoints by meshSIFT. Descriptor vectors of gallery samples form the gallery dictionary. Given a probe 3D face scan, its descriptors are extracted at first and then its identity can be determined by using a multitask SRC. The proposed 3DMKDSRC approach does not require the pre-alignment between two face scans and is quite robust to the problems of missing data, occlusions and expressions. Its superiority over the other leading 3D face recognition schemes has been corroborated by extensive experiments conducted on three benchmark databases, Bosphorus, GavabDB, and FRGC2.0. The Matlab source code for 3DMKDSRC and the related evaluation results are publicly available at http://sse.tongji.edu.cn/linzhang/3dmkdsrcface/3dmkdsrc.htm. PMID:24940876
Dim moving target tracking algorithm based on particle discriminative sparse representation
NASA Astrophysics Data System (ADS)
Li, Zhengzhou; Li, Jianing; Ge, Fengzeng; Shao, Wanxing; Liu, Bing; Jin, Gang
2016-03-01
The small dim moving target usually submerged in strong noise, and its motion observability is debased by numerous false alarms for low signal-to-noise ratio (SNR). A target tracking algorithm based on particle filter and discriminative sparse representation is proposed in this paper to cope with the uncertainty of dim moving target tracking. The weight of every particle is the crucial factor to ensuring the accuracy of dim target tracking for particle filter (PF) that can achieve excellent performance even under the situation of non-linear and non-Gaussian motion. In discriminative over-complete dictionary constructed according to image sequence, the target dictionary describes target signal and the background dictionary embeds background clutter. The difference between target particle and background particle is enhanced to a great extent, and the weight of every particle is then measured by means of the residual after reconstruction using the prescribed number of target atoms and their corresponding coefficients. The movement state of dim moving target is then estimated and finally tracked by these weighted particles. Meanwhile, the subspace of over-complete dictionary is updated online by the stochastic estimation algorithm. Some experiments are induced and the experimental results show the proposed algorithm could improve the performance of moving target tracking by enhancing the consistency between the posteriori probability distribution and the moving target state.
NASA Astrophysics Data System (ADS)
Zhang, Min; Zhou, Xiangrong; Goshima, Satoshi; Chen, Huayue; Muramatsu, Chisako; Hara, Takeshi; Yokoyama, Ryujiro; Kanematsu, Masayuki; Fujita, Hiroshi
2013-03-01
In this paper, we present a texture classification method based on texton learned via sparse representation (SR) with new feature histogram maps in the classification of emphysema. First, an overcomplete dictionary of textons is learned via KSVD learning on every class image patches in the training dataset. In this stage, high-pass filter is introduced to exclude patches in smooth area to speed up the dictionary learning process. Second, 3D joint-SR coefficients and intensity histograms of the test images are used for characterizing regions of interest (ROIs) instead of conventional feature histograms constructed from SR coefficients of the test images over the dictionary. Classification is then performed using a classifier with distance as a histogram dissimilarity measure. Four hundreds and seventy annotated ROIs extracted from 14 test subjects, including 6 paraseptal emphysema (PSE) subjects, 5 centrilobular emphysema (CLE) subjects and 3 panlobular emphysema (PLE) subjects, are used to evaluate the effectiveness and robustness of the proposed method. The proposed method is tested on 167 PSE, 240 CLE and 63 PLE ROIs consisting of mild, moderate and severe pulmonary emphysema. The accuracy of the proposed system is around 74%, 88% and 89% for PSE, CLE and PLE, respectively.
NASA Astrophysics Data System (ADS)
Zhang, Min; Zhou, Xiangrong; Goshima, Satoshi; Chen, Huayue; Muramatsu, Chisako; Hara, Takeshi; Yokoyama, Ryojiro; Kanematsu, Masayuki; Fujita, Hiroshi
2012-03-01
We aim at using a new texton based texture classification method in the classification of pulmonary emphysema in computed tomography (CT) images of the lungs. Different from conventional computer-aided diagnosis (CAD) pulmonary emphysema classification methods, in this paper, firstly, the dictionary of texton is learned via applying sparse representation(SR) to image patches in the training dataset. Then the SR coefficients of the test images over the dictionary are used to construct the histograms for texture presentations. Finally, classification is performed by using a nearest neighbor classifier with a histogram dissimilarity measure as distance. The proposed approach is tested on 3840 annotated regions of interest consisting of normal tissue and mild, moderate and severe pulmonary emphysema of three subtypes. The performance of the proposed system, with an accuracy of about 88%, is comparably higher than state of the art method based on the basic rotation invariant local binary pattern histograms and the texture classification method based on texton learning by k-means, which performs almost the best among other approaches in the literature.
Yuan, Yuan; Lin, Jianzhe; Wang, Qi
2016-12-01
Hyperspectral image (HSI) classification is a crucial issue in remote sensing. Accurate classification benefits a large number of applications such as land use analysis and marine resource utilization. But high data correlation brings difficulty to reliable classification, especially for HSI with abundant spectral information. Furthermore, the traditional methods often fail to well consider the spatial coherency of HSI that also limits the classification performance. To address these inherent obstacles, a novel spectral-spatial classification scheme is proposed in this paper. The proposed method mainly focuses on multitask joint sparse representation (MJSR) and a stepwise Markov random filed framework, which are claimed to be two main contributions in this procedure. First, the MJSR not only reduces the spectral redundancy, but also retains necessary correlation in spectral field during classification. Second, the stepwise optimization further explores the spatial correlation that significantly enhances the classification accuracy and robustness. As far as several universal quality evaluation indexes are concerned, the experimental results on Indian Pines and Pavia University demonstrate the superiority of our method compared with the state-of-the-art competitors.
Yu, Kai; Yin, Ming; Luo, Ji-An; Wang, Yingguan; Bao, Ming; Hu, Yu-Hen; Wang, Zhi
2016-01-01
A compressive sensing joint sparse representation direction of arrival estimation (CSJSR-DoA) approach is proposed for wireless sensor array networks (WSAN). By exploiting the joint spatial and spectral correlations of acoustic sensor array data, the CSJSR-DoA approach provides reliable DoA estimation using randomly-sampled acoustic sensor data. Since random sampling is performed at remote sensor arrays, less data need to be transmitted over lossy wireless channels to the fusion center (FC), and the expensive source coding operation at sensor nodes can be avoided. To investigate the spatial sparsity, an upper bound of the coherence of incoming sensor signals is derived assuming a linear sensor array configuration. This bound provides a theoretical constraint on the angular separation of acoustic sources to ensure the spatial sparsity of the received acoustic sensor array signals. The Crame´r–Rao bound of the CSJSR-DoA estimator that quantifies the theoretical DoA estimation performance is also derived. The potential performance of the CSJSR-DoA approach is validated using both simulations and field experiments on a prototype WSAN platform. Compared to existing compressive sensing-based DoA estimation methods, the CSJSR-DoA approach shows significant performance improvement. PMID:27223287
NASA Astrophysics Data System (ADS)
Moonon, Altan-Ulzii; Hu, Jianwen; Li, Shutao
2015-12-01
The remote sensing image fusion is an important preprocessing technique in remote sensing image processing. In this paper, a remote sensing image fusion method based on the nonsubsampled shearlet transform (NSST) with sparse representation (SR) is proposed. Firstly, the low resolution multispectral (MS) image is upsampled and color space is transformed from Red-Green-Blue (RGB) to Intensity-Hue-Saturation (IHS). Then, the high resolution panchromatic (PAN) image and intensity component of MS image are decomposed by NSST to high and low frequency coefficients. The low frequency coefficients of PAN and the intensity component are fused by the SR with the learned dictionary. The high frequency coefficients of intensity component and PAN image are fused by local energy based fusion rule. Finally, the fused result is obtained by performing inverse NSST and inverse IHS transform. The experimental results on IKONOS and QuickBird satellites demonstrate that the proposed method provides better spectral quality and superior spatial information in the fused image than other remote sensing image fusion methods both in visual effect and object evaluation.
Yu, Kai; Yin, Ming; Luo, Ji-An; Wang, Yingguan; Bao, Ming; Hu, Yu-Hen; Wang, Zhi
2016-05-23
A compressive sensing joint sparse representation direction of arrival estimation (CSJSR-DoA) approach is proposed for wireless sensor array networks (WSAN). By exploiting the joint spatial and spectral correlations of acoustic sensor array data, the CSJSR-DoA approach provides reliable DoA estimation using randomly-sampled acoustic sensor data. Since random sampling is performed at remote sensor arrays, less data need to be transmitted over lossy wireless channels to the fusion center (FC), and the expensive source coding operation at sensor nodes can be avoided. To investigate the spatial sparsity, an upper bound of the coherence of incoming sensor signals is derived assuming a linear sensor array configuration. This bound provides a theoretical constraint on the angular separation of acoustic sources to ensure the spatial sparsity of the received acoustic sensor array signals. The Cram e ´ r-Rao bound of the CSJSR-DoA estimator that quantifies the theoretical DoA estimation performance is also derived. The potential performance of the CSJSR-DoA approach is validated using both simulations and field experiments on a prototype WSAN platform. Compared to existing compressive sensing-based DoA estimation methods, the CSJSR-DoA approach shows significant performance improvement.
Sparse representation and Bayesian detection of genome copy number alterations from microarray data
Pique-Regi, Roger; Monso-Varona, Jordi; Ortega, Antonio; Seeger, Robert C.; Triche, Timothy J.; Asgharzadeh, Shahab
2008-01-01
Motivation: Genomic instability in cancer leads to abnormal genome copy number alterations (CNA) that are associated with the development and behavior of tumors. Advances in microarray technology have allowed for greater resolution in detection of DNA copy number changes (amplifications or deletions) across the genome. However, the increase in number of measured signals and accompanying noise from the array probes present a challenge in accurate and fast identification of breakpoints that define CNA. This article proposes a novel detection technique that exploits the use of piece wise constant (PWC) vectors to represent genome copy number and sparse Bayesian learning (SBL) to detect CNA breakpoints. Methods: First, a compact linear algebra representation for the genome copy number is developed from normalized probe intensities. Second, SBL is applied and optimized to infer locations where copy number changes occur. Third, a backward elimination (BE) procedure is used to rank the inferred breakpoints; and a cut-off point can be efficiently adjusted in this procedure to control for the false discovery rate (FDR). Results: The performance of our algorithm is evaluated using simulated and real genome datasets and compared to other existing techniques. Our approach achieves the highest accuracy and lowest FDR while improving computational speed by several orders of magnitude. The proposed algorithm has been developed into a free standing software application (GADA, Genome Alteration Detection Algorithm). Availability: http://biron.usc.edu/~piquereg/GADA Contact: jpei@chop.swmed.edu and rpique@ieee.org Supplementary information: Supplementary data are available at Bioinformatics online. PMID:18203770
3D face recognition based on multiple keypoint descriptors and sparse representation.
Zhang, Lin; Ding, Zhixuan; Li, Hongyu; Shen, Ying; Lu, Jianwei
2014-01-01
Recent years have witnessed a growing interest in developing methods for 3D face recognition. However, 3D scans often suffer from the problems of missing parts, large facial expressions, and occlusions. To be useful in real-world applications, a 3D face recognition approach should be able to handle these challenges. In this paper, we propose a novel general approach to deal with the 3D face recognition problem by making use of multiple keypoint descriptors (MKD) and the sparse representation-based classification (SRC). We call the proposed method 3DMKDSRC for short. Specifically, with 3DMKDSRC, each 3D face scan is represented as a set of descriptor vectors extracted from keypoints by meshSIFT. Descriptor vectors of gallery samples form the gallery dictionary. Given a probe 3D face scan, its descriptors are extracted at first and then its identity can be determined by using a multitask SRC. The proposed 3DMKDSRC approach does not require the pre-alignment between two face scans and is quite robust to the problems of missing data, occlusions and expressions. Its superiority over the other leading 3D face recognition schemes has been corroborated by extensive experiments conducted on three benchmark databases, Bosphorus, GavabDB, and FRGC2.0. The Matlab source code for 3DMKDSRC and the related evaluation results are publicly available at http://sse.tongji.edu.cn/linzhang/3dmkdsrcface/3dmkdsrc.htm.
NASA Astrophysics Data System (ADS)
Choi, Jae-Seok; Bae, Sung-Ho; Kim, Munchurl
2015-03-01
In recent years, perceptually-driven super-resolution (SR) methods have been proposed to lower computational complexity. Furthermore, sparse representation based super-resolution is known to produce competitive high-resolution images with lower computational costs compared to other SR methods. Nevertheless, super-resolution is still difficult to be implemented with substantially low processing power for real-time applications. In order to speed up the processing time of SR, much effort has been made with efficient methods, which selectively incorporate elaborate computation algorithms for perceptually sensitive image regions based on a metric, such as just noticeable distortion (JND). Inspired by the previous works, we first propose a novel fast super-resolution method with sparse representation, which incorporates a no-reference just noticeable blur (JNB) metric. That is, the proposed fast super-resolution method efficiently generates super-resolution images by selectively applying a sparse representation method for perceptually sensitive image areas which are detected based on the JNB metric. Experimental results show that our JNB-based fast super-resolution method is about 4 times faster than a non-perceptual sparse representation based SR method for 256× 256 test LR images. Compared to a JND-based SR method, the proposed fast JNB-based SR method is about 3 times faster, with approximately 0.1 dB higher PSNR and a slightly higher SSIM value in average. This indicates that our proposed perceptual JNB-based SR method generates high-quality SR images with much lower computational costs, opening a new possibility for real-time hardware implementations.
Generating virtual training samples for sparse representation of face images and face recognition
NASA Astrophysics Data System (ADS)
Du, Yong; Wang, Yu
2016-03-01
There are many challenges in face recognition. In real-world scenes, images of the same face vary with changing illuminations, different expressions and poses, multiform ornaments, or even altered mental status. Limited available training samples cannot convey these possible changes in the training phase sufficiently, and this has become one of the restrictions to improve the face recognition accuracy. In this article, we view the multiplication of two images of the face as a virtual face image to expand the training set and devise a representation-based method to perform face recognition. The generated virtual samples really reflect some possible appearance and pose variations of the face. By multiplying a training sample with another sample from the same subject, we can strengthen the facial contour feature and greatly suppress the noise. Thus, more human essential information is retained. Also, uncertainty of the training data is simultaneously reduced with the increase of the training samples, which is beneficial for the training phase. The devised representation-based classifier uses both the original and new generated samples to perform the classification. In the classification phase, we first determine K nearest training samples for the current test sample by calculating the Euclidean distances between the test sample and training samples. Then, a linear combination of these selected training samples is used to represent the test sample, and the representation result is used to classify the test sample. The experimental results show that the proposed method outperforms some state-of-the-art face recognition methods.
Ren, Yudan; Fang, Jun; Lv, Jinglei; Hu, Xintao; Guo, Cong Christine; Guo, Lei; Xu, Jiansong; Potenza, Marc N; Liu, Tianming
2016-10-04
Assessing functional brain activation patterns in neuropsychiatric disorders such as cocaine dependence (CD) or pathological gambling (PG) under naturalistic stimuli has received rising interest in recent years. In this paper, we propose and apply a novel group-wise sparse representation framework to assess differences in neural responses to naturalistic stimuli across multiple groups of participants (healthy control, cocaine dependence, pathological gambling). Specifically, natural stimulus fMRI (N-fMRI) signals from all three groups of subjects are aggregated into a big data matrix, which is then decomposed into a common signal basis dictionary and associated weight coefficient matrices via an effective online dictionary learning and sparse coding method. The coefficient matrices associated with each common dictionary atom are statistically assessed for each group separately. With the inter-group comparisons based on the group-wise correspondence established by the common dictionary, our experimental results demonstrated that the group-wise sparse coding and representation strategy can effectively and specifically detect brain networks/regions affected by different pathological conditions of the brain under naturalistic stimuli.
A sparse Bayesian representation for super-resolution of cardiac MR images.
Velasco, Nelson F; Rueda, Andrea; Santa Marta, Cristina; Romero, Eduardo
2017-02-01
High-quality cardiac magnetic resonance (CMR) images can be hardly obtained when intrinsic noise sources are present, namely heart and breathing movements. Yet heart images may be acquired in real time, the image quality is really limited and most sequences use ECG gating to capture images at each stage of the cardiac cycle during several heart beats. This paper presents a novel super-resolution algorithm that improves the cardiac image quality using a sparse Bayesian approach. The high-resolution version of the cardiac image is constructed by combining the information of the low-resolution series -observations from different non-orthogonal series composed of anisotropic voxels - with a prior distribution of the high-resolution local coefficients that enforces sparsity. In addition, a global prior, extracted from the observed data, regularizes the solution. Quantitative and qualitative validations were performed in synthetic and real images w.r.t to a baseline, showing an average increment between 2.8 and 3.2 dB in the Peak Signal-to-Noise Ratio (PSNR), between 1.8% and 2.6% in the Structural Similarity Index (SSIM) and 2.% to 4% in quality assessment (IL-NIQE). The obtained results demonstrated that the proposed method is able to accurately reconstruct a cardiac image, recovering the original shape with less artifacts and low noise.
Hyperspectral Image Super-resolution via Non-negative Structured Sparse Representation.
Dong, Weisheng; Fu, Fazuo; Shi, Guangming; Cao, Xun; Wu, Jinjian; Li, Guangyu; Li, Xin
2016-03-22
Hyperspectral imaging has many applications from agriculture and astronomy to surveillance and mineralogy. However, it is often challenging to obtain High-resolution (HR) hyperspectral images using existing hyperspectral imaging techniques due to various hardware limitations. In this paper, we propose a new Hyperspectral image super-resolution method from a low-resolution (LR) image and a HR reference image of the same scene. The estimation of the HR hyperspectral image is formulated as a joint estimation of the hyperspectral dictionary and the sparse codes based on the prior knowledge of the spatialspectral sparsity of the hyperspectral image. The hyperspectral dictionary representing prototype reflectance spectra vectors of the scene is first learned from the input LR image. Specifically, an efficient non-negative dictionary learning algorithm using the block-coordinate descent optimization technique is proposed. Then, sparse codes of the desired HR hyperspectral image with respect to learned hyperspectral basis are estimated from the pair of LR and HR reference images. To improve the accuracy of non-negtative sparse coding, a clustering-based structured sparse coding method is proposed to exploit the spatial correlation among the learned sparse codes. Experimental results on both public datasets and real LR hypspectral images suggest that the proposed method substantially outperforms several existing HR hyperspectral image recovery techniques in the literature in terms of both objective quality metrics and computational efficiency.
Hyperspectral Image Super-Resolution via Non-Negative Structured Sparse Representation.
Dong, Weisheng; Fu, Fazuo; Shi, Guangming; Cao, Xun; Wu, Jinjian; Li, Guangyu; Li, Guangyu
2016-05-01
Hyperspectral imaging has many applications from agriculture and astronomy to surveillance and mineralogy. However, it is often challenging to obtain high-resolution (HR) hyperspectral images using existing hyperspectral imaging techniques due to various hardware limitations. In this paper, we propose a new hyperspectral image super-resolution method from a low-resolution (LR) image and a HR reference image of the same scene. The estimation of the HR hyperspectral image is formulated as a joint estimation of the hyperspectral dictionary and the sparse codes based on the prior knowledge of the spatial-spectral sparsity of the hyperspectral image. The hyperspectral dictionary representing prototype reflectance spectra vectors of the scene is first learned from the input LR image. Specifically, an efficient non-negative dictionary learning algorithm using the block-coordinate descent optimization technique is proposed. Then, the sparse codes of the desired HR hyperspectral image with respect to learned hyperspectral basis are estimated from the pair of LR and HR reference images. To improve the accuracy of non-negative sparse coding, a clustering-based structured sparse coding method is proposed to exploit the spatial correlation among the learned sparse codes. The experimental results on both public datasets and real LR hypspectral images suggest that the proposed method substantially outperforms several existing HR hyperspectral image recovery techniques in the literature in terms of both objective quality metrics and computational efficiency.
Paiton, Dylan M.; Kenyon, Garrett T.; Brumby, Steven P.; Schultz, Peter F.; George, John S.
2015-07-28
An approach to detecting objects in an image dataset may combine texture/color detection, shape/contour detection, and/or motion detection using sparse, generative, hierarchical models with lateral and top-down connections. A first independent representation of objects in an image dataset may be produced using a color/texture detection algorithm. A second independent representation of objects in the image dataset may be produced using a shape/contour detection algorithm. A third independent representation of objects in the image dataset may be produced using a motion detection algorithm. The first, second, and third independent representations may then be combined into a single coherent output using a combinatorial algorithm.
Paiton, Dylan M.; Kenyon, Garrett T.; Brumby, Steven P.; Schultz, Peter F.; George, John S.
2016-10-25
An approach to detecting objects in an image dataset may combine texture/color detection, shape/contour detection, and/or motion detection using sparse, generative, hierarchical models with lateral and top-down connections. A first independent representation of objects in an image dataset may be produced using a color/texture detection algorithm. A second independent representation of objects in the image dataset may be produced using a shape/contour detection algorithm. A third independent representation of objects in the image dataset may be produced using a motion detection algorithm. The first, second, and third independent representations may then be combined into a single coherent output using a combinatorial algorithm.
Wang, Li; Shi, Feng; Gao, Yaozong; Li, Gang; Gilmore, John H; Lin, Weili; Shen, Dinggang
2014-04-01
Segmentation of infant brain MR images is challenging due to poor spatial resolution, severe partial volume effect, and the ongoing maturation and myelination processes. During the first year of life, the brain image contrast between white and gray matters undergoes dramatic changes. In particular, the image contrast inverses around 6-8months of age, where the white and gray matter tissues are isointense in T1 and T2 weighted images and hence exhibit the extremely low tissue contrast, posing significant challenges for automated segmentation. In this paper, we propose a general framework that adopts sparse representation to fuse the multi-modality image information and further incorporate the anatomical constraints for brain tissue segmentation. Specifically, we first derive an initial segmentation from a library of aligned images with ground-truth segmentations by using sparse representation in a patch-based fashion for the multi-modality T1, T2 and FA images. The segmentation result is further iteratively refined by integration of the anatomical constraint. The proposed method was evaluated on 22 infant brain MR images acquired at around 6months of age by using a leave-one-out cross-validation, as well as other 10 unseen testing subjects. Our method achieved a high accuracy for the Dice ratios that measure the volume overlap between automated and manual segmentations, i.e., 0.889±0.008 for white matter and 0.870±0.006 for gray matter.
Zhang, Shu; Li, Xiang; Lv, Jinglei; Jiang, Xi; Guo, Lei; Liu, Tianming
2016-03-01
A relatively underexplored question in fMRI is whether there are intrinsic differences in terms of signal composition patterns that can effectively characterize and differentiate task-based or resting state fMRI (tfMRI or rsfMRI) signals. In this paper, we propose a novel two-stage sparse representation framework to examine the fundamental difference between tfMRI and rsfMRI signals. Specifically, in the first stage, the whole-brain tfMRI or rsfMRI signals of each subject were composed into a big data matrix, which was then factorized into a subject-specific dictionary matrix and a weight coefficient matrix for sparse representation. In the second stage, all of the dictionary matrices from both tfMRI/rsfMRI data across multiple subjects were composed into another big data-matrix, which was further sparsely represented by a cross-subjects common dictionary and a weight matrix. This framework has been applied on the recently publicly released Human Connectome Project (HCP) fMRI data and experimental results revealed that there are distinctive and descriptive atoms in the cross-subjects common dictionary that can effectively characterize and differentiate tfMRI and rsfMRI signals, achieving 100% classification accuracy. Moreover, our methods and results can be meaningfully interpreted, e.g., the well-known default mode network (DMN) activities can be recovered from the very noisy and heterogeneous aggregated big-data of tfMRI and rsfMRI signals across all subjects in HCP Q1 release.
NASA Astrophysics Data System (ADS)
Mahrooghy, Majid; Ashraf, Ahmed B.; Daye, Dania; Mies, Carolyn; Rosen, Mark; Feldman, Michael; Kontos, Despina
2014-03-01
We evaluate the prognostic value of sparse representation-based features by applying the K-SVD algorithm on multiparametric kinetic, textural, and morphologic features in breast dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). K-SVD is an iterative dimensionality reduction method that optimally reduces the initial feature space by updating the dictionary columns jointly with the sparse representation coefficients. Therefore, by using K-SVD, we not only provide sparse representation of the features and condense the information in a few coefficients but also we reduce the dimensionality. The extracted K-SVD features are evaluated by a machine learning algorithm including a logistic regression classifier for the task of classifying high versus low breast cancer recurrence risk as determined by a validated gene expression assay. The features are evaluated using ROC curve analysis and leave one-out cross validation for different sparse representation and dimensionality reduction numbers. Optimal sparse representation is obtained when the number of dictionary elements is 4 (K=4) and maximum non-zero coefficients is 2 (L=2). We compare K-SVD with ANOVA based feature selection for the same prognostic features. The ROC results show that the AUC of the K-SVD based (K=4, L=2), the ANOVA based, and the original features (i.e., no dimensionality reduction) are 0.78, 0.71. and 0.68, respectively. From the results, it can be inferred that by using sparse representation of the originally extracted multi-parametric, high-dimensional data, we can condense the information on a few coefficients with the highest predictive value. In addition, the dimensionality reduction introduced by K-SVD can prevent models from over-fitting.
Overcoming the Curse of Dimension: Methods Based on Sparse Representation and Adaptive Sampling
2011-02-28
carried out mainly by him, together with our joint post-doc Haijun Yu. Please refer to his report for the progress made in this direction. 3 Exploring...multiscale modeling using sparse representation”, Comm. Comp. Phys., 4(5), pp. 1025–1033 (2008). [3] X. Zhou and W. Ren and W. E, “Adaptive minimum...action method for the study of rare events”, J. Chem. Phys., 128, 10, 2008. [4] X. Wan, X. Zhou and W. E, “Noise-induced transitions in the Kuramoto-Sivashinsky equation”, preprint, submitted. 4
NASA Astrophysics Data System (ADS)
Yang, Yongchao; Nagarajaiah, Satish
2016-06-01
Randomly missing data of structural vibration responses time history often occurs in structural dynamics and health monitoring. For example, structural vibration responses are often corrupted by outliers or erroneous measurements due to sensor malfunction; in wireless sensing platforms, data loss during wireless communication is a common issue. Besides, to alleviate the wireless data sampling or communication burden, certain accounts of data are often discarded during sampling or before transmission. In these and other applications, recovery of the randomly missing structural vibration responses from the available, incomplete data, is essential for system identification and structural health monitoring; it is an ill-posed inverse problem, however. This paper explicitly harnesses the data structure itself-of the structural vibration responses-to address this (inverse) problem. What is relevant is an empirical, but often practically true, observation, that is, typically there are only few modes active in the structural vibration responses; hence a sparse representation (in frequency domain) of the single-channel data vector, or, a low-rank structure (by singular value decomposition) of the multi-channel data matrix. Exploiting such prior knowledge of data structure (intra-channel sparse or inter-channel low-rank), the new theories of ℓ1-minimization sparse recovery and nuclear-norm-minimization low-rank matrix completion enable recovery of the randomly missing or corrupted structural vibration response data. The performance of these two alternatives, in terms of recovery accuracy and computational time under different data missing rates, is investigated on a few structural vibration response data sets-the seismic responses of the super high-rise Canton Tower and the structural health monitoring accelerations of a real large-scale cable-stayed bridge. Encouraging results are obtained and the applicability and limitation of the presented methods are discussed.
A tight and explicit representation of Q in sparse QR factorization
Ng, E.G.; Peyton, B.W.
1992-05-01
In QR factorization of a sparse m{times}n matrix A (m {ge} n) the orthogonal factor Q is often stored implicitly as a lower trapezoidal matrix H known as the Householder matrix. This paper presents a simple characterization of the row structure of Q, which could be used as the basis for a sparse data structure that can store Q explicitly. The new characterization is a simple extension of a well known row-oriented characterization of the structure of H. Hare, Johnson, Olesky, and van den Driessche have recently provided a complete sparsity analysis of the QR factorization. Let U be the matrix consisting of the first n columns of Q. Using results from, we show that the data structures for H and U resulting from our characterizations are tight when A is a strong Hall matrix. We also show that H and the lower trapezoidal part of U have the same sparsity characterization when A is strong Hall. We then show that this characterization can be extended to any weak Hall matrix that has been permuted into block upper triangular form. Finally, we show that permuting to block triangular form never increases the fill incurred during the factorization.
NASA Astrophysics Data System (ADS)
Testa, D.; Carfantan, H.; Albergante, M.; Blanchard, P.; Bourguignon, S.; Fasoli, A.; Goodyear, A.; Klein, A.; Lister, J. B.; Panis, T.; Contributors, JET
2016-12-01
Efficient, real-time and automated data analysis is one of the key elements for achieving scientific success in complex engineering and physical systems, two examples of which include the JET and ITER tokamaks. One problem which is common to these fields is the determination of the pulsation modes from an irregularly sampled time series. To this end, there are a wealth of signal processing techniques that are being applied to post-pulse and real-time data analysis in such complex systems. Here, we wish to present a review of the applications of a method based on the sparse representation of signals, using examples of the synergies that can be exploited when combining ideas and methods from very different fields, such as astronomy, astrophysics and thermonuclear fusion plasmas. Examples of this work in astronomy and astrophysics are the analysis of pulsation modes in various classes of stars and the orbit determination software of the Pioneer spacecraft. Two examples of this work in thermonuclear fusion plasmas include the detection of magneto-hydrodynamic instabilities, which is now performed routinely in JET in real-time on a sub-millisecond time scale, and the studies leading to the optimization of the magnetic diagnostic system in ITER and TCV. These questions have been solved by formulating them as inverse problems, despite the fact that these applicative frameworks are extremely different from the classical use of sparse representations, from both the theoretical and computational point of view. The requirements, prospects and ideas for the signal processing and real-time data analysis applications of this method to the routine operation of ITER will also be discussed. Finally, a very recent development has been the attempt to apply this method to the deconvolution of the measurement of electric potential performed during a ground-based survey of a proto-Villanovian necropolis in central Italy.
Sparse Representation and Dictionary Learning as Feature Extraction in Vessel Imagery
2014-12-01
rotated images have all the ships pointing up, the cropped images have some excess background removed, and the resized images are all 300× 150 pixels...sets using either BOW or ScSPM image representations. BOW K = 1000 K = 2000 Original Data 76.0 79.4 Rotated 91.5 91.5 Cropped 92.3 92.8 Resized 94.0...94.8 ScSPM K = 1024 K = 2048 Original Data 72.0 73.3 Rotated 90.3 90.0 Cropped 93.9 94.1 Resized 95.0 94.3 4. CONCLUSION This report contains results
Concept Abstractness and the Representation of Noun-Noun Combinations
ERIC Educational Resources Information Center
Xu, Xu; Paulson, Lisa
2013-01-01
Research on noun-noun combinations has been largely focusing on concrete concepts. Three experiments examined the role of concept abstractness in the representation of noun-noun combinations. In Experiment 1, participants provided written interpretations for phrases constituted by nouns of varying degrees of abstractness. Interpretive focus (the…
Combination of direct matching and collaborative representation for face recognition
NASA Astrophysics Data System (ADS)
Zhang, Chongyang
2013-06-01
It has been proved that representation-based classification (RBC) can achieve high accuracy in face recognition. However, conventional RBC has a very high computational cost. Collaborative representation proposed in [1] not only has the advantages of RBC but also is computationally very efficient. In this paper, a combination of direct matching of images and collaborative representation is proposed for face recognition. Experimental results show that the proposed method can always classify more accurately than collaborative representation! The underlying reason is that direct matching of images and collaborative representation use different ways to calculate the dissimilarity between the test sample and training sample. As a result, the score obtained using direct matching of images is very complementary to the score obtained using collaborative representation. Actually, the analysis shows that the matching scores generated from direct matching of images and collaborative representation always have a low correlation. This allows the proposed method to exploit more information for face recognition and to produce a better result.
NASA Astrophysics Data System (ADS)
Diaz-Hernandez, R.; Ortiz-Esquivel, A.; Peregrina-Barreto, H.; Altamirano-Robles, L.; Gonzalez-Bernal, J.
2016-06-01
The observation of celestial objects in the sky is a practice that helps astronomers to understand the way in which the Universe is structured. However, due to the large number of observed objects with modern telescopes, the analysis of these by hand is a difficult task. An important part in galaxy research is the morphological structure classification based on the Hubble sequence. In this research, we present an approach to solve the morphological galaxy classification problem in an automatic way by using the Sparse Representation technique and dictionary learning with K-SVD. For the tests in this work, we use a database of galaxies extracted from the Principal Galaxy Catalog (PGC) and the APM Equatorial Catalogue of Galaxies obtaining a total of 2403 useful galaxies. In order to represent each galaxy frame, we propose to calculate a set of 20 features such as Hu's invariant moments, galaxy nucleus eccentricity, gabor galaxy ratio and some other features commonly used in galaxy classification. A stage of feature relevance analysis was performed using Relief-f in order to determine which are the best parameters for the classification tests using 2, 3, 4, 5, 6 and 7 galaxy classes making signal vectors of different length values with the most important features. For the classification task, we use a 20-random cross-validation technique to evaluate classification accuracy with all signal sets achieving a score of 82.27 % for 2 galaxy classes and up to 44.27 % for 7 galaxy classes.
Sparse Representation of Multimodality Sensing Databases for Data Mining and Retrieval
2015-04-09
SUPPLEMENTARY NOTES 12. DISTRIBUTION AVAILIBILITY STATEMENT 6. AUTHORS 7. PERFORMING ORGANIZATION NAMES AND ADDRESSES 15. SUBJECT TERMS b. ABSTRACT 2. REPORT...SPONSOR/MONITOR’S ACRONYM(S) ARO 8. PERFORMING ORGANIZATION REPORT NUMBER 19a. NAME OF RESPONSIBLE PERSON 19b. TELEPHONE NUMBER Alfred Hero Alfred O...robustness to noise and other distortions. Experimental validation will be performed by a combination of simulation and experiment on multimodality
NASA Astrophysics Data System (ADS)
Rowland, J. C.; Moody, D. I.; Brumby, S.; Gangodagamage, C.
2012-12-01
Techniques for automated feature extraction, including neuroscience-inspired machine vision, are of great interest for landscape characterization and change detection in support of global climate change science and modeling. Successful application of novel unsupervised feature extraction and clustering algorithms for use in Land Cover Classification requires the ability to determine what landscape attributes are represented by automated clustering. A closely related challenge is learning how to precondition the input data streams to the unsupervised classification algorithms in order to obtain clusters that represent Land Cover category of relevance to landsurface change and modeling applications. We present results from an ongoing effort to apply novel clustering methodologies developed primarily for neuroscience machine vision applications to the environmental sciences. We use a Hebbian learning rule to build spectral-textural dictionaries that are adapted to the data. We learn our dictionaries from millions of overlapping image patches and then use a pursuit search to generate sparse classification features. These sparse representations of pixel patches are used to perform unsupervised k-means clustering. In our application, we use 8-band multispectral Worldview-2 data from three arctic study areas: Barrow, Alaska; the Selawik River, Alaska; and a watershed near the Mackenzie River delta in northwest Canada. Our goal is to develop a robust classification methodology that will allow for the automated discretization of the landscape into distinct units based on attributes such as vegetation, surface hydrological properties (e.g. soil moisture and inundation), and topographic/geomorphic characteristics. The challenge of developing a meaningful land cover classification includes both learning how optimize the clustering algorithm and successfully interpreting the results. In applying the unsupervised clustering, we have the flexibility of selecting both the window
NASA Astrophysics Data System (ADS)
Ponomaryov, Volodymyr I.; Chavez-Roman, Herminio; Gonzalez-Huitron, Victor
2014-05-01
The paper presents the design and hardware implementation of novel framework for image resolution enhancement employing the wavelet domain. The principal idea of resolution enhancement consists of using edge preservation procedure and mutual interpolation between the input low-resolution (LR) image and the HF sub-band images performed via the Discrete Wavelet Transform (DWT). The LR image is used in the sparse representation for the resolutionenhancement process, employing a 1-D interpolation in set of angle directions; following, the computations of the new samples are found, estimating the missing samples. Finally, pixels are performed via the Lanczos interpolation. To preserve more edge information additional edge extraction in HF sub-bands is performed in the DWT decomposition of input image. The differences between the LL sub-band image and LR input image is used to correct the HF component, generating a significantly sharper reconstructed image. All sub-band images are used to generate the new HR image applying the inverse DWT (IDWT). Additionally, the novel framework employs a denoising procedure by using the Non-Local Means for the input LR image. An efficiency analysis of the designed and other state-of-the-art filters have been performed on the DSP TMS320DM648 by Texas Instruments through MATLAB's Simulink module and on the video card (NVIDIA®Quadro® K2000), showing that novel SR procedure can be used in real-time processing applications. Experimental results have confirmed that implemented framework outperforms existing SR algorithms in terms of objective criteria (PSNR, MAE and SSIM) as well as in subjective perception, justifying better image resolution.
NASA Astrophysics Data System (ADS)
Chen, Huasong; Qu, Xiangju; Jin, Ying; Li, Zhenhua; He, Anzhi
2016-10-01
Image deblurring is a fundamental problem in image processing. Conventional methods often deal with the degraded image as a whole while ignoring that an image contains two different components: cartoon and texture. Recently, total variation (TV) based image decomposition methods are introduced into image deblurring problem. However, these methods often suffer from the well-known stair-casing effects of TV. In this paper, a new cartoon -texture based sparsity regularization method is proposed for non-blind image deblurring. Based on image decomposition, it respectively regularizes the cartoon with a combined term including framelet-domain-based sparse prior and a quadratic regularization and the texture with the sparsity of discrete cosine transform domain. Then an adaptive alternative split Bregman iteration is proposed to solve the new multi-term sparsity regularization model. Experimental results demonstrate that our method can recover both cartoon and texture of images simultaneously, and therefore can improve the visual effect, the PSNR and the SSIM of the deblurred image efficiently than TV and the undecomposed methods.
Wen, Dong; Jia, Peilei; Lian, Qiusheng; Zhou, Yanhong; Lu, Chengbiao
2016-01-01
At present, the sparse representation-based classification (SRC) has become an important approach in electroencephalograph (EEG) signal analysis, by which the data is sparsely represented on the basis of a fixed dictionary or learned dictionary and classified based on the reconstruction criteria. SRC methods have been used to analyze the EEG signals of epilepsy, cognitive impairment and brain computer interface (BCI), which made rapid progress including the improvement in computational accuracy, efficiency and robustness. However, these methods have deficiencies in real-time performance, generalization ability and the dependence of labeled sample in the analysis of the EEG signals. This mini review described the advantages and disadvantages of the SRC methods in the EEG signal analysis with the expectation that these methods can provide the better tools for analyzing EEG signals. PMID:27458376
NASA Astrophysics Data System (ADS)
Prabhakar, Sunil Kumar; Rajaguru, Harikumar
2015-12-01
The most common and frequently occurring neurological disorder is epilepsy and the main method useful for the diagnosis of epilepsy is electroencephalogram (EEG) signal analysis. Due to the length of EEG recordings, EEG signal analysis method is quite time-consuming when it is processed manually by an expert. This paper proposes the application of Linear Graph Embedding (LGE) concept as a dimensionality reduction technique for processing the epileptic encephalographic signals and then it is classified using Sparse Representation Classifiers (SRC). SRC is used to analyze the classification of epilepsy risk levels from EEG signals and the parameters such as Sensitivity, Specificity, Time Delay, Quality Value, Performance Index and Accuracy are analyzed.
Sex Education Representations in Spanish Combined Biology and Geology Textbooks
NASA Astrophysics Data System (ADS)
García-Cabeza, Belén; Sánchez-Bello, Ana
2013-07-01
Sex education is principally dealt with as part of the combined subject of Biology and Geology in the Spanish school curriculum. Teachers of this subject are not specifically trained to teach sex education, and thus the contents of their assigned textbooks are the main source of information available to them in this field. The main goal of this study was to determine what information Biology and Geology textbooks provide with regard to sex education and the vision of sexuality they give, but above all to reveal which perspectives of sex education they legitimise and which they silence. We analysed the textbooks in question by interpreting both visual and text representations, as a means of enabling us to investigate the nature of the discourse on sex education. With this aim, we have used a qualitative methodology, based on the content analysis. The main analytical tool was an in-house grid constructed to allow us to analyse the visual and textual representations. Our analysis of the combined Biology and Geology textbooks for Secondary Year 3 revealed that there is a tendency to reproduce models of sex education that take place within a framework of the more traditional discourses. Besides, the results suggested that the most of the sample chosen for this study makes a superficial, incomplete, incorrect or biased approach to sex education.
Visual tracking via robust multitask sparse prototypes
NASA Astrophysics Data System (ADS)
Zhang, Huanlong; Hu, Shiqiang; Yu, Junyang
2015-03-01
Sparse representation has been applied to an online subspace learning-based tracking problem. To handle partial occlusion effectively, some researchers introduce l1 regularization to principal component analysis (PCA) reconstruction. However, in these traditional tracking methods, the representation of each object observation is often viewed as an individual task so the inter-relationship between PCA basis vectors is ignored. We propose a new online visual tracking algorithm with multitask sparse prototypes, which combines multitask sparse learning with PCA-based subspace representation. We first extend a visual tracking algorithm with sparse prototypes in multitask learning framework to mine inter-relations between subtasks. Then, to avoid the problem that enforcing all subtasks to share the same structure may result in degraded tracking results, we impose group sparse constraints on the coefficients of PCA basis vectors and element-wise sparse constraints on the error coefficients, respectively. Finally, we show that the proposed optimization problem can be effectively solved using the accelerated proximal gradient method with the fast convergence. Experimental results compared with the state-of-the-art tracking methods demonstrate that the proposed algorithm achieves favorable performance when the object undergoes partial occlusion, motion blur, and illumination changes.
Guo, Qiang; Qi, Liangang
2017-04-10
In the coexistence of multiple types of interfering signals, the performance of interference suppression methods based on time and frequency domains is degraded seriously, and the technique using an antenna array requires a large enough size and huge hardware costs. To combat multi-type interferences better for GNSS receivers, this paper proposes a cascaded multi-type interferences mitigation method combining improved double chain quantum genetic matching pursuit (DCQGMP)-based sparse decomposition and an MPDR beamformer. The key idea behind the proposed method is that the multiple types of interfering signals can be excised by taking advantage of their sparse features in different domains. In the first stage, the single-tone (multi-tone) and linear chirp interfering signals are canceled by sparse decomposition according to their sparsity in the over-complete dictionary. In order to improve the timeliness of matching pursuit (MP)-based sparse decomposition, a DCQGMP is introduced by combining an improved double chain quantum genetic algorithm (DCQGA) and the MP algorithm, and the DCQGMP algorithm is extended to handle the multi-channel signals according to the correlation among the signals in different channels. In the second stage, the minimum power distortionless response (MPDR) beamformer is utilized to nullify the residuary interferences (e.g., wideband Gaussian noise interferences). Several simulation results show that the proposed method can not only improve the interference mitigation degree of freedom (DoF) of the array antenna, but also effectively deal with the interference arriving from the same direction with the GNSS signal, which can be sparse represented in the over-complete dictionary. Moreover, it does not bring serious distortions into the navigation signal.
NASA Astrophysics Data System (ADS)
Eybpoosh, Matineh; Berges, Mario; Noh, Hae Young
2017-01-01
This work addresses the main challenges in real-world application of guided-waves for damage detection of pipelines, namely their complex nature and sensitivity to environmental and operational conditions (EOCs). Different propagation characteristics of the wave modes, their distinctive sensitivities to different types and ranges of EOCs, and to different damage scenarios, make the interpretation of diffuse-field guided-wave signals a challenging task. This paper proposes an unsupervised feature-extraction method for online damage detection of pipelines under varying EOCs. The objective is to simplify diffuse-field guided-wave signals to a sparse subset of the arrivals that contains the majority of the energy carried by the signal. We show that such a subset is less affected by EOCs compared to the complete time-traces of the signals. Moreover, it is shown that the effects of damage on the energy of this subset suppress those of EOCs. A set of signals from the undamaged state of a pipe are used as reference records. The reference dataset is used to extract the aforementioned sparse representation. During the monitoring stage, the sparse subset, representing the undamaged pipe, will not accurately reconstruct the energy of a signal from a damaged pipe. In other words, such a sparse representation of guided-waves is sensitive to occurrence of damage. Therefore, the energy estimation errors are used as damage-sensitive features for damage detection purposes. A diverse set of experimental analyses are conducted to verify the hypotheses of the proposed feature-extraction approach, and to validate the detection performance of the damage-sensitive features. The empirical validation of the proposed method includes (1) detecting a structural abnormality in an aluminum pipe, under varying temperature at different ranges, (2) detecting multiple small damages of different types, at different locations, in a steel pipe, under varying temperature, (3) detecting a structural
Ma, Xu; Cheng, Yongmei; Hao, Shuai
2016-12-10
Automatic classification of terrain surfaces from an aerial image is essential for an autonomous unmanned aerial vehicle (UAV) landing at an unprepared site by using vision. Diverse terrain surfaces may show similar spectral properties due to the illumination and noise that easily cause poor classification performance. To address this issue, a multi-stage classification algorithm based on low-rank recovery and multi-feature fusion sparse representation is proposed. First, color moments and Gabor texture feature are extracted from training data and stacked as column vectors of a dictionary. Then we perform low-rank matrix recovery for the dictionary by using augmented Lagrange multipliers and construct a multi-stage terrain classifier. Experimental results on an aerial map database that we prepared verify the classification accuracy and robustness of the proposed method.
NASA Astrophysics Data System (ADS)
Wang, Yan; Zhang, Pei; An, Le; Ma, Guangkai; Kang, Jiayin; Shi, Feng; Wu, Xi; Zhou, Jiliu; Lalush, David S.; Lin, Weili; Shen, Dinggang
2016-01-01
Positron emission tomography (PET) has been widely used in clinical diagnosis for diseases and disorders. To obtain high-quality PET images requires a standard-dose radionuclide (tracer) injection into the human body, which inevitably increases risk of radiation exposure. One possible solution to this problem is to predict the standard-dose PET image from its low-dose counterpart and its corresponding multimodal magnetic resonance (MR) images. Inspired by the success of patch-based sparse representation (SR) in super-resolution image reconstruction, we propose a mapping-based SR (m-SR) framework for standard-dose PET image prediction. Compared with the conventional patch-based SR, our method uses a mapping strategy to ensure that the sparse coefficients, estimated from the multimodal MR images and low-dose PET image, can be applied directly to the prediction of standard-dose PET image. As the mapping between multimodal MR images (or low-dose PET image) and standard-dose PET images can be particularly complex, one step of mapping is often insufficient. To this end, an incremental refinement framework is therefore proposed. Specifically, the predicted standard-dose PET image is further mapped to the target standard-dose PET image, and then the SR is performed again to predict a new standard-dose PET image. This procedure can be repeated for prediction refinement of the iterations. Also, a patch selection based dictionary construction method is further used to speed up the prediction process. The proposed method is validated on a human brain dataset. The experimental results show that our method can outperform benchmark methods in both qualitative and quantitative measures.
Double shrinking sparse dimension reduction.
Zhou, Tianyi; Tao, Dacheng
2013-01-01
Learning tasks such as classification and clustering usually perform better and cost less (time and space) on compressed representations than on the original data. Previous works mainly compress data via dimension reduction. In this paper, we propose "double shrinking" to compress image data on both dimensionality and cardinality via building either sparse low-dimensional representations or a sparse projection matrix for dimension reduction. We formulate a double shrinking model (DSM) as an l(1) regularized variance maximization with constraint ||x||(2)=1, and develop a double shrinking algorithm (DSA) to optimize DSM. DSA is a path-following algorithm that can build the whole solution path of locally optimal solutions of different sparse levels. Each solution on the path is a "warm start" for searching the next sparser one. In each iteration of DSA, the direction, the step size, and the Lagrangian multiplier are deduced from the Karush-Kuhn-Tucker conditions. The magnitudes of trivial variables are shrunk and the importances of critical variables are simultaneously augmented along the selected direction with the determined step length. Double shrinking can be applied to manifold learning and feature selections for better interpretation of features, and can be combined with classification and clustering to boost their performance. The experimental results suggest that double shrinking produces efficient and effective data compression.
Tseng, Yi-Li; Lin, Keng-Sheng; Jaw, Fu-Shan
2016-01-01
An automatic method is presented for detecting myocardial ischemia, which can be considered as the early symptom of acute coronary events. Myocardial ischemia commonly manifests as ST- and T-wave changes on ECG signals. The methods in this study are proposed to detect abnormal ECG beats using knowledge-based features and classification methods. A novel classification method, sparse representation-based classification (SRC), is involved to improve the performance of the existing algorithms. A comparison was made between two classification methods, SRC and support-vector machine (SVM), using rule-based vectors as input feature space. The two methods are proposed with quantitative evaluation to validate their performances. The results of SRC method encompassed with rule-based features demonstrate higher sensitivity than that of SVM. However, the specificity and precision are a trade-off. Moreover, SRC method is less dependent on the selection of rule-based features and can achieve high performance using fewer features. The overall performances of the two methods proposed in this study are better than the previous methods. PMID:26925158
Image fusion using sparse overcomplete feature dictionaries
Brumby, Steven P.; Bettencourt, Luis; Kenyon, Garrett T.; Chartrand, Rick; Wohlberg, Brendt
2015-10-06
Approaches for deciding what individuals in a population of visual system "neurons" are looking for using sparse overcomplete feature dictionaries are provided. A sparse overcomplete feature dictionary may be learned for an image dataset and a local sparse representation of the image dataset may be built using the learned feature dictionary. A local maximum pooling operation may be applied on the local sparse representation to produce a translation-tolerant representation of the image dataset. An object may then be classified and/or clustered within the translation-tolerant representation of the image dataset using a supervised classification algorithm and/or an unsupervised clustering algorithm.
Brayanov, Jordan B; Press, Daniel Z; Smith, Maurice A
2012-10-24
Actions can be planned in either an intrinsic (body-based) reference frame or an extrinsic (world-based) frame, and understanding how the internal representations associated with these frames contribute to the learning of motor actions is a key issue in motor control. We studied the internal representation of this learning in human subjects by analyzing generalization patterns across an array of different movement directions and workspaces after training a visuomotor rotation in a single movement direction in one workspace. This provided a dense sampling of the generalization function across intrinsic and extrinsic reference frames, which allowed us to dissociate intrinsic and extrinsic representations and determine the manner in which they contributed to the motor memory for a trained action. A first experiment showed that the generalization pattern reflected a memory that was intermediate between intrinsic and extrinsic representations. A second experiment showed that this intermediate representation could not arise from separate intrinsic and extrinsic learning. Instead, we find that the representation of learning is based on a gain-field combination of local representations in intrinsic and extrinsic coordinates. This gain-field representation generalizes between actions by effectively computing similarity based on the (Mahalanobis) distance across intrinsic and extrinsic coordinates and is in line with neural recordings showing mixed intrinsic-extrinsic representations in motor and parietal cortices.
NASA Astrophysics Data System (ADS)
Hanuš, J.; Ďurech, J.; Brož, M.; Marciniak, A.; Warner, B. D.; Pilcher, F.; Stephens, R.; Behrend, R.; Carry, B.; Čapek, D.; Antonini, P.; Audejean, M.; Augustesen, K.; Barbotin, E.; Baudouin, P.; Bayol, A.; Bernasconi, L.; Borczyk, W.; Bosch, J.-G.; Brochard, E.; Brunetto, L.; Casulli, S.; Cazenave, A.; Charbonnel, S.; Christophe, B.; Colas, F.; Coloma, J.; Conjat, M.; Cooney, W.; Correira, H.; Cotrez, V.; Coupier, A.; Crippa, R.; Cristofanelli, M.; Dalmas, Ch.; Danavaro, C.; Demeautis, C.; Droege, T.; Durkee, R.; Esseiva, N.; Esteban, M.; Fagas, M.; Farroni, G.; Fauvaud, M.; Fauvaud, S.; Del Freo, F.; Garcia, L.; Geier, S.; Godon, C.; Grangeon, K.; Hamanowa, H.; Hamanowa, H.; Heck, N.; Hellmich, S.; Higgins, D.; Hirsch, R.; Husarik, M.; Itkonen, T.; Jade, O.; Kamiński, K.; Kankiewicz, P.; Klotz, A.; Koff, R. A.; Kryszczyńska, A.; Kwiatkowski, T.; Laffont, A.; Leroy, A.; Lecacheux, J.; Leonie, Y.; Leyrat, C.; Manzini, F.; Martin, A.; Masi, G.; Matter, D.; Michałowski, J.; Michałowski, M. J.; Michałowski, T.; Michelet, J.; Michelsen, R.; Morelle, E.; Mottola, S.; Naves, R.; Nomen, J.; Oey, J.; Ogłoza, W.; Oksanen, A.; Oszkiewicz, D.; Pääkkönen, P.; Paiella, M.; Pallares, H.; Paulo, J.; Pavic, M.; Payet, B.; Polińska, M.; Polishook, D.; Poncy, R.; Revaz, Y.; Rinner, C.; Rocca, M.; Roche, A.; Romeuf, D.; Roy, R.; Saguin, H.; Salom, P. A.; Sanchez, S.; Santacana, G.; Santana-Ros, T.; Sareyan, J.-P.; Sobkowiak, K.; Sposetti, S.; Starkey, D.; Stoss, R.; Strajnic, J.; Teng, J.-P.; Trégon, B.; Vagnozzi, A.; Velichko, F. P.; Waelchli, N.; Wagrez, K.; Wücher, H.
2013-03-01
Context. The larger number of models of asteroid shapes and their rotational states derived by the lightcurve inversion give us better insight into both the nature of individual objects and the whole asteroid population. With a larger statistical sample we can study the physical properties of asteroid populations, such as main-belt asteroids or individual asteroid families, in more detail. Shape models can also be used in combination with other types of observational data (IR, adaptive optics images, stellar occultations), e.g., to determine sizes and thermal properties. Aims: We use all available photometric data of asteroids to derive their physical models by the lightcurve inversion method and compare the observed pole latitude distributions of all asteroids with known convex shape models with the simulated pole latitude distributions. Methods: We used classical dense photometric lightcurves from several sources (Uppsala Asteroid Photometric Catalogue, Palomar Transient Factory survey, and from individual observers) and sparse-in-time photometry from the U.S. Naval Observatory in Flagstaff, Catalina Sky Survey, and La Palma surveys (IAU codes 689, 703, 950) in the lightcurve inversion method to determine asteroid convex models and their rotational states. We also extended a simple dynamical model for the spin evolution of asteroids used in our previous paper. Results: We present 119 new asteroid models derived from combined dense and sparse-in-time photometry. We discuss the reliability of asteroid shape models derived only from Catalina Sky Survey data (IAU code 703) and present 20 such models. By using different values for a scaling parameter cYORP (corresponds to the magnitude of the YORP momentum) in the dynamical model for the spin evolution and by comparing synthetic and observed pole-latitude distributions, we were able to constrain the typical values of the cYORP parameter as between 0.05 and 0.6. Table 3 is available in electronic form at http://www.aanda.org
NASA Astrophysics Data System (ADS)
Uebbing, Bernd; Roscher, Ribana; Kusche, Jürgen
2016-04-01
Satellite radar altimeters allow global monitoring of mean sea level changes over the last two decades. However, coastal regions are less well observed due to influences on the returned signal energy by land located inside the altimeter footprint. The altimeter emits a radar pulse, which is reflected at the nadir-surface and measures the two-way travel time, as well as the returned energy as a function of time, resulting in a return waveform. Over the open ocean the waveform shape corresponds to a theoretical model which can be used to infer information on range corrections, significant wave height or wind speed. However, in coastal areas the shape of the waveform is significantly influenced by return signals from land, located in the altimeter footprint, leading to peaks which tend to bias the estimated parameters. Recently, several approaches dealing with this problem have been published, including utilizing only parts of the waveform (sub-waveforms), estimating the parameters in two steps or estimating additional peak parameters. We present a new approach in estimating sub-waveforms using conditional random fields (CRF) based on spatio-temporal waveform information. The CRF piece-wise approximates the measured waveforms based on a pre-derived dictionary of theoretical waveforms for various combinations of the geophysical parameters; neighboring range gates are likely to be assigned to the same underlying sub-waveform model. Depending on the choice of hyperparameters in the CRF estimation, the classification into sub-waveforms can either be more fine or coarse resulting in multiple sub-waveform hypotheses. After the sub-waveforms have been detected, existing retracking algorithms can be applied to derive water heights or other desired geophysical parameters from particular sub-waveforms. To identify the optimal heights from the multiple hypotheses, instead of utilizing a known reference height, we apply a Dijkstra-algorithm to find the "shortest path" of all
Image fusion via nonlocal sparse K-SVD dictionary learning.
Li, Ying; Li, Fangyi; Bai, Bendu; Shen, Qiang
2016-03-01
Image fusion aims to merge two or more images captured via various sensors of the same scene to construct a more informative image by integrating their details. Generally, such integration is achieved through the manipulation of the representations of the images concerned. Sparse representation plays an important role in the effective description of images, offering a great potential in a variety of image processing tasks, including image fusion. Supported by sparse representation, in this paper, an approach for image fusion by the use of a novel dictionary learning scheme is proposed. The nonlocal self-similarity property of the images is exploited, not only at the stage of learning the underlying description dictionary but during the process of image fusion. In particular, the property of nonlocal self-similarity is combined with the traditional sparse dictionary. This results in an improved learned dictionary, hereafter referred to as the nonlocal sparse K-SVD dictionary (where K-SVD stands for the K times singular value decomposition that is commonly used in the literature), and abbreviated to NL_SK_SVD. The performance of the NL_SK_SVD dictionary is applied for image fusion using simultaneous orthogonal matching pursuit. The proposed approach is evaluated with different types of images, and compared with a number of alternative image fusion techniques. The resultant superior fused images using the present approach demonstrates the efficacy of the NL_SK_SVD dictionary in sparse image representation.
Sajjad, Muhammad; Mehmood, Irfan; Baik, Sung Wook
2014-01-01
Visual sensor networks (VSNs) usually generate a low-resolution (LR) frame-sequence due to energy and processing constraints. These LR-frames are not very appropriate for use in certain surveillance applications. It is very important to enhance the resolution of the captured LR-frames using resolution enhancement schemes. In this paper, an effective framework for a super-resolution (SR) scheme is proposed that enhances the resolution of LR key-frames extracted from frame-sequences captured by visual-sensors. In a VSN, a visual processing hub (VPH) collects a huge amount of visual data from camera sensors. In the proposed framework, at the VPH, key-frames are extracted using our recent key-frame extraction technique and are streamed to the base station (BS) after compression. A novel effective SR scheme is applied at BS to produce a high-resolution (HR) output from the received key-frames. The proposed SR scheme uses optimized orthogonal matching pursuit (OOMP) for sparse-representation recovery in SR. OOMP does better in terms of detecting true sparsity than orthogonal matching pursuit (OMP). This property of the OOMP helps produce a HR image which is closer to the original image. The K-SVD dictionary learning procedure is incorporated for dictionary learning. Batch-OMP improves the dictionary learning process by removing the limitation in handling a large set of observed signals. Experimental results validate the effectiveness of the proposed scheme and show its superiority over other state-of-the-art schemes. PMID:24566632
Sajjad, Muhammad; Mehmood, Irfan; Baik, Sung Wook
2014-02-21
Visual sensor networks (VSNs) usually generate a low-resolution (LR) frame-sequence due to energy and processing constraints. These LR-frames are not very appropriate for use in certain surveillance applications. It is very important to enhance the resolution of the captured LR-frames using resolution enhancement schemes. In this paper, an effective framework for a super-resolution (SR) scheme is proposed that enhances the resolution of LR key-frames extracted from frame-sequences captured by visual-sensors. In a VSN, a visual processing hub (VPH) collects a huge amount of visual data from camera sensors. In the proposed framework, at the VPH, key-frames are extracted using our recent key-frame extraction technique and are streamed to the base station (BS) after compression. A novel effective SR scheme is applied at BS to produce a high-resolution (HR) output from the received key-frames. The proposed SR scheme uses optimized orthogonal matching pursuit (OOMP) for sparse-representation recovery in SR. OOMP does better in terms of detecting true sparsity than orthogonal matching pursuit (OMP). This property of the OOMP helps produce a HR image which is closer to the original image. The K-SVD dictionary learning procedure is incorporated for dictionary learning. Batch-OMP improves the dictionary learning process by removing the limitation in handling a large set of observed signals. Experimental results validate the effectiveness of the proposed scheme and show its superiority over other state-of-the-art schemes.
Structured sparse models for classification
NASA Astrophysics Data System (ADS)
Castrodad, Alexey
The main focus of this thesis is the modeling and classification of high dimensional data using structured sparsity. Sparse models, where data is assumed to be well represented as a linear combination of a few elements from a dictionary, have gained considerable attention in recent years, and its use has led to state-of-the-art results in many signal and image processing tasks. The success of sparse modeling is highly due to its ability to efficiently use the redundancy of the data and find its underlying structure. On a classification setting, we capitalize on this advantage to properly model and separate the structure of the classes. We design and validate modeling solutions to challenging problems arising in computer vision and remote sensing. We propose both supervised and unsupervised schemes for the modeling of human actions from motion imagery under a wide variety of acquisition condi- tions. In the supervised case, the main goal is to classify the human actions in the video given a predefined set of actions to learn from. In the unsupervised case, the main goal is to an- alyze the spatio-temporal dynamics of the individuals in the scene without having any prior information on the actions themselves. We also propose a model for remotely sensed hysper- spectral imagery, where the main goal is to perform automatic spectral source separation and mapping at the subpixel level. Finally, we present a sparse model for sensor fusion to exploit the common structure and enforce collaboration of hyperspectral with LiDAR data for better mapping capabilities. In all these scenarios, we demonstrate that these data can be expressed as a combination of atoms from a class-structured dictionary. These data representation becomes essentially a "mixture of classes," and by directly exploiting the sparse codes, one can attain highly accurate classification performance with relatively unsophisticated classifiers.
Multiresolution image representation using combined 2-D and 1-D directional filter banks.
Tanaka, Yuichi; Ikehara, Masaaki; Nguyen, Truong Q
2009-02-01
In this paper, effective multiresolution image representations using a combination of 2-D filter bank (FB) and directional wavelet transform (WT) are presented. The proposed methods yield simple implementation and low computation costs compared to previous 1-D and 2-D FB combinations or adaptive directional WT methods. Furthermore, they are nonredundant transforms and realize quad-tree like multiresolution representations. In applications on nonlinear approximation, image coding, and denoising, the proposed filter banks show visual quality improvements and have higher PSNR than the conventional separable WT or the contourlet.
Sparse Methods for Biomedical Data.
Ye, Jieping; Liu, Jun
2012-06-01
Following recent technological revolutions, the investigation of massive biomedical data with growing scale, diversity, and complexity has taken a center stage in modern data analysis. Although complex, the underlying representations of many biomedical data are often sparse. For example, for a certain disease such as leukemia, even though humans have tens of thousands of genes, only a few genes are relevant to the disease; a gene network is sparse since a regulatory pathway involves only a small number of genes; many biomedical signals are sparse or compressible in the sense that they have concise representations when expressed in a proper basis. Therefore, finding sparse representations is fundamentally important for scientific discovery. Sparse methods based on the [Formula: see text] norm have attracted a great amount of research efforts in the past decade due to its sparsity-inducing property, convenient convexity, and strong theoretical guarantees. They have achieved great success in various applications such as biomarker selection, biological network construction, and magnetic resonance imaging. In this paper, we review state-of-the-art sparse methods and their applications to biomedical data.
Sparse Methods for Biomedical Data
Ye, Jieping; Liu, Jun
2013-01-01
Following recent technological revolutions, the investigation of massive biomedical data with growing scale, diversity, and complexity has taken a center stage in modern data analysis. Although complex, the underlying representations of many biomedical data are often sparse. For example, for a certain disease such as leukemia, even though humans have tens of thousands of genes, only a few genes are relevant to the disease; a gene network is sparse since a regulatory pathway involves only a small number of genes; many biomedical signals are sparse or compressible in the sense that they have concise representations when expressed in a proper basis. Therefore, finding sparse representations is fundamentally important for scientific discovery. Sparse methods based on the ℓ1 norm have attracted a great amount of research efforts in the past decade due to its sparsity-inducing property, convenient convexity, and strong theoretical guarantees. They have achieved great success in various applications such as biomarker selection, biological network construction, and magnetic resonance imaging. In this paper, we review state-of-the-art sparse methods and their applications to biomedical data. PMID:24076585
Building Hierarchical Representations for Oracle Character and Sketch Recognition.
Jun Guo; Changhu Wang; Roman-Rangel, Edgar; Hongyang Chao; Yong Rui
2016-01-01
In this paper, we study oracle character recognition and general sketch recognition. First, a data set of oracle characters, which are the oldest hieroglyphs in China yet remain a part of modern Chinese characters, is collected for analysis. Second, typical visual representations in shape- and sketch-related works are evaluated. We analyze the problems suffered when addressing these representations and determine several representation design criteria. Based on the analysis, we propose a novel hierarchical representation that combines a Gabor-related low-level representation and a sparse-encoder-related mid-level representation. Extensive experiments show the effectiveness of the proposed representation in both oracle character recognition and general sketch recognition. The proposed representation is also complementary to convolutional neural network (CNN)-based models. We introduce a solution to combine the proposed representation with CNN-based models, and achieve better performances over both approaches. This solution has beaten humans at recognizing general sketches.
van Tulder, Gijs; de Bruijne, Marleen
2016-05-01
The choice of features greatly influences the performance of a tissue classification system. Despite this, many systems are built with standard, predefined filter banks that are not optimized for that particular application. Representation learning methods such as restricted Boltzmann machines may outperform these standard filter banks because they learn a feature description directly from the training data. Like many other representation learning methods, restricted Boltzmann machines are unsupervised and are trained with a generative learning objective; this allows them to learn representations from unlabeled data, but does not necessarily produce features that are optimal for classification. In this paper we propose the convolutional classification restricted Boltzmann machine, which combines a generative and a discriminative learning objective. This allows it to learn filters that are good both for describing the training data and for classification. We present experiments with feature learning for lung texture classification and airway detection in CT images. In both applications, a combination of learning objectives outperformed purely discriminative or generative learning, increasing, for instance, the lung tissue classification accuracy by 1 to 8 percentage points. This shows that discriminative learning can help an otherwise unsupervised feature learner to learn filters that are optimized for classification.
Serino, Andrea; Canzoneri, Elisa; Marzolla, Marilena; di Pellegrino, Giuseppe; Magosso, Elisa
2015-01-01
Stimuli from different sensory modalities occurring on or close to the body are integrated in a multisensory representation of the space surrounding the body, i.e., peripersonal space (PPS). PPS dynamically modifies depending on experience, e.g., it extends after using a tool to reach far objects. However, the neural mechanism underlying PPS plasticity after tool use is largely unknown. Here we use a combined computational-behavioral approach to propose and test a possible mechanism accounting for PPS extension. We first present a neural network model simulating audio-tactile representation in the PPS around one hand. Simulation experiments showed that our model reproduced the main property of PPS neurons, i.e., selective multisensory response for stimuli occurring close to the hand. We used the neural network model to simulate the effects of a tool-use training. In terms of sensory inputs, tool use was conceptualized as a concurrent tactile stimulation from the hand, due to holding the tool, and an auditory stimulation from the far space, due to tool-mediated action. Results showed that after exposure to those inputs, PPS neurons responded also to multisensory stimuli far from the hand. The model thus suggests that synchronous pairing of tactile hand stimulation and auditory stimulation from the far space is sufficient to extend PPS, such as after tool-use. Such prediction was confirmed by a behavioral experiment, where we used an audio-tactile interaction paradigm to measure the boundaries of PPS representation. We found that PPS extended after synchronous tactile-hand stimulation and auditory-far stimulation in a group of healthy volunteers. Control experiments both in simulation and behavioral settings showed that the same amount of tactile and auditory inputs administered out of synchrony did not change PPS representation. We conclude by proposing a simple, biological-plausible model to explain plasticity in PPS representation after tool-use, which is
Serino, Andrea; Canzoneri, Elisa; Marzolla, Marilena; di Pellegrino, Giuseppe; Magosso, Elisa
2015-01-01
Stimuli from different sensory modalities occurring on or close to the body are integrated in a multisensory representation of the space surrounding the body, i.e., peripersonal space (PPS). PPS dynamically modifies depending on experience, e.g., it extends after using a tool to reach far objects. However, the neural mechanism underlying PPS plasticity after tool use is largely unknown. Here we use a combined computational-behavioral approach to propose and test a possible mechanism accounting for PPS extension. We first present a neural network model simulating audio-tactile representation in the PPS around one hand. Simulation experiments showed that our model reproduced the main property of PPS neurons, i.e., selective multisensory response for stimuli occurring close to the hand. We used the neural network model to simulate the effects of a tool-use training. In terms of sensory inputs, tool use was conceptualized as a concurrent tactile stimulation from the hand, due to holding the tool, and an auditory stimulation from the far space, due to tool-mediated action. Results showed that after exposure to those inputs, PPS neurons responded also to multisensory stimuli far from the hand. The model thus suggests that synchronous pairing of tactile hand stimulation and auditory stimulation from the far space is sufficient to extend PPS, such as after tool-use. Such prediction was confirmed by a behavioral experiment, where we used an audio-tactile interaction paradigm to measure the boundaries of PPS representation. We found that PPS extended after synchronous tactile-hand stimulation and auditory-far stimulation in a group of healthy volunteers. Control experiments both in simulation and behavioral settings showed that the same amount of tactile and auditory inputs administered out of synchrony did not change PPS representation. We conclude by proposing a simple, biological-plausible model to explain plasticity in PPS representation after tool-use, which is
Online Dictionary Learning for Sparse Coding
2009-04-01
cessing tasks such as denoising (Elad & Aharon, 2006) as well as higher-level tasks such as classification (Raina et al., 2007; Mairal et al., 2008a...Bruckstein, A. M. (2006). The K- SVD : An algorithm for designing of overcomplete dic- tionaries for sparse representations. IEEE Trans. SP...Tibshirani, R. (2004). Least angle regression. Ann. Statist. Elad, M., & Aharon, M. (2006). Image denoising via sparse and redundant representations
Sparse recovery via convex optimization
NASA Astrophysics Data System (ADS)
Randall, Paige Alicia
This thesis considers the problem of estimating a sparse signal from a few (possibly noisy) linear measurements. In other words, we have y = Ax + z where A is a measurement matrix with more columns than rows, x is a sparse signal to be estimated, z is a noise vector, and y is a vector of measurements. This setup arises frequently in many problems ranging from MRI imaging to genomics to compressed sensing.We begin by relating our setup to an error correction problem over the reals, where a received encoded message is corrupted by a few arbitrary errors, as well as smaller dense errors. We show that under suitable conditions on the encoding matrix and on the number of arbitrary errors, one is able to accurately recover the message.We next show that we are able to achieve oracle optimality for x, up to a log factor and a factor of sqrt{s}, when we require the matrix A to obey an incoherence property. The incoherence property is novel in that it allows the coherence of A to be as large as O(1/ log n) and still allows sparsities as large as O(m/log n). This is in contrast to other existing results involving coherence where the coherence can only be as large as O(1/sqrt{m}) to allow sparsities as large as O(sqrt{m}). We also do not make the common assumption that the matrix A obeys a restricted eigenvalue condition.We then show that we can recover a (non-sparse) signal from a few linear measurements when the signal has an exactly sparse representation in an overcomplete dictionary. We again only require that the dictionary obey an incoherence property.Finally, we introduce the method of l_1 analysis and show that it is guaranteed to give good recovery of a signal from a few measurements, when the signal can be well represented in a dictionary. We require that the combined measurement/dictionary matrix satisfies a uniform uncertainty principle and we compare our results with the more standard l_1 synthesis approach.All our methods involve solving an l_1 minimization
A combined representation method for use in band structure calculations. 1: Method
NASA Technical Reports Server (NTRS)
Friedli, C.; Ashcroft, N. W.
1975-01-01
A representation was described whose basis levels combine the important physical aspects of a finite set of plane waves with those of a set of Bloch tight-binding levels. The chosen combination has a particularly simple dependence on the wave vector within the Brillouin Zone, and its use in reducing the standard one-electron band structure problem to the usual secular equation has the advantage that the lattice sums involved in the calculation of the matrix elements are actually independent of the wave vector. For systems with complicated crystal structures, for which the Korringa-Kohn-Rostoker (KKR), Augmented-Plane Wave (APW) and Orthogonalized-Plane Wave (OPW) methods are difficult to apply, the present method leads to results with satisfactory accuracy and convergence.
Combination of geodetic measurements by means of a multi-resolution representation
NASA Astrophysics Data System (ADS)
Goebel, G.; Schmidt, M. G.; Börger, K.; List, H.; Bosch, W.
2010-12-01
Recent and in particular current satellite gravity missions provide important contributions for global Earth gravity models, and these global models can be refined by airborne and terrestrial gravity observations. The most common representation of a gravity field model in terms of spherical harmonics has the disadvantages that it is difficult to represent small spatial details and cannot handle data gaps appropriately. An adequate modeling using a multi-resolution representation (MRP) is necessary in order to exploit the highest degree of information out of all these mentioned measurements. The MRP provides a simple hierarchical framework for identifying the properties of a signal. The procedure starts from the measurements, performs the decomposition into frequency-dependent detail signals by applying a pyramidal algorithm and allows for data compression and filtering, i.e. data manipulations. Since different geodetic measurement types (terrestrial, airborne, spaceborne) cover different parts of the frequency spectrum, it seems reasonable to calculate the detail signals of the lower levels mainly from satellite data, the detail signals of medium levels mainly from airborne and the detail signals of the higher levels mainly from terrestrial data. A concept is presented how these different measurement types can be combined within the MRP. In this presentation the basic principles on strategies and concepts for the generation of MRPs will be shown. Examples of regional gravity field determination are presented.
Gao, Yang; Bian, Zhaoying; Huang, Jing; Zhang, Yunwan; Niu, Shanzhou; Feng, Qianjin; Chen, Wufan; Liang, Zhengrong; Ma, Jianhua
2014-01-01
To realize low-dose imaging in X-ray computed tomography (CT) examination, lowering milliampere-seconds (low-mAs) or reducing the required number of projection views (sparse-view) per rotation around the body has been widely studied as an easy and effective approach. In this study, we are focusing on low-dose CT image reconstruction from the sinograms acquired with a combined low-mAs and sparse-view protocol and propose a two-step image reconstruction strategy. Specifically, to suppress significant statistical noise in the noisy and insufficient sinograms, an adaptive sinogram restoration (ASR) method is first proposed with consideration of the statistical property of sinogram data, and then to further acquire a high-quality image, a total variation based projection onto convex sets (TV-POCS) method is adopted with a slight modification. For simplicity, the present reconstruction strategy was termed as “ASR-TV-POCS.” To evaluate the present ASR-TV-POCS method, both qualitative and quantitative studies were performed on a physical phantom. Experimental results have demonstrated that the present ASR-TV-POCS method can achieve promising gains over other existing methods in terms of the noise reduction, contrast-to-noise ratio, and edge detail preservation. PMID:24977611
NASA Technical Reports Server (NTRS)
Wambsganss, J.; Witt, H. J.; Schneider, P.
1992-01-01
We present a combination of two very different methods for numerically calculating the effects of gravitational microlensing: the backward-ray-tracing that results in two-dimensional magnification patterns, and the parametric representation of caustic lines; they are in a way complementary to each other. The combination of these methods is much more powerful than the sum of its parts. It allows to determine the total magnification and the number of microimages as a function of source position. The mean number of microimages is calculated analytically and compared to the numerical results. The peaks in the lightcurves, as obtained from one-dimensional tracks through the magnification pattern, can now be divided into two groups: those which correspond to a source crossing a caustic, and those which are due to sources passing outside cusps. We determine the frequencies of those two types of events as a function of the surface mass density, and the probability distributions of their magnitudes. We find that for low surface mass density as many as 40 percent of all events in a lightcurve are not due to caustic crossings, but rather due to passings outside cusps.
NASA Astrophysics Data System (ADS)
Liu, Bin; Wang, Shanyi; Wang, Xiaolong
2015-10-01
DNA-binding proteins play an important role in most cellular processes. Therefore, it is necessary to develop an efficient predictor for identifying DNA-binding proteins only based on the sequence information of proteins. The bottleneck for constructing a useful predictor is to find suitable features capturing the characteristics of DNA binding proteins. We applied PseAAC to DNA binding protein identification, and PseAAC was further improved by incorporating the evolutionary information by using profile-based protein representation. Finally, Combined with Support Vector Machines (SVMs), a predictor called iDNAPro-PseAAC was proposed. Experimental results on an updated benchmark dataset showed that iDNAPro-PseAAC outperformed some state-of-the-art approaches, and it can achieve stable performance on an independent dataset. By using an ensemble learning approach to incorporate more negative samples (non-DNA binding proteins) in the training process, the performance of iDNAPro-PseAAC was further improved. The web server of iDNAPro-PseAAC is available at http://bioinformatics.hitsz.edu.cn/iDNAPro-PseAAC/.
JiTTree: A Just-in-Time Compiled Sparse GPU Volume Data Structure.
Labschütz, Matthias; Bruckner, Stefan; Gröller, M Eduard; Hadwiger, Markus; Rautek, Peter
2016-01-01
Sparse volume data structures enable the efficient representation of large but sparse volumes in GPU memory for computation and visualization. However, the choice of a specific data structure for a given data set depends on several factors, such as the memory budget, the sparsity of the data, and data access patterns. In general, there is no single optimal sparse data structure, but a set of several candidates with individual strengths and drawbacks. One solution to this problem are hybrid data structures which locally adapt themselves to the sparsity. However, they typically suffer from increased traversal overhead which limits their utility in many applications. This paper presents JiTTree, a novel sparse hybrid volume data structure that uses just-in-time compilation to overcome these problems. By combining multiple sparse data structures and reducing traversal overhead we leverage their individual advantages. We demonstrate that hybrid data structures adapt well to a large range of data sets. They are especially superior to other sparse data structures for data sets that locally vary in sparsity. Possible optimization criteria are memory, performance and a combination thereof. Through just-in-time (JIT) compilation, JiTTree reduces the traversal overhead of the resulting optimal data structure. As a result, our hybrid volume data structure enables efficient computations on the GPU, while being superior in terms of memory usage when compared to non-hybrid data structures.
Rescigno, Thomas N.; Horner, Daniel A.; Yip, Frank L.; McCurdy,C. William
2005-08-29
Gaussian basis functions, routinely employed in molecular electronic structure calculations, can be combined with numerical grid-based functions in a discrete variable representation to provide an efficient method for computing molecular continuum wave functions. This approach, combined with exterior complex scaling, obviates the need for slowly convergent single-center expansions, and allows one to study a variety of electron-molecule collision problems. The method is illustrated by computation of various bound and continuum properties of H2+.
Inversion of magnetotelluric data in a sparse model domain
NASA Astrophysics Data System (ADS)
Nittinger, Christian G.; Becken, Michael
2016-08-01
The inversion of magnetotelluric data into subsurface electrical conductivity poses an ill-posed problem. Smoothing constraints are widely employed to estimate a regularized solution. Here, we present an alternative inversion scheme that estimates a sparse representation of the model in a wavelet basis. The objective of the inversion is to determine the few non-zero wavelet coefficients which are required to fit the data. This approach falls into the class of sparsity constrained inversion schemes and minimizes the combination of the data misfit in a least-squares ℓ2 sense and of a model coefficient norm in an ℓ1 sense (ℓ2-ℓ1 minimization). The ℓ1 coefficient norm renders the solution sparse in a suitable representation such as the multiresolution wavelet basis, but does not impose explicit structural penalties on the model as it is the case for ℓ2 regularization. The presented numerical algorithm solves the mixed ℓ2-ℓ1 norm minimization problem for the nonlinear magnetotelluric inverse problem. We demonstrate the feasibility of our algorithm on synthetic 2-D MT data as well as on a real data example. We found that sparse models can be estimated by inversion and that the spatial distribution of non-vanishing coefficients indicates regions in the model which are resolved.
NASA Astrophysics Data System (ADS)
Zheng, Maoteng; Zhang, Yongjun; Zhou, Shunping; Zhu, Junfeng; Xiong, Xiaodong
2016-07-01
In recent years, new platforms and sensors in photogrammetry, remote sensing and computer vision areas have become available, such as Unmanned Aircraft Vehicles (UAV), oblique camera systems, common digital cameras and even mobile phone cameras. Images collected by all these kinds of sensors could be used as remote sensing data sources. These sensors can obtain large-scale remote sensing data which consist of a great number of images. Bundle block adjustment of large-scale data with conventional algorithm is very time and space (memory) consuming due to the super large normal matrix arising from large-scale data. In this paper, an efficient Block-based Sparse Matrix Compression (BSMC) method combined with the Preconditioned Conjugate Gradient (PCG) algorithm is chosen to develop a stable and efficient bundle block adjustment system in order to deal with the large-scale remote sensing data. The main contribution of this work is the BSMC-based PCG algorithm which is more efficient in time and memory than the traditional algorithm without compromising the accuracy. Totally 8 datasets of real data are used to test our proposed method. Preliminary results have shown that the BSMC method can efficiently decrease the time and memory requirement of large-scale data.
Sparse Representation of Smooth Linear Operators
1990-08-01
received study by many authors, resulting in constructions with a variety of properties. Meyer [13] constructed orthonormal wavelets for which h E CI(R...Lemmas 2.3 and 2.4; in fact, substitution of the finite sums which determine the elements of UTUT for the integrals in those lemmas yields the...some k the orthogonal matrices U1,..., U, defined in Section 4.1 have been computed (1 = log2(n/k)). We now present a procedure for computation of UTUT
Sparse Representations for Limited Data Tomography (PREPRINT)
2007-11-01
dictionary. Let αk∈ RJ denote the k- th row of α. The K- SVD algorithm for denoising of gray scale images essentially minimizes the objective function in...predefined (such as wavelets) or learned (e.g., by the K- SVD algorithm [8]), as in this work. Due to its highly effectiveness for tasks such as image... denoising , demosaicing, and inpainting, in particular when the dictionary is learned [9, 10], here we extend this idea to tomographic reconstruction. To
Sparse Representation for Color Image Restoration (PREPRINT)
2006-10-01
learning dictionaries for color images and extend the K- SVD -based grayscale image denoising algorithm that appears in [2]. This work puts forward...extend the K- SVD -based gray- scale image denoising algorithm that appears in [2]. This work puts forward ways for handling non- homogeneous noise and...brief description of the K- SVD -based gray-scale image denoising algorithm as proposed in [2]. Section 4 describes the novelties offered in this paper
Collins, Tom; Tillmann, Barbara; Barrett, Frederick S; Delbé, Charles; Janata, Petr
2014-01-01
Listeners' expectations for melodies and harmonies in tonal music are perhaps the most studied aspect of music cognition. Long debated has been whether faster response times (RTs) to more strongly primed events (in a music theoretic sense) are driven by sensory or cognitive mechanisms, such as repetition of sensory information or activation of cognitive schemata that reflect learned tonal knowledge, respectively. We analyzed over 300 stimuli from 7 priming experiments comprising a broad range of musical material, using a model that transforms raw audio signals through a series of plausible physiological and psychological representations spanning a sensory-cognitive continuum. We show that RTs are modeled, in part, by information in periodicity pitch distributions, chroma vectors, and activations of tonal space--a representation on a toroidal surface of the major/minor key relationships in Western tonal music. We show that in tonal space, melodies are grouped by their tonal rather than timbral properties, whereas the reverse is true for the periodicity pitch representation. While tonal space variables explained more of the variation in RTs than did periodicity pitch variables, suggesting a greater contribution of cognitive influences to tonal expectation, a stepwise selection model contained variables from both representations and successfully explained the pattern of RTs across stimulus categories in 4 of the 7 experiments. The addition of closure--a cognitive representation of a specific syntactic relationship--succeeded in explaining results from all 7 experiments. We conclude that multiple representational stages along a sensory-cognitive continuum combine to shape tonal expectations in music.
NASA Astrophysics Data System (ADS)
He, Fei; Han, Ye; Gong, Jianting; Song, Jiazhi; Wang, Han; Li, Yanwen
2017-03-01
Small interfering RNAs (siRNAs) may induce to targeted gene knockdown, and the gene silencing effectiveness relies on the efficacy of the siRNA. Therefore, the task of this paper is to construct an effective siRNA prediction method. In our work, we try to describe siRNA from both quantitative and qualitative aspects. For quantitative analyses, we form four groups of effective features, including nucleotide frequencies, thermodynamic stability profile, thermodynamic of siRNA-mRNA interaction, and mRNA related features, as a new mixed representation, in which thermodynamic of siRNA-mRNA interaction is introduced to siRNA efficacy prediction for the first time to our best knowledge. And then an F-score based feature selection is employed to investigate the contribution of each feature and remove the weak relevant features. Meanwhile, we encode the siRNA sequence and existed empirical design rules as a qualitative siRNA representation. These two kinds of siRNA representations are combined to predict siRNA efficacy by supported Vector Regression (SVR) at score level. The experimental results indicate that our method may select the features with powerful discriminative ability and make the two kinds of siRNA representations work at full capacity. The prediction results also demonstrate that our method can outperform other popular siRNA efficacy prediction algorithms.
He, Fei; Han, Ye; Gong, Jianting; Song, Jiazhi; Wang, Han; Li, Yanwen
2017-01-01
Small interfering RNAs (siRNAs) may induce to targeted gene knockdown, and the gene silencing effectiveness relies on the efficacy of the siRNA. Therefore, the task of this paper is to construct an effective siRNA prediction method. In our work, we try to describe siRNA from both quantitative and qualitative aspects. For quantitative analyses, we form four groups of effective features, including nucleotide frequencies, thermodynamic stability profile, thermodynamic of siRNA-mRNA interaction, and mRNA related features, as a new mixed representation, in which thermodynamic of siRNA-mRNA interaction is introduced to siRNA efficacy prediction for the first time to our best knowledge. And then an F-score based feature selection is employed to investigate the contribution of each feature and remove the weak relevant features. Meanwhile, we encode the siRNA sequence and existed empirical design rules as a qualitative siRNA representation. These two kinds of siRNA representations are combined to predict siRNA efficacy by supported Vector Regression (SVR) at score level. The experimental results indicate that our method may select the features with powerful discriminative ability and make the two kinds of siRNA representations work at full capacity. The prediction results also demonstrate that our method can outperform other popular siRNA efficacy prediction algorithms. PMID:28317874
Structured sparse priors for image classification.
Srinivas, Umamahesh; Suo, Yuanming; Dao, Minh; Monga, Vishal; Tran, Trac D
2015-06-01
Model-based compressive sensing (CS) exploits the structure inherent in sparse signals for the design of better signal recovery algorithms. This information about structure is often captured in the form of a prior on the sparse coefficients, with the Laplacian being the most common such choice (leading to l1 -norm minimization). Recent work has exploited the discriminative capability of sparse representations for image classification by employing class-specific dictionaries in the CS framework. Our contribution is a logical extension of these ideas into structured sparsity for classification. We introduce the notion of discriminative class-specific priors in conjunction with class specific dictionaries, specifically the spike-and-slab prior widely applied in Bayesian sparse regression. Significantly, the proposed framework takes the burden off the demand for abundant training image samples necessary for the success of sparsity-based classification schemes. We demonstrate this practical benefit of our approach in important applications, such as face recognition and object categorization.
Guided wavefield reconstruction from sparse measurements
NASA Astrophysics Data System (ADS)
Mesnil, Olivier; Ruzzene, Massimo
2016-02-01
Guided wave measurements are at the basis of several Non-Destructive Evaluation (NDE) techniques. Although sparse measurements of guided wave obtained using piezoelectric sensors can efficiently detect and locate defects, extensive informa-tion on the shape and subsurface location of defects can be extracted from full-field measurements acquired by Laser Doppler Vibrometers (LDV). Wavefield acquisition from LDVs is generally a slow operation due to the fact that the wave propagation to record must be repeated for each point measurement and the initial conditions must be reached between each measurement. In this research, a Sparse Wavefield Reconstruction (SWR) process using Compressed Sensing is developed. The goal of this technique is to reduce the number of point measurements needed to apply NDE techniques by at least one order of magnitude by extrapolating the knowledge of a few randomly chosen measured pixels over an over-sampled grid. To achieve this, the Lamb wave propagation equation is used to formulate a basis of shape functions in which the wavefield has a sparse representation, in order to comply with the Compressed Sensing requirements and use l1-minimization solvers. The main assumption of this reconstruction process is that every material point of the studied area is a potential source. The Compressed Sensing matrix is defined as being the contribution that would have been received at a measurement location from each possible source, using the dispersion relations of the specimen computed using a Semi-Analytical Finite Element technique. The measurements are then processed through an l1-minimizer to find a minimum corresponding to the set of active sources and their corresponding excitation functions. This minimum represents the best combination of the parameters of the model matching the sparse measurements. Wavefields are then reconstructed using the propagation equation. The set of active sources found by minimization contains all the wave
Deep ensemble learning of sparse regression models for brain disease diagnosis.
Suk, Heung-Il; Lee, Seong-Whan; Shen, Dinggang
2017-04-01
Recent studies on brain imaging analysis witnessed the core roles of machine learning techniques in computer-assisted intervention for brain disease diagnosis. Of various machine-learning techniques, sparse regression models have proved their effectiveness in handling high-dimensional data but with a small number of training samples, especially in medical problems. In the meantime, deep learning methods have been making great successes by outperforming the state-of-the-art performances in various applications. In this paper, we propose a novel framework that combines the two conceptually different methods of sparse regression and deep learning for Alzheimer's disease/mild cognitive impairment diagnosis and prognosis. Specifically, we first train multiple sparse regression models, each of which is trained with different values of a regularization control parameter. Thus, our multiple sparse regression models potentially select different feature subsets from the original feature set; thereby they have different powers to predict the response values, i.e., clinical label and clinical scores in our work. By regarding the response values from our sparse regression models as target-level representations, we then build a deep convolutional neural network for clinical decision making, which thus we call 'Deep Ensemble Sparse Regression Network.' To our best knowledge, this is the first work that combines sparse regression models with deep neural network. In our experiments with the ADNI cohort, we validated the effectiveness of the proposed method by achieving the highest diagnostic accuracies in three classification tasks. We also rigorously analyzed our results and compared with the previous studies on the ADNI cohort in the literature.
Vergeer, Mark; Moors, Pieter; Wagemans, Johan; van Ee, Raymond
2016-09-01
Our visual system faces the challenging task to construct integrated visual representations from the visual input projected on our retinae. Previous research has provided mixed evidence as to whether visual awareness of the stimulus parts is required for such integration to occur. Here, we address this issue by taking a novel approach in which we combine a monocular rivalry stimulus (i.e., a bistable rotating cylinder) with binocular rivalry. The results of Experiment 1 show that in a rivalry condition, where one half of the cylinder is perceptually suppressed, significantly more perceptual switches occur that are consistent with visual integration of the whole cylinder than occur in a control condition, where only half of the cylinder is presented at a time and the presentation of the two images is physically alternated. In Experiment 2, stimulation in the observer's dominant eye was kept dominant by presenting the half cylinder in this eye at higher contrast and by surrounding it with a flickering context. Results show that the strong convexity bias that was found in a control condition, where no stimulus was presented in the suppressed eye, almost completely disappears when the unseen half is presented in the suppressed eye, indicating that both halves visually integrate and, subsequently, compete for convexity. These findings provide evidence that unseen visual information is biased towards a representation that is congruent with the current visible representation and, hence, that principles of perceptual organization also apply to parts of the visual input that remain unseen by the observer.
Multimodal visual dictionary learning via heterogeneous latent semantic sparse coding
NASA Astrophysics Data System (ADS)
Li, Chenxiao; Ding, Guiguang; Zhou, Jile; Guo, Yuchen; Liu, Qiang
2014-11-01
Visual dictionary learning as a crucial task of image representation has gained increasing attention. Specifically, sparse coding is widely used due to its intrinsic advantage. In this paper, we propose a novel heterogeneous latent semantic sparse coding model. The central idea is to bridge heterogeneous modalities by capturing their common sparse latent semantic structure so that the learned visual dictionary is able to describe both the visual and textual properties of training data. Experiments on both image categorization and retrieval tasks demonstrate that our model shows superior performance over several recent methods such as K-means and Sparse Coding.
Optimal Dictionaries for Sparse Solutions of Multi-frame Blind Deconvolution
2014-09-01
Optimal Dictionaries for Sparse Solutions of Multi-frame Blind Deconvolution B. R. Hunt...overcomplete dictionaries from atmospheric turbulence data. Implications for blind - deconvolution of turbulent images are discussed. The application of sparse...dictionaries is demonstrated by the employment of sparse PSF representations to formulate a multi-frame blind deconvolution (MFBD) algorithm. We
Natural image sequences constrain dynamic receptive fields and imply a sparse code.
Häusler, Chris; Susemihl, Alex; Nawrot, Martin P
2013-11-06
In their natural environment, animals experience a complex and dynamic visual scenery. Under such natural stimulus conditions, neurons in the visual cortex employ a spatially and temporally sparse code. For the input scenario of natural still images, previous work demonstrated that unsupervised feature learning combined with the constraint of sparse coding can predict physiologically measured receptive fields of simple cells in the primary visual cortex. This convincingly indicated that the mammalian visual system is adapted to the natural spatial input statistics. Here, we extend this approach to the time domain in order to predict dynamic receptive fields that can account for both spatial and temporal sparse activation in biological neurons. We rely on temporal restricted Boltzmann machines and suggest a novel temporal autoencoding training procedure. When tested on a dynamic multi-variate benchmark dataset this method outperformed existing models of this class. Learning features on a large dataset of natural movies allowed us to model spatio-temporal receptive fields for single neurons. They resemble temporally smooth transformations of previously obtained static receptive fields and are thus consistent with existing theories. A neuronal spike response model demonstrates how the dynamic receptive field facilitates temporal and population sparseness. We discuss the potential mechanisms and benefits of a spatially and temporally sparse representation of natural visual input.
Analog system for computing sparse codes
Rozell, Christopher John; Johnson, Don Herrick; Baraniuk, Richard Gordon; Olshausen, Bruno A.; Ortman, Robert Lowell
2010-08-24
A parallel dynamical system for computing sparse representations of data, i.e., where the data can be fully represented in terms of a small number of non-zero code elements, and for reconstructing compressively sensed images. The system is based on the principles of thresholding and local competition that solves a family of sparse approximation problems corresponding to various sparsity metrics. The system utilizes Locally Competitive Algorithms (LCAs), nodes in a population continually compete with neighboring units using (usually one-way) lateral inhibition to calculate coefficients representing an input in an over complete dictionary.
Sparse Coding and Counting for Robust Visual Tracking
Liu, Risheng; Wang, Jing; Shang, Xiaoke; Wang, Yiyang; Su, Zhixun; Cai, Yu
2016-01-01
In this paper, we propose a novel sparse coding and counting method under Bayesian framework for visual tracking. In contrast to existing methods, the proposed method employs the combination of L0 and L1 norm to regularize the linear coefficients of incrementally updated linear basis. The sparsity constraint enables the tracker to effectively handle difficult challenges, such as occlusion or image corruption. To achieve real-time processing, we propose a fast and efficient numerical algorithm for solving the proposed model. Although it is an NP-hard problem, the proposed accelerated proximal gradient (APG) approach is guaranteed to converge to a solution quickly. Besides, we provide a closed solution of combining L0 and L1 regularized representation to obtain better sparsity. Experimental results on challenging video sequences demonstrate that the proposed method achieves state-of-the-art results both in accuracy and speed. PMID:27992474
Combined numerical and linguistic knowledge representation and its application to medical diagnosis
NASA Astrophysics Data System (ADS)
Meesad, Phayung; Yen, Gary G.
2002-07-01
In this study, we propose a novel hybrid intelligent system (HIS) which provides a unified integration of numerical and linguistic knowledge representations. The proposed HIS is hierarchical integration of an incremental learning fuzzy neural network (ILFN) and a linguistic model, i.e., fuzzy expert system, optimized via the genetic algorithm. The ILFN is a self-organizing network with the capability of fast, one-pass, online, and incremental learning. The linguistic model is constructed based on knowledge embedded in the trained ILFN or provided by the domain expert. The knowledge captured from the low-level ILFN can be mapped to the higher-level linguistic model and vice versa. The GA is applied to optimize the linguistic model to maintain high accuracy, comprehensibility, completeness, compactness, and consistency. After the system being completely constructed, it can incrementally learn new information in both numerical and linguistic forms. To evaluate the system's performance, the well-known benchmark Wisconsin breast cancer data set was studied for an application to medical diagnosis. The simulation results have shown that the prosed HIS perform better than the individual standalone systems. The comparison results show that the linguistic rules extracted are competitive with or even superior to some well-known methods.
Vectorized Sparse Elimination.
1984-03-01
Grids," Proc. 6th Symposium on Reservoir Simulation , New Orleans, Feb. 1-2, 1982, pp. 489-506. [51 Arya, S., and D. A. Calahan, "Optimal Scheduling of...of Computer Architecture on Direct Sparse Matrix Routines in Petroleum Reservoir Simulation ," Sparse Matrix Symposium, Fairfield Glade, TE, October
Adaptive feature extraction using sparse coding for machinery fault diagnosis
NASA Astrophysics Data System (ADS)
Liu, Haining; Liu, Chengliang; Huang, Yixiang
2011-02-01
In the signal processing domain, there has been growing interest in sparse coding with a learned dictionary instead of a predefined one, which is advocated as an effective mathematical description for the underlying principle of mammalian sensory systems in processing information. In this paper, sparse coding is introduced as a feature extraction technique for machinery fault diagnosis and an adaptive feature extraction scheme is proposed based on it. The two core problems of sparse coding, i.e., dictionary learning and coefficients solving, are discussed in detail. A natural extension of sparse coding, shift-invariant sparse coding, is also introduced. Then, the vibration signals of rolling element bearings are taken as the target signals to verify the proposed scheme, and shift-invariant sparse coding is used for vibration analysis. With the purpose of diagnosing the different fault conditions of bearings, features are extracted following the proposed scheme: basis functions are separately learned from each class of vibration signals trying to capture the defective impulses; a redundant dictionary is built by merging all the learned basis functions; based on the redundant dictionary, the diagnostic information is made explicit in the solved sparse representations of vibration signals; sparse features are formulated in terms of activations of atoms. The multiclass linear discriminant analysis (LDA) classifier is used to test the discriminability of the extracted sparse features and the adaptability of the learned atoms. The experiments show that sparse coding is an effective feature extraction technique for machinery fault diagnosis.
ERIC Educational Resources Information Center
Lem, Stephanie; Kempen, Goya; Ceulemans, Eva; Onghena, Patrick; Verschaffel, Lieven; Van Dooren, Wim
2015-01-01
Box plots are frequently misinterpreted and educational attempts to correct these misinterpretations have not been successful. In this study, we used two instructional techniques that seemed powerful to change the misinterpretation of the area of the box in box plots, both separately and in combination, leading to three experimental conditions,…
ERIC Educational Resources Information Center
Gagne, Christina L.; Spalding, Thomas L.; Ji, Hongbo
2005-01-01
In a recent study of conceptual combination, Estes (2003) presented evidence for the priming of relational information in the absence of shared constituents between the prime and target (e.g., "pancake spatula" was interpreted more quickly following "bacon tongs" than following "city riots"). He argued that these data support the view that…
NASA Astrophysics Data System (ADS)
Pinski, Peter; Riplinger, Christoph; Valeev, Edward F.; Neese, Frank
2015-07-01
In this work, a systematic infrastructure is described that formalizes concepts implicit in previous work and greatly simplifies computer implementation of reduced-scaling electronic structure methods. The key concept is sparse representation of tensors using chains of sparse maps between two index sets. Sparse map representation can be viewed as a generalization of compressed sparse row, a common representation of a sparse matrix, to tensor data. By combining few elementary operations on sparse maps (inversion, chaining, intersection, etc.), complex algorithms can be developed, illustrated here by a linear-scaling transformation of three-center Coulomb integrals based on our compact code library that implements sparse maps and operations on them. The sparsity of the three-center integrals arises from spatial locality of the basis functions and domain density fitting approximation. A novel feature of our approach is the use of differential overlap integrals computed in linear-scaling fashion for screening products of basis functions. Finally, a robust linear scaling domain based local pair natural orbital second-order Möller-Plesset (DLPNO-MP2) method is described based on the sparse map infrastructure that only depends on a minimal number of cutoff parameters that can be systematically tightened to approach 100% of the canonical MP2 correlation energy. With default truncation thresholds, DLPNO-MP2 recovers more than 99.9% of the canonical resolution of the identity MP2 (RI-MP2) energy while still showing a very early crossover with respect to the computational effort. Based on extensive benchmark calculations, relative energies are reproduced with an error of typically <0.2 kcal/mol. The efficiency of the local MP2 (LMP2) method can be drastically improved by carrying out the LMP2 iterations in a basis of pair natural orbitals. While the present work focuses on local electron correlation, it is of much broader applicability to computation with sparse tensors in
Pinski, Peter; Riplinger, Christoph; Neese, Frank E-mail: frank.neese@cec.mpg.de; Valeev, Edward F. E-mail: frank.neese@cec.mpg.de
2015-07-21
In this work, a systematic infrastructure is described that formalizes concepts implicit in previous work and greatly simplifies computer implementation of reduced-scaling electronic structure methods. The key concept is sparse representation of tensors using chains of sparse maps between two index sets. Sparse map representation can be viewed as a generalization of compressed sparse row, a common representation of a sparse matrix, to tensor data. By combining few elementary operations on sparse maps (inversion, chaining, intersection, etc.), complex algorithms can be developed, illustrated here by a linear-scaling transformation of three-center Coulomb integrals based on our compact code library that implements sparse maps and operations on them. The sparsity of the three-center integrals arises from spatial locality of the basis functions and domain density fitting approximation. A novel feature of our approach is the use of differential overlap integrals computed in linear-scaling fashion for screening products of basis functions. Finally, a robust linear scaling domain based local pair natural orbital second-order Möller-Plesset (DLPNO-MP2) method is described based on the sparse map infrastructure that only depends on a minimal number of cutoff parameters that can be systematically tightened to approach 100% of the canonical MP2 correlation energy. With default truncation thresholds, DLPNO-MP2 recovers more than 99.9% of the canonical resolution of the identity MP2 (RI-MP2) energy while still showing a very early crossover with respect to the computational effort. Based on extensive benchmark calculations, relative energies are reproduced with an error of typically <0.2 kcal/mol. The efficiency of the local MP2 (LMP2) method can be drastically improved by carrying out the LMP2 iterations in a basis of pair natural orbitals. While the present work focuses on local electron correlation, it is of much broader applicability to computation with sparse tensors in
Pinski, Peter; Riplinger, Christoph; Valeev, Edward F; Neese, Frank
2015-07-21
In this work, a systematic infrastructure is described that formalizes concepts implicit in previous work and greatly simplifies computer implementation of reduced-scaling electronic structure methods. The key concept is sparse representation of tensors using chains of sparse maps between two index sets. Sparse map representation can be viewed as a generalization of compressed sparse row, a common representation of a sparse matrix, to tensor data. By combining few elementary operations on sparse maps (inversion, chaining, intersection, etc.), complex algorithms can be developed, illustrated here by a linear-scaling transformation of three-center Coulomb integrals based on our compact code library that implements sparse maps and operations on them. The sparsity of the three-center integrals arises from spatial locality of the basis functions and domain density fitting approximation. A novel feature of our approach is the use of differential overlap integrals computed in linear-scaling fashion for screening products of basis functions. Finally, a robust linear scaling domain based local pair natural orbital second-order Möller-Plesset (DLPNO-MP2) method is described based on the sparse map infrastructure that only depends on a minimal number of cutoff parameters that can be systematically tightened to approach 100% of the canonical MP2 correlation energy. With default truncation thresholds, DLPNO-MP2 recovers more than 99.9% of the canonical resolution of the identity MP2 (RI-MP2) energy while still showing a very early crossover with respect to the computational effort. Based on extensive benchmark calculations, relative energies are reproduced with an error of typically <0.2 kcal/mol. The efficiency of the local MP2 (LMP2) method can be drastically improved by carrying out the LMP2 iterations in a basis of pair natural orbitals. While the present work focuses on local electron correlation, it is of much broader applicability to computation with sparse tensors in
Evolving sparse stellar populations
NASA Astrophysics Data System (ADS)
Bruzual, Gustavo; Gladis Magris, C.; Hernández-Pérez, Fabiola
2017-03-01
We examine the role that stochastic fluctuations in the IMF and in the number of interacting binaries have on the spectro-photometric properties of sparse stellar populations as a function of age and metallicity.
Local structure preserving sparse coding for infrared target recognition
Han, Jing; Yue, Jiang; Zhang, Yi; Bai, Lianfa
2017-01-01
Sparse coding performs well in image classification. However, robust target recognition requires a lot of comprehensive template images and the sparse learning process is complex. We incorporate sparsity into a template matching concept to construct a local sparse structure matching (LSSM) model for general infrared target recognition. A local structure preserving sparse coding (LSPSc) formulation is proposed to simultaneously preserve the local sparse and structural information of objects. By adding a spatial local structure constraint into the classical sparse coding algorithm, LSPSc can improve the stability of sparse representation for targets and inhibit background interference in infrared images. Furthermore, a kernel LSPSc (K-LSPSc) formulation is proposed, which extends LSPSc to the kernel space to weaken the influence of the linear structure constraint in nonlinear natural data. Because of the anti-interference and fault-tolerant capabilities, both LSPSc- and K-LSPSc-based LSSM can implement target identification based on a simple template set, which just needs several images containing enough local sparse structures to learn a sufficient sparse structure dictionary of a target class. Specifically, this LSSM approach has stable performance in the target detection with scene, shape and occlusions variations. High performance is demonstrated on several datasets, indicating robust infrared target recognition in diverse environments and imaging conditions. PMID:28323824
Symbol Systems and Pictorial Representations
NASA Astrophysics Data System (ADS)
Diederich, Joachim; Wright, Susan
All problem-solvers are subject to the same ultimate constraints -- limitations on space, time, and materials (Minsky, 1985). He introduces two principles: (1) Economics: Every intelligence must develop symbol-systems for representing objects, causes and goals, and (2) Sparseness: Every evolving intelligence will eventually encounter certain very special ideas -- e.g., about arithmetic, causal reasoning, and economics -- because these particular ideas are very much simpler than other ideas with similar uses. An extra-terrestrial intelligence (ETI) would have developed symbol systems to express these ideas and would have the capacity of multi-modal processing. Vakoch (1998) states that ...``ETI may rely significantly on other sensory modalities (than vision). Particularly useful representations would be ones that may be intelligible through more than one sensory modality. For instance, the information used to create a three-dimensional representation of an object might be intelligible to ETI heavily reliant on either visual or tactile sensory processes.'' The cross-modal representations Vakoch (1998) describes and the symbol systems Minsky (1985) proposes are called ``metaphors'' when combined. Metaphors allow for highly efficient communication. Metaphors are compact, condensed ways of expressing an idea: words, sounds, gestures or images are used in novel ways to refer to something they do not literally denote. Due to the importance of Minsky's ``economics'' principle, it is therefore possible that a message heavily relies on metaphors.
Menon, Samir; Brantner, Gerald; Aholt, Chris; Kay, Kendrick; Khatib, Oussama
2013-01-01
A challenging problem in motor control neuroimaging studies is the inability to perform complex human motor tasks given the Magnetic Resonance Imaging (MRI) scanner's disruptive magnetic fields and confined workspace. In this paper, we propose a novel experimental platform that combines Functional MRI (fMRI) neuroimaging, haptic virtual simulation environments, and an fMRI-compatible haptic device for real-time haptic interaction across the scanner workspace (above torso ∼ .65×.40×.20m(3)). We implement this Haptic fMRI platform with a novel haptic device, the Haptic fMRI Interface (HFI), and demonstrate its suitability for motor neuroimaging studies. HFI has three degrees-of-freedom (DOF), uses electromagnetic motors to enable high-fidelity haptic rendering (>350Hz), integrates radio frequency (RF) shields to prevent electromagnetic interference with fMRI (temporal SNR >100), and is kinematically designed to minimize currents induced by the MRI scanner's magnetic field during motor displacement (<2cm). HFI possesses uniform inertial and force transmission properties across the workspace, and has low friction (.05-.30N). HFI's RF noise levels, in addition, are within a 3 Tesla fMRI scanner's baseline noise variation (∼.85±.1%). Finally, HFI is haptically transparent and does not interfere with human motor tasks (tested for .4m reaches). By allowing fMRI experiments involving complex three-dimensional manipulation with haptic interaction, Haptic fMRI enables-for the first time-non-invasive neuroscience experiments involving interactive motor tasks, object manipulation, tactile perception, and visuo-motor integration.
A sparse Bayesian learning based scheme for multi-movement recognition using sEMG.
Ding, Shuai; Wang, Liang
2016-03-01
This paper proposed a feature extraction scheme based on sparse representation considering the non-stationary property of surface electromyography (sEMG). Sparse Bayesian learning was introduced to extract the feature with optimal class separability to improve recognition accuracy of multi-movement patterns. The extracted feature, sparse representation coefficients (SRC), represented time-varying characteristics of sEMG effectively because of the compressibility (or weak sparsity) of the signal in some transformed domains. We investigated the effect of the proposed feature by comparing with other fourteen individual features in offline recognition. The results demonstrated the proposed feature revealed important dynamic information in the sEMG signals. The multi-feature sets formed by the SRC and other single feature yielded more superior performance on recognition accuracy, compared with the single features. The best average recognition accuracy of 94.33% was gained by using SVM classifier with the multi-feature set combining the feature SRC, Williston amplitude (WAMP), wavelength (WL) and the coefficients of the fourth order autoregressive model (ARC4) via multiple kernel learning framework. The proposed feature extraction scheme (known as SRC + WAMP + WL + ARC4) is a promising method for multi-movement recognition with high accuracy.
On the decoding of intracranial data using sparse orthonormalized partial least squares
NASA Astrophysics Data System (ADS)
van Gerven, Marcel A. J.; Chao, Zenas C.; Heskes, Tom
2012-04-01
It has recently been shown that robust decoding of motor output from electrocorticogram signals in monkeys over prolonged periods of time has become feasible (Chao et al 2010 Front. Neuroeng. 3 1-10 ). In order to achieve these results, multivariate partial least-squares (PLS) regression was used. PLS uses a set of latent variables, referred to as components, to model the relationship between the input and the output data and is known to handle high-dimensional and possibly strongly correlated inputs and outputs well. We developed a new decoding method called sparse orthonormalized partial least squares (SOPLS) which was tested on a subset of the data used in Chao et al (2010) (freely obtainable from neurotycho.org (Nagasaka et al 2011 PLoS ONE 6 e22561)). We show that SOPLS reaches the same decoding performance as PLS using just two sparse components which can each be interpreted as encoding particular combinations of motor parameters. Furthermore, the sparse solution afforded by the SOPLS model allowed us to show the functional involvement of beta and gamma band responses in premotor and motor cortex for predicting the first component. Based on the literature, we conjecture that this first component is involved in the encoding of movement direction. Hence, the sparse and compact representation afforded by the SOPLS model facilitates interpretation of which spectral, spatial and temporal components are involved in successful decoding. These advantages make the proposed decoding method an important new tool in neuroprosthetics.
Local Sparse Structure Denoising for Low-Light-Level Image.
Han, Jing; Yue, Jiang; Zhang, Yi; Bai, Lianfa
2015-12-01
Sparse and redundant representations perform well in image denoising. However, sparsity-based methods fail to denoise low-light-level (LLL) images because of heavy and complex noise. They consider sparsity on image patches independently and tend to lose the texture structures. To suppress noises and maintain textures simultaneously, it is necessary to embed noise invariant features into the sparse decomposition process. We, therefore, used a local structure preserving sparse coding (LSPSc) formulation to explore the local sparse structures (both the sparsity and local structure) in image. It was found that, with the introduction of spatial local structure constraint into the general sparse coding algorithm, LSPSc could improve the robustness of sparse representation for patches in serious noise. We further used a kernel LSPSc (K-LSPSc) formulation, which extends LSPSc into the kernel space to weaken the influence of linear structure constraint in nonlinear data. Based on the robust LSPSc and K-LSPSc algorithms, we constructed a local sparse structure denoising (LSSD) model for LLL images, which was demonstrated to give high performance in the natural LLL images denoising, indicating that both the LSPSc- and K-LSPSc-based LSSD models have the stable property of noise inhibition and texture details preservation.
Sparse distributed memory overview
NASA Technical Reports Server (NTRS)
Raugh, Mike
1990-01-01
The Sparse Distributed Memory (SDM) project is investigating the theory and applications of massively parallel computing architecture, called sparse distributed memory, that will support the storage and retrieval of sensory and motor patterns characteristic of autonomous systems. The immediate objectives of the project are centered in studies of the memory itself and in the use of the memory to solve problems in speech, vision, and robotics. Investigation of methods for encoding sensory data is an important part of the research. Examples of NASA missions that may benefit from this work are Space Station, planetary rovers, and solar exploration. Sparse distributed memory offers promising technology for systems that must learn through experience and be capable of adapting to new circumstances, and for operating any large complex system requiring automatic monitoring and control. Sparse distributed memory is a massively parallel architecture motivated by efforts to understand how the human brain works. Sparse distributed memory is an associative memory, able to retrieve information from cues that only partially match patterns stored in the memory. It is able to store long temporal sequences derived from the behavior of a complex system, such as progressive records of the system's sensory data and correlated records of the system's motor controls.
Li, Dengwang; Liu, Li; Chen, Jinhu; Li, Hongsheng
2014-06-01
Purpose: The aiming of this study was to extract liver structures for daily Cone beam CT (CBCT) images automatically. Methods: Datasets were collected from 50 intravenous contrast planning CT images, which were regarded as training dataset for probabilistic atlas and shape prior model construction. Firstly, probabilistic atlas and shape prior model based on sparse shape composition (SSC) were constructed by iterative deformable registration. Secondly, the artifacts and noise were removed from the daily CBCT image by an edge-preserving filtering using total variation with L1 norm (TV-L1). Furthermore, the initial liver region was obtained by registering the incoming CBCT image with the atlas utilizing edge-preserving deformable registration with multi-scale strategy, and then the initial liver region was converted to surface meshing which was registered with the shape model where the major variation of specific patient was modeled by sparse vectors. At the last stage, the shape and intensity information were incorporated into joint probabilistic model, and finally the liver structure was extracted by maximum a posteriori segmentation.Regarding the construction process, firstly the manually segmented contours were converted into meshes, and then arbitrary patient data was chosen as reference image to register with the rest of training datasets by deformable registration algorithm for constructing probabilistic atlas and prior shape model. To improve the efficiency of proposed method, the initial probabilistic atlas was used as reference image to register with other patient data for iterative construction for removing bias caused by arbitrary selection. Results: The experiment validated the accuracy of the segmentation results quantitatively by comparing with the manually ones. The volumetric overlap percentage between the automatically generated liver contours and the ground truth were on an average 88%–95% for CBCT images. Conclusion: The experiment demonstrated
Slowness and sparseness have diverging effects on complex cell learning.
Lies, Jörn-Philipp; Häfner, Ralf M; Bethge, Matthias
2014-03-01
Following earlier studies which showed that a sparse coding principle may explain the receptive field properties of complex cells in primary visual cortex, it has been concluded that the same properties may be equally derived from a slowness principle. In contrast to this claim, we here show that slowness and sparsity drive the representations towards substantially different receptive field properties. To do so, we present complete sets of basis functions learned with slow subspace analysis (SSA) in case of natural movies as well as translations, rotations, and scalings of natural images. SSA directly parallels independent subspace analysis (ISA) with the only difference that SSA maximizes slowness instead of sparsity. We find a large discrepancy between the filter shapes learned with SSA and ISA. We argue that SSA can be understood as a generalization of the Fourier transform where the power spectrum corresponds to the maximally slow subspace energies in SSA. Finally, we investigate the trade-off between slowness and sparseness when combined in one objective function.
Parto Dezfouli, Mohammad Ali; Parto Dezfouli, Mohsen; Ahmadian, Alireza; Frangi, Alejandro F; Esmaeili Rad, Melika; Saligheh Rad, Hamidreza
2017-02-01
MRS is an analytical approach used for both quantitative and qualitative analysis of human body metabolites. The accurate and robust quantification capability of proton MRS ((1) H-MRS) enables the accurate estimation of living tissue metabolite concentrations. However, such methods can be efficiently employed for quantification of metabolite concentrations only if the overlapping nature of metabolites, existing static field inhomogeneity and low signal-to-noise ratio (SNR) are taken into consideration. Representation of (1) H-MRS signals in the time-frequency domain enables us to handle the baseline and noise better. This is possible because the MRS signal of each metabolite is sparsely represented, with only a few peaks, in the frequency domain, but still along with specific time-domain features such as distinct decay constant associated with T2 relaxation rate. The baseline, however, has a smooth behavior in the frequency domain. In this study, we proposed a quantification method using continuous wavelet transformation of (1) H-MRS signals in combination with sparse representation of features in the time-frequency domain. Estimation of the sparse representations of MR spectra is performed according to the dictionaries constructed from metabolite profiles. Results on simulated and phantom data show that the proposed method is able to quantify the concentration of metabolites in (1) H-MRS signals with high accuracy and robustness. This is achieved for both low SNR (5 dB) and low signal-to-baseline ratio (-5 dB) regimes.
A unified approach to sparse signal processing
NASA Astrophysics Data System (ADS)
Marvasti, Farokh; Amini, Arash; Haddadi, Farzan; Soltanolkotabi, Mahdi; Khalaj, Babak Hossein; Aldroubi, Akram; Sanei, Saeid; Chambers, Janathon
2012-12-01
A unified view of the area of sparse signal processing is presented in tutorial form by bringing together various fields in which the property of sparsity has been successfully exploited. For each of these fields, various algorithms and techniques, which have been developed to leverage sparsity, are described succinctly. The common potential benefits of significant reduction in sampling rate and processing manipulations through sparse signal processing are revealed. The key application domains of sparse signal processing are sampling, coding, spectral estimation, array processing, component analysis, and multipath channel estimation. In terms of the sampling process and reconstruction algorithms, linkages are made with random sampling, compressed sensing, and rate of innovation. The redundancy introduced by channel coding in finite and real Galois fields is then related to over-sampling with similar reconstruction algorithms. The error locator polynomial (ELP) and iterative methods are shown to work quite effectively for both sampling and coding applications. The methods of Prony, Pisarenko, and MUltiple SIgnal Classification (MUSIC) are next shown to be targeted at analyzing signals with sparse frequency domain representations. Specifically, the relations of the approach of Prony to an annihilating filter in rate of innovation and ELP in coding are emphasized; the Pisarenko and MUSIC methods are further improvements of the Prony method under noisy environments. The iterative methods developed for sampling and coding applications are shown to be powerful tools in spectral estimation. Such narrowband spectral estimation is then related to multi-source location and direction of arrival estimation in array processing. Sparsity in unobservable source signals is also shown to facilitate source separation in sparse component analysis; the algorithms developed in this area such as linear programming and matching pursuit are also widely used in compressed sensing. Finally
Structured Sparse Method for Hyperspectral Unmixing
NASA Astrophysics Data System (ADS)
Zhu, Feiyun; Wang, Ying; Xiang, Shiming; Fan, Bin; Pan, Chunhong
2014-02-01
Hyperspectral Unmixing (HU) has received increasing attention in the past decades due to its ability of unveiling information latent in hyperspectral data. Unfortunately, most existing methods fail to take advantage of the spatial information in data. To overcome this limitation, we propose a Structured Sparse regularized Nonnegative Matrix Factorization (SS-NMF) method based on the following two aspects. First, we incorporate a graph Laplacian to encode the manifold structures embedded in the hyperspectral data space. In this way, the highly similar neighboring pixels can be grouped together. Second, the lasso penalty is employed in SS-NMF for the fact that pixels in the same manifold structure are sparsely mixed by a common set of relevant bases. These two factors act as a new structured sparse constraint. With this constraint, our method can learn a compact space, where highly similar pixels are grouped to share correlated sparse representations. Experiments on real hyperspectral data sets with different noise levels demonstrate that our method outperforms the state-of-the-art methods significantly.
NASA Astrophysics Data System (ADS)
Ressler, Gerhard; Eicker, Annette; Lieb, Verena; Schmidt, Michael; Seitz, Florian; Shang, Kun; Shum, Che-Kwan
2015-04-01
Regionally changing hydrological conditions and their link to the availability of water for human consumption and agriculture is a challenging topic in the context of global change that is receiving increasing attention. Gravity field changes related to signals of land hydrology have been observed by the Gravity Recovery And Climate Experiment (GRACE) satellite mission over a period of more than 12 years. These changes are being analysed in our studies with respect to changing hydrological conditions, especially as a consequence of extreme weather situations and/or a change of climatic conditions. Typically, variations of the Earth's gravity field are modeled as a series expansion in terms of global spherical harmonics with time dependent harmonic coefficients. In order to investigate specific structures in the signal we alternatively apply a wavelet-based multi-resolution technique for the determination of regional spatiotemporal variations of the Earth's gravitational potential in combination with principal component analysis (PCA) for detailed evaluation of these structures. The multi-resolution representation (MRR) i.e. the composition of a signal considering different resolution levels is a suitable approach for spatial gravity modeling especially in case of inhomogeneous distribution of observation data on the one hand and because of the inhomogeneous structure of the Earth's gravity field itself on the other hand. In the MRR the signal is split into detail signals by applying low- and band-pass filters realized e.g. by spherical scaling and wavelet functions. Each detail signal is related to a specific resolution level and covers a certain part of the signal spectrum. Principal component analysis (PCA) enables for revealing specific signal patterns in the space as well as the time domain like trends and seasonal as well as semi seasonal variations. We apply the above mentioned combined technique to GRACE L1C residual potential differences that have been
A Multiobjective Sparse Feature Learning Model for Deep Neural Networks.
Gong, Maoguo; Liu, Jia; Li, Hao; Cai, Qing; Su, Linzhi
2015-12-01
Hierarchical deep neural networks are currently popular learning models for imitating the hierarchical architecture of human brain. Single-layer feature extractors are the bricks to build deep networks. Sparse feature learning models are popular models that can learn useful representations. But most of those models need a user-defined constant to control the sparsity of representations. In this paper, we propose a multiobjective sparse feature learning model based on the autoencoder. The parameters of the model are learnt by optimizing two objectives, reconstruction error and the sparsity of hidden units simultaneously to find a reasonable compromise between them automatically. We design a multiobjective induced learning procedure for this model based on a multiobjective evolutionary algorithm. In the experiments, we demonstrate that the learning procedure is effective, and the proposed multiobjective model can learn useful sparse features.
Stable Restoration and Separation of Approximately Sparse Signals
2011-07-02
dictionary. Particular applications covered by our framework include the restoration of signals impaired by impulse noise , narrowband interference, or...representation in a second general dictionary. Particular applications covered by our framework include the restoration of signals impaired by impulse noise ...applications (see [1–17] and references therein), including: • Impulse noise : The recovery of approximately sparse signals corrupted by impulse noise [13
Feature-Enhanced, Model-Based Sparse Aperture Imaging
2008-03-01
imaging, anisotropy characterization, feature-enhanced imaging, inverse problems, superresolution , anisotropy, sparse signal representation, overcomplete...number of such activities ourselves, and we provide very brief information on some of them here. We have developed a superresolution technique for...enhanced, superresolution image reconstruction. This framework provides a number of desirable features including preservation of anisotropic scatterers
Sparse inpainting and isotropy
NASA Astrophysics Data System (ADS)
Feeney, Stephen M.; Marinucci, Domenico; McEwen, Jason D.; Peiris, Hiranya V.; Wandelt, Benjamin D.; Cammarota, Valentina
2014-01-01
Sparse inpainting techniques are gaining in popularity as a tool for cosmological data analysis, in particular for handling data which present masked regions and missing observations. We investigate here the relationship between sparse inpainting techniques using the spherical harmonic basis as a dictionary and the isotropy properties of cosmological maps, as for instance those arising from cosmic microwave background (CMB) experiments. In particular, we investigate the possibility that inpainted maps may exhibit anisotropies in the behaviour of higher-order angular polyspectra. We provide analytic computations and simulations of inpainted maps for a Gaussian isotropic model of CMB data, suggesting that the resulting angular trispectrum may exhibit small but non-negligible deviations from isotropy.
Learning Discriminative Sparse Models for Source Separation and Mapping of Hyperspectral Imagery
2010-10-01
in a modification of (4), that allows sparse representations under bounded noise, known as basis pursuit denoising [21], or the Lasso [22]. Its...the atoms of the dictionary. There are several ways for learning these atoms. The K- SVD and Method of Orthogonal Directions (MOD) algorithms [25...Computation, vol. 12, no. 2, pp. 337–365, 2000. [27] M. Elad and M. Aharon, “Image denoising via learned dictionaries and sparse representation,” in
NASA Technical Reports Server (NTRS)
Kanerva, Pentti
1988-01-01
Theoretical models of the human brain and proposed neural-network computers are developed analytically. Chapters are devoted to the mathematical foundations, background material from computer science, the theory of idealized neurons, neurons as address decoders, and the search of memory for the best match. Consideration is given to sparse memory, distributed storage, the storage and retrieval of sequences, the construction of distributed memory, and the organization of an autonomous learning system.
Sparse matrix test collections
Duff, I.
1996-12-31
This workshop will discuss plans for coordinating and developing sets of test matrices for the comparison and testing of sparse linear algebra software. We will talk of plans for the next release (Release 2) of the Harwell-Boeing Collection and recent work on improving the accessibility of this Collection and others through the World Wide Web. There will only be three talks of about 15 to 20 minutes followed by a discussion from the floor.
Yin, Junming; Chen, Xi; Xing, Eric P.
2016-01-01
We consider the problem of sparse variable selection in nonparametric additive models, with the prior knowledge of the structure among the covariates to encourage those variables within a group to be selected jointly. Previous works either study the group sparsity in the parametric setting (e.g., group lasso), or address the problem in the nonparametric setting without exploiting the structural information (e.g., sparse additive models). In this paper, we present a new method, called group sparse additive models (GroupSpAM), which can handle group sparsity in additive models. We generalize the ℓ1/ℓ2 norm to Hilbert spaces as the sparsity-inducing penalty in GroupSpAM. Moreover, we derive a novel thresholding condition for identifying the functional sparsity at the group level, and propose an efficient block coordinate descent algorithm for constructing the estimate. We demonstrate by simulation that GroupSpAM substantially outperforms the competing methods in terms of support recovery and prediction accuracy in additive models, and also conduct a comparative experiment on a real breast cancer dataset.
Nanni, Loris; Lumini, Alessandra; Brahnam, Sheryl
2013-03-01
Many domains have a stake in the development of reliable systems for automatic protein classification. Of particular interest in recent studies of automatic protein classification is the exploration of new methods for extracting features from a protein that enhance classification for specific problems. These methods have proven very useful in one or two domains, but they have failed to generalize well across several domains (i.e. classification problems). In this paper, we evaluate several feature extraction approaches for representing proteins with the aim of sequence-based protein classification. Several protein representations are evaluated, those starting from: the position specific scoring matrix (PSSM) of the proteins; the amino-acid sequence; a matrix representation of the protein, of dimension (length of the protein) ×20, obtained using the substitution matrices for representing each amino-acid as a vector. A valuable result is that a texture descriptor can be extracted from the PSSM protein representation which improves the performance of standard descriptors based on the PSSM representation. Experimentally, we develop our systems by comparing several protein descriptors on nine different datasets. Each descriptor is used to train a support vector machine (SVM) or an ensemble of SVM. Although different stand-alone descriptors work well on some datasets (but not on others), we have discovered that fusion among classifiers trained using different descriptors obtains a good performance across all the tested datasets. Matlab code/Datasets used in the proposed paper are available at http://www.bias.csr.unibo.it\
Tang, Xin; Feng, Guo-can; Li, Xiao-xin; Cai, Jia-xin
2015-01-01
Face recognition is challenging especially when the images from different persons are similar to each other due to variations in illumination, expression, and occlusion. If we have sufficient training images of each person which can span the facial variations of that person under testing conditions, sparse representation based classification (SRC) achieves very promising results. However, in many applications, face recognition often encounters the small sample size problem arising from the small number of available training images for each person. In this paper, we present a novel face recognition framework by utilizing low-rank and sparse error matrix decomposition, and sparse coding techniques (LRSE+SC). Firstly, the low-rank matrix recovery technique is applied to decompose the face images per class into a low-rank matrix and a sparse error matrix. The low-rank matrix of each individual is a class-specific dictionary and it captures the discriminative feature of this individual. The sparse error matrix represents the intra-class variations, such as illumination, expression changes. Secondly, we combine the low-rank part (representative basis) of each person into a supervised dictionary and integrate all the sparse error matrix of each individual into a within-individual variant dictionary which can be applied to represent the possible variations between the testing and training images. Then these two dictionaries are used to code the query image. The within-individual variant dictionary can be shared by all the subjects and only contribute to explain the lighting conditions, expressions, and occlusions of the query image rather than discrimination. At last, a reconstruction-based scheme is adopted for face recognition. Since the within-individual dictionary is introduced, LRSE+SC can handle the problem of the corrupted training data and the situation that not all subjects have enough samples for training. Experimental results show that our method achieves the
Xu, Yuan; Ding, Kun; Huo, Chunlei; Zhong, Zisha; Li, Haichang; Pan, Chunhong
2015-01-01
Very high resolution (VHR) image change detection is challenging due to the low discriminative ability of change feature and the difficulty of change decision in utilizing the multilevel contextual information. Most change feature extraction techniques put emphasis on the change degree description (i.e., in what degree the changes have happened), while they ignore the change pattern description (i.e., how the changes changed), which is of equal importance in characterizing the change signatures. Moreover, the simultaneous consideration of the classification robust to the registration noise and the multiscale region-consistent fusion is often neglected in change decision. To overcome such drawbacks, in this paper, a novel VHR image change detection method is proposed based on sparse change descriptor and robust discriminative dictionary learning. Sparse change descriptor combines the change degree component and the change pattern component, which are encoded by the sparse representation error and the morphological profile feature, respectively. Robust change decision is conducted by multiscale region-consistent fusion, which is implemented by the superpixel-level cosparse representation with robust discriminative dictionary and the conditional random field model. Experimental results confirm the effectiveness of the proposed change detection technique. PMID:25918748
Constructing a Nonnegative Low-Rank and Sparse Graph With Data-Adaptive Features.
Zhuang, Liansheng; Gao, Shenghua; Tang, Jinhui; Wang, Jingjing; Lin, Zhouchen; Ma, Yi; Yu, Nenghai
2015-11-01
This paper aims at constructing a good graph to discover the intrinsic data structures under a semisupervised learning setting. First, we propose to build a nonnegative low-rank and sparse (referred to as NNLRS) graph for the given data representation. In particular, the weights of edges in the graph are obtained by seeking a nonnegative low-rank and sparse reconstruction coefficients matrix that represents each data sample as a linear combination of others. The so-obtained NNLRS-graph captures both the global mixture of subspaces structure (by the low-rankness) and the locally linear structure (by the sparseness) of the data, hence it is both generative and discriminative. Second, as good features are extremely important for constructing a good graph, we propose to learn the data embedding matrix and construct the graph simultaneously within one framework, which is termed as NNLRS with embedded features (referred to as NNLRS-EF). Extensive NNLRS experiments on three publicly available data sets demonstrate that the proposed method outperforms the state-of-the-art graph construction method by a large margin for both semisupervised classification and discriminative analysis, which verifies the effectiveness of our proposed method.
and Drayton Munster, Miroslav Stoyanov
2013-09-20
Sparse Grids are the family of methods of choice for multidimensional integration and interpolation in low to moderate number of dimensions. The method is to select extend a one dimensional set of abscissas, weights and basis functions by taking a subset of all possible tensor products. The module provides the ability to create global and local approximations based on polynomials and wavelets. The software has three components, a library, a wrapper for the library that provides a command line interface via text files ad a MATLAB interface via the command line tool.
NASA Astrophysics Data System (ADS)
Fang, Jun; Zhang, Lizao; Duan, Huiping; Huang, Lei; Li, Hongbin
2016-05-01
The application of sparse representation to SAR/ISAR imaging has attracted much attention over the past few years. This new class of sparse representation based imaging methods present a number of unique advantages over conventional range-Doppler methods, the basic idea behind these works is to formulate SAR/ISAR imaging as a sparse signal recovery problem. In this paper, we propose a new two-dimensional pattern-coupled sparse Bayesian learning(SBL) method to capture the underlying cluster patterns of the ISAR target images. Based on this model, an expectation-maximization (EM) algorithm is developed to infer the maximum a posterior (MAP) estimate of the hyperparameters, along with the posterior distribution of the sparse signal. Experimental results demonstrate that the proposed method is able to achieve a substantial performance improvement over existing algorithms, including the conventional SBL method.
Cerebellar Functional Parcellation Using Sparse Dictionary Learning Clustering
Wang, Changqing; Kipping, Judy; Bao, Chenglong; Ji, Hui; Qiu, Anqi
2016-01-01
The human cerebellum has recently been discovered to contribute to cognition and emotion beyond the planning and execution of movement, suggesting its functional heterogeneity. We aimed to identify the functional parcellation of the cerebellum using information from resting-state functional magnetic resonance imaging (rs-fMRI). For this, we introduced a new data-driven decomposition-based functional parcellation algorithm, called Sparse Dictionary Learning Clustering (SDLC). SDLC integrates dictionary learning, sparse representation of rs-fMRI, and k-means clustering into one optimization problem. The dictionary is comprised of an over-complete set of time course signals, with which a sparse representation of rs-fMRI signals can be constructed. Cerebellar functional regions were then identified using k-means clustering based on the sparse representation of rs-fMRI signals. We solved SDLC using a multi-block hybrid proximal alternating method that guarantees strong convergence. We evaluated the reliability of SDLC and benchmarked its classification accuracy against other clustering techniques using simulated data. We then demonstrated that SDLC can identify biologically reasonable functional regions of the cerebellum as estimated by their cerebello-cortical functional connectivity. We further provided new insights into the cerebello-cortical functional organization in children. PMID:27199650
Biomedical time series clustering based on non-negative sparse coding and probabilistic topic model.
Wang, Jin; Liu, Ping; F H She, Mary; Nahavandi, Saeid; Kouzani, Abbas
2013-09-01
Biomedical time series clustering that groups a set of unlabelled temporal signals according to their underlying similarity is very useful for biomedical records management and analysis such as biosignals archiving and diagnosis. In this paper, a new framework for clustering of long-term biomedical time series such as electrocardiography (ECG) and electroencephalography (EEG) signals is proposed. Specifically, local segments extracted from the time series are projected as a combination of a small number of basis elements in a trained dictionary by non-negative sparse coding. A Bag-of-Words (BoW) representation is then constructed by summing up all the sparse coefficients of local segments in a time series. Based on the BoW representation, a probabilistic topic model that was originally developed for text document analysis is extended to discover the underlying similarity of a collection of time series. The underlying similarity of biomedical time series is well captured attributing to the statistic nature of the probabilistic topic model. Experiments on three datasets constructed from publicly available EEG and ECG signals demonstrates that the proposed approach achieves better accuracy than existing state-of-the-art methods, and is insensitive to model parameters such as length of local segments and dictionary size.
Adaptive sparse signal processing of on-orbit lightning data using learned dictionaries
NASA Astrophysics Data System (ADS)
Moody, Daniela I.; Smith, David A.; Hamlin, Timothy D.; Light, Tess E.; Suszcynsky, David M.
2013-05-01
For the past two decades, there has been an ongoing research effort at Los Alamos National Laboratory to learn more about the Earth's radiofrequency (RF) background utilizing satellite-based RF observations of terrestrial lightning. The Fast On-orbit Recording of Transient Events (FORTE) satellite provided a rich RF lighting database, comprising of five years of data recorded from its two RF payloads. While some classification work has been done previously on the FORTE RF database, application of modern pattern recognition techniques may advance lightning research in the scientific community and potentially improve on-orbit processing and event discrimination capabilities for future satellite payloads. We now develop and implement new event classification capability on the FORTE database using state-of-the-art adaptive signal processing combined with compressive sensing and machine learning techniques. The focus of our work is improved feature extraction using sparse representations in learned dictionaries. Conventional localized data representations for RF transients using analytical dictionaries, such as a short-time Fourier basis or wavelets, can be suitable for analyzing some types of signals, but not others. Instead, we learn RF dictionaries directly from data, without relying on analytical constraints or additional knowledge about the signal characteristics, using several established machine learning algorithms. Sparse classification features are extracted via matching pursuit search over the learned dictionaries, and used in conjunction with a statistical classifier to distinguish between lightning types. We present preliminary results of our work and discuss classification scenarios and future development.
Flexible sparse regularization
NASA Astrophysics Data System (ADS)
Lorenz, Dirk A.; Resmerita, Elena
2017-01-01
The seminal paper of Daubechies, Defrise, DeMol made clear that {{\\ell }}p spaces with p\\in [1,2) and p-powers of the corresponding norms are appropriate settings for dealing with reconstruction of sparse solutions of ill-posed problems by regularization. It seems that the case p = 1 provides the best results in most of the situations compared to the cases p\\in (1,2). An extensive literature gives great credit also to using {{\\ell }}p spaces with p\\in (0,1) together with the corresponding quasi-norms, although one has to tackle challenging numerical problems raised by the non-convexity of the quasi-norms. In any of these settings, either superlinear, linear or sublinear, the question of how to choose the exponent p has been not only a numerical issue, but also a philosophical one. In this work we introduce a more flexible way of sparse regularization by varying exponents. We introduce the corresponding functional analytic framework, that leaves the setting of normed spaces but works with so-called F-norms. One curious result is that there are F-norms which generate the ℓ 1 space, but they are strictly convex, while the ℓ 1-norm is just convex.
Resistant multiple sparse canonical correlation.
Coleman, Jacob; Replogle, Joseph; Chandler, Gabriel; Hardin, Johanna
2016-04-01
Canonical correlation analysis (CCA) is a multivariate technique that takes two datasets and forms the most highly correlated possible pairs of linear combinations between them. Each subsequent pair of linear combinations is orthogonal to the preceding pair, meaning that new information is gleaned from each pair. By looking at the magnitude of coefficient values, we can find out which variables can be grouped together, thus better understanding multiple interactions that are otherwise difficult to compute or grasp intuitively. CCA appears to have quite powerful applications to high-throughput data, as we can use it to discover, for example, relationships between gene expression and gene copy number variation. One of the biggest problems of CCA is that the number of variables (often upwards of 10,000) makes biological interpretation of linear combinations nearly impossible. To limit variable output, we have employed a method known as sparse canonical correlation analysis (SCCA), while adding estimation which is resistant to extreme observations or other types of deviant data. In this paper, we have demonstrated the success of resistant estimation in variable selection using SCCA. Additionally, we have used SCCA to find multiple canonical pairs for extended knowledge about the datasets at hand. Again, using resistant estimators provided more accurate estimates than standard estimators in the multiple canonical correlation setting. R code is available and documented at https://github.com/hardin47/rmscca.
Adaptive sparse grid expansions of the vibrational Hamiltonian
Strobusch, D.; Scheurer, Ch.
2014-02-21
The vibrational Hamiltonian involves two high dimensional operators, the kinetic energy operator (KEO), and the potential energy surface (PES). Both must be approximated for systems involving more than a few atoms. Adaptive approximation schemes are not only superior to truncated Taylor or many-body expansions (MBE), they also allow for error estimates, and thus operators of predefined precision. To this end, modified sparse grids (SG) are developed that can be combined with adaptive MBEs. This MBE/SG hybrid approach yields a unified, fully adaptive representation of the KEO and the PES. Refinement criteria, based on the vibrational self-consistent field (VSCF) and vibrational configuration interaction (VCI) methods, are presented. The combination of the adaptive MBE/SG approach and the VSCF plus VCI methods yields a black box like procedure to compute accurate vibrational spectra. This is demonstrated on a test set of molecules, comprising water, formaldehyde, methanimine, and ethylene. The test set is first employed to prove convergence for semi-empirical PM3-PESs and subsequently to compute accurate vibrational spectra from CCSD(T)-PESs that agree well with experimental values.
A comparison of methods for representing sparsely sampled random quantities.
Romero, Vicente Jose; Swiler, Laura Painton; Urbina, Angel; Mullins, Joshua
2013-09-01
This report discusses the treatment of uncertainties stemming from relatively few samples of random quantities. The importance of this topic extends beyond experimental data uncertainty to situations involving uncertainty in model calibration, validation, and prediction. With very sparse data samples it is not practical to have a goal of accurately estimating the underlying probability density function (PDF). Rather, a pragmatic goal is that the uncertainty representation should be conservative so as to bound a specified percentile range of the actual PDF, say the range between 0.025 and .975 percentiles, with reasonable reliability. A second, opposing objective is that the representation not be overly conservative; that it minimally over-estimate the desired percentile range of the actual PDF. The presence of the two opposing objectives makes the sparse-data uncertainty representation problem interesting and difficult. In this report, five uncertainty representation techniques are characterized for their performance on twenty-one test problems (over thousands of trials for each problem) according to these two opposing objectives and other performance measures. Two of the methods, statistical Tolerance Intervals and a kernel density approach specifically developed for handling sparse data, exhibit significantly better overall performance than the others.
Duarte-Carvajalino, Julio Martin; Sapiro, Guillermo
2009-07-01
Sparse signal representation, analysis, and sensing have received a lot of attention in recent years from the signal processing, optimization, and learning communities. On one hand, learning overcomplete dictionaries that facilitate a sparse representation of the data as a liner combination of a few atoms from such dictionary leads to state-of-the-art results in image and video restoration and classification. On the other hand, the framework of compressed sensing (CS) has shown that sparse signals can be recovered from far less samples than those required by the classical Shannon-Nyquist Theorem. The samples used in CS correspond to linear projections obtained by a sensing projection matrix. It has been shown that, for example, a nonadaptive random sampling matrix satisfies the fundamental theoretical requirements of CS, enjoying the additional benefit of universality. On the other hand, a projection sensing matrix that is optimally designed for a certain class of signals can further improve the reconstruction accuracy or further reduce the necessary number of samples. In this paper, we introduce a framework for the joint design and optimization, from a set of training images, of the nonparametric dictionary and the sensing matrix. We show that this joint optimization outperforms both the use of random sensing matrices and those matrices that are optimized independently of the learning of the dictionary. Particular cases of the proposed framework include the optimization of the sensing matrix for a given dictionary as well as the optimization of the dictionary for a predefined sensing environment. The presentation of the framework and its efficient numerical optimization is complemented with numerous examples on classical image datasets.
Percolation on Sparse Networks
NASA Astrophysics Data System (ADS)
Karrer, Brian; Newman, M. E. J.; Zdeborová, Lenka
2014-11-01
We study percolation on networks, which is used as a model of the resilience of networked systems such as the Internet to attack or failure and as a simple model of the spread of disease over human contact networks. We reformulate percolation as a message passing process and demonstrate how the resulting equations can be used to calculate, among other things, the size of the percolating cluster and the average cluster size. The calculations are exact for sparse networks when the number of short loops in the network is small, but even on networks with many short loops we find them to be highly accurate when compared with direct numerical simulations. By considering the fixed points of the message passing process, we also show that the percolation threshold on a network with few loops is given by the inverse of the leading eigenvalue of the so-called nonbacktracking matrix.
NASA Technical Reports Server (NTRS)
Denning, Peter J.
1989-01-01
Sparse distributed memory was proposed be Pentti Kanerva as a realizable architecture that could store large patterns and retrieve them based on partial matches with patterns representing current sensory inputs. This memory exhibits behaviors, both in theory and in experiment, that resemble those previously unapproached by machines - e.g., rapid recognition of faces or odors, discovery of new connections between seemingly unrelated ideas, continuation of a sequence of events when given a cue from the middle, knowing that one doesn't know, or getting stuck with an answer on the tip of one's tongue. These behaviors are now within reach of machines that can be incorporated into the computing systems of robots capable of seeing, talking, and manipulating. Kanerva's theory is a break with the Western rationalistic tradition, allowing a new interpretation of learning and cognition that respects biology and the mysteries of individual human beings.
NASA Astrophysics Data System (ADS)
Moody, Daniela I.; Smith, David A.
2015-05-01
For over two decades, Los Alamos National Laboratory programs have included an active research effort utilizing satellite observations of terrestrial lightning to learn more about the Earth's RF background. The FORTE satellite provided a rich satellite lightning database, which has been previously used for some event classification, and remains relevant for advancing lightning research. Lightning impulses are dispersed as they travel through the ionosphere, appearing as nonlinear chirps at the receiver on orbit. The data processing challenge arises from the combined complexity of the lightning source model, the propagation medium nonlinearities, and the sensor artifacts. We continue to develop modern event classification capability on the FORTE database using adaptive signal processing combined with compressive sensing techniques. The focus of our work is improved feature extraction using sparse representations in overcomplete analytical dictionaries. We explore two possible techniques for detecting lightning events, and showcase the algorithms on few representative data examples. We present preliminary results of our work and discuss future development.
Group-Constrained Sparse FMRI Connectivity Modeling for Mild Cognitive Impairment Identification
Wee, Chong-Yaw; Yap, Pew-Thian; Zhang, Daoqiang; Wang, Lihong; Shen, Dinggang
2013-01-01
Emergence of advanced network analysis techniques utilizing resting-state functional Magnetic Resonance Imaging (R-fMRI) has enabled a more comprehensive understanding of neurological disorders at a whole-brain level. However, inferring brain connectivity from R-fMRI is a challenging task, particularly when the ultimate goal is to achieve good control-patient classification performance, owing to perplexing noise effects, curse of dimensionality, and inter-subject variability. Incorporating sparsity into connectivity modeling may be a possible solution to partially remedy this problem since most biological networks are intrinsically sparse. Nevertheless, sparsity constraint, when applied at an individual level, will inevitably cause inter-subject variability and hence degrade classification performance. To this end, we formulate the R-fMRI time series of each region-of-interest (ROI) as a linear representation of time series of other ROIs to infer sparse connectivity networks that are topologically identical across individuals. This formulation allows simultaneous selection of a common set of ROIs across subjects so that their linear combination is best in estimating the time series of the considered ROI. Specifically, l1-norm is imposed on each subject to filter out spurious or insignificant connections to produce sparse networks. A group-constraint is hence imposed via multi-task learning using a l2-norm to encourage consistent non-zero connections across subjects. This group-constraint is crucial since the network topology is identical for all subjects while still preserving individual information via different connectivity values. We validated the proposed modeling in mild cognitive impairment (MCI) identification and promising results achieved demonstrate its superiority in disease characterization, particularly greater sensitivity to early stage brain pathologies. The inferred group-constrained sparse network is found to be biologically plausible and is highly
NASA Astrophysics Data System (ADS)
Takayanagi, Toshiyuki; Shiga, Motoyuki
2002-08-01
The structures and vibrational frequencies of Cl 2-helium clusters have been studied using the path integral molecular dynamics method combined with the discrete-variable-representation approach. It is found that the Cl 2-helium clusters form clear shell structures comprised of rings around the Cl 2 bond. The vibrational frequencies calculated show a monotonically increasing red shift with an increase in cluster size. It can be concluded that the first solvation shell and its density around T-shaped configurations play the most important role in the observed frequency shifts.
Sparse nonnegative matrix factorization with ℓ0-constraints
Peharz, Robert; Pernkopf, Franz
2012-01-01
Although nonnegative matrix factorization (NMF) favors a sparse and part-based representation of nonnegative data, there is no guarantee for this behavior. Several authors proposed NMF methods which enforce sparseness by constraining or penalizing the ℓ1-norm of the factor matrices. On the other hand, little work has been done using a more natural sparseness measure, the ℓ0-pseudo-norm. In this paper, we propose a framework for approximate NMF which constrains the ℓ0-norm of the basis matrix, or the coefficient matrix, respectively. For this purpose, techniques for unconstrained NMF can be easily incorporated, such as multiplicative update rules, or the alternating nonnegative least-squares scheme. In experiments we demonstrate the benefits of our methods, which compare to, or outperform existing approaches. PMID:22505792
Spatiotemporal System Identification With Continuous Spatial Maps and Sparse Estimation.
Aram, Parham; Kadirkamanathan, Visakan; Anderson, Sean R
2015-11-01
We present a framework for the identification of spatiotemporal linear dynamical systems. We use a state-space model representation that has the following attributes: 1) the number of spatial observation locations are decoupled from the model order; 2) the model allows for spatial heterogeneity; 3) the model representation is continuous over space; and 4) the model parameters can be identified in a simple and sparse estimation procedure. The model identification procedure we propose has four steps: 1) decomposition of the continuous spatial field using a finite set of basis functions where spatial frequency analysis is used to determine basis function width and spacing, such that the main spatial frequency contents of the underlying field can be captured; 2) initialization of states in closed form; 3) initialization of state-transition and input matrix model parameters using sparse regression-the least absolute shrinkage and selection operator method; and 4) joint state and parameter estimation using an iterative Kalman-filter/sparse-regression algorithm. To investigate the performance of the proposed algorithm we use data generated by the Kuramoto model of spatiotemporal cortical dynamics. The identification algorithm performs successfully, predicting the spatiotemporal field with high accuracy, whilst the sparse regression leads to a compact model.
Sparse PDF Volumes for Consistent Multi-Resolution Volume Rendering
Sicat, Ronell; Krüger, Jens; Möller, Torsten; Hadwiger, Markus
2015-01-01
This paper presents a new multi-resolution volume representation called sparse pdf volumes, which enables consistent multi-resolution volume rendering based on probability density functions (pdfs) of voxel neighborhoods. These pdfs are defined in the 4D domain jointly comprising the 3D volume and its 1D intensity range. Crucially, the computation of sparse pdf volumes exploits data coherence in 4D, resulting in a sparse representation with surprisingly low storage requirements. At run time, we dynamically apply transfer functions to the pdfs using simple and fast convolutions. Whereas standard low-pass filtering and down-sampling incur visible differences between resolution levels, the use of pdfs facilitates consistent results independent of the resolution level used. We describe the efficient out-of-core computation of large-scale sparse pdf volumes, using a novel iterative simplification procedure of a mixture of 4D Gaussians. Finally, our data structure is optimized to facilitate interactive multi-resolution volume rendering on GPUs. PMID:26146475
Estimating sparse precision matrices
NASA Astrophysics Data System (ADS)
Padmanabhan, Nikhil; White, Martin; Zhou, Harrison H.; O'Connell, Ross
2016-08-01
We apply a method recently introduced to the statistical literature to directly estimate the precision matrix from an ensemble of samples drawn from a corresponding Gaussian distribution. Motivated by the observation that cosmological precision matrices are often approximately sparse, the method allows one to exploit this sparsity of the precision matrix to more quickly converge to an asymptotic 1/sqrt{N_sim} rate while simultaneously providing an error model for all of the terms. Such an estimate can be used as the starting point for further regularization efforts which can improve upon the 1/sqrt{N_sim} limit above, and incorporating such additional steps is straightforward within this framework. We demonstrate the technique with toy models and with an example motivated by large-scale structure two-point analysis, showing significant improvements in the rate of convergence. For the large-scale structure example, we find errors on the precision matrix which are factors of 5 smaller than for the sample precision matrix for thousands of simulations or, alternatively, convergence to the same error level with more than an order of magnitude fewer simulations.
Tiwari, Pallavi; Kurhanewicz, John; Viswanath, Satish; Sridhar, Akshay; Madabhushi, Anant
2011-01-01
Rationale and Objectives To develop a computerized data integration framework (MaWERiC) for quantitatively combining structural and metabolic information from different Magnetic Resonance (MR) imaging modalities. Materials and Methods In this paper, we present a novel computerized support system that we call Multimodal Wavelet Embedding Representation for data Combination (MaWERiC) which (1) employs wavelet theory and dimensionality reduction for providing a common, uniform representation of the different imaging (T2-w) and non-imaging (spectroscopy) MRI channels, and (2) leverages a random forest classifier for automated prostate cancer detection on a per voxel basis from combined 1.5 Tesla in vivo MRI and MRS. Results A total of 36 1.5 T endorectal in vivo T2-w MRI, MRS patient studies were evaluated on a per-voxel via MaWERiC, using a three-fold cross validation scheme across 25 iterations. Ground truth for evaluation of the results was obtained via ex-vivo whole-mount histology sections which served as the gold standard for expert radiologist annotations of prostate cancer on a per-voxel basis. The results suggest that MaWERiC based MRS-T2-w meta-classifier (mean AUC, μ = 0.89 ± 0.02) significantly outperformed (i) a T2-w MRI (employing wavelet texture features) classifier (μ = 0.55± 0.02), (ii) a MRS (employing metabolite ratios) classifier (μ= 0.77 ± 0.03), (iii) a decision-fusion classifier, obtained by combining individual T2-w MRI and MRS classifier outputs (μ = 0.85 ± 0.03) and (iv) a data combination scheme involving combination of metabolic MRS and MR signal intensity features (μ = 0.66± 0.02). Conclusion A novel data integration framework, MaWERiC, for combining imaging and non-imaging MRI channels was presented. Application to prostate cancer detection via combination of T2-w MRI and MRS data demonstrated significantly higher AUC and accuracy values compared to the individual T2-w MRI, MRS modalities and other data integration strategies
NASA Astrophysics Data System (ADS)
Zhu, Qiqi; Zhong, Yanfei; Zhang, Liangpei
2016-06-01
Topic modeling has been an increasingly mature method to bridge the semantic gap between the low-level features and high-level semantic information. However, with more and more high spatial resolution (HSR) images to deal with, conventional probabilistic topic model (PTM) usually presents the images with a dense semantic representation. This consumes more time and requires more storage space. In addition, due to the complex spectral and spatial information, a combination of multiple complementary features is proved to be an effective strategy to improve the performance for HSR image scene classification. But it should be noticed that how the distinct features are fused to fully describe the challenging HSR images, which is a critical factor for scene classification. In this paper, a semantic-feature fusion fully sparse topic model (SFF-FSTM) is proposed for HSR imagery scene classification. In SFF-FSTM, three heterogeneous features - the mean and standard deviation based spectral feature, wavelet based texture feature, and dense scale-invariant feature transform (SIFT) based structural feature are effectively fused at the latent semantic level. The combination of multiple semantic-feature fusion strategy and sparse based FSTM is able to provide adequate feature representations, and can achieve comparable performance with limited training samples. Experimental results on the UC Merced dataset and Google dataset of SIRI-WHU demonstrate that the proposed method can improve the performance of scene classification compared with other scene classification methods for HSR imagery.
ERIC Educational Resources Information Center
Schultz, James E.; Waters, Michael S.
2000-01-01
Discusses representations in the context of solving a system of linear equations. Views representations (concrete, tables, graphs, algebraic, matrices) from perspectives of understanding, technology, generalization, exact versus approximate solution, and learning style. (KHR)
Learn Sparse Dictionaries for Edit Propagation.
Xiaowu Chen; Jianwei Li; Dongqing Zou; Qinping Zhao
2016-04-01
With the increasing availability of high-resolution images, videos, and 3D models, the demand for scalable large data processing techniques increases. We introduce a method of sparse dictionary learning for edit propagation of large input data. Previous approaches for edit propagation typically employ a global optimization over the whole set of pixels (or vertexes), incurring a prohibitively high memory and time-consumption for large input data. Rather than propagating an edit pixel by pixel, we follow the principle of sparse representation to obtain a representative and compact dictionary and perform edit propagation on the dictionary instead. The sparse dictionary provides an intrinsic basis for input data, and the coding coefficients capture the linear relationship between all pixels and the dictionary atoms. The learned dictionary is then optimized by a novel scheme, which maximizes the Kullback-Leibler divergence between each atom pair to remove redundant atoms. To enable local edit propagation for images or videos with similar appearance, a dictionary learning strategy is proposed by considering range constraint to better account for the global distribution of pixels in their feature space. We show several applications of the sparsity-based edit propagation, including video recoloring, theme editing, and seamless cloning, operating on both color and texture features. Our approach can also be applied to computer graphics tasks, such as 3D surface deformation. We demonstrate that with an atom-to-pixel ratio in the order of 0.01% signifying a significant reduction on memory consumption, our method still maintains a high degree of visual fidelity.
Blind source separation by sparse decomposition
NASA Astrophysics Data System (ADS)
Zibulevsky, Michael; Pearlmutter, Barak A.
2000-04-01
The blind source separation problem is to extract the underlying source signals from a set of their linear mixtures, where the mixing matrix is unknown. This situation is common, eg in acoustics, radio, and medical signal processing. We exploit the property of the sources to have a sparse representation in a corresponding signal dictionary. Such a dictionary may consist of wavelets, wavelet packets, etc., or be obtained by learning from a given family of signals. Starting from the maximum a posteriori framework, which is applicable to the case of more sources than mixtures, we derive a few other categories of objective functions, which provide faster and more robust computations, when there are an equal number of sources and mixtures. Our experiments with artificial signals and with musical sounds demonstrate significantly better separation than other known techniques.
Integer sparse distributed memory: analysis and results.
Snaider, Javier; Franklin, Stan; Strain, Steve; George, E Olusegun
2013-10-01
Sparse distributed memory is an auto-associative memory system that stores high dimensional Boolean vectors. Here we present an extension of the original SDM, the Integer SDM that uses modular arithmetic integer vectors rather than binary vectors. This extension preserves many of the desirable properties of the original SDM: auto-associativity, content addressability, distributed storage, and robustness over noisy inputs. In addition, it improves the representation capabilities of the memory and is more robust over normalization. It can also be extended to support forgetting and reliable sequence storage. We performed several simulations that test the noise robustness property and capacity of the memory. Theoretical analyses of the memory's fidelity and capacity are also presented.
Joint Sparse Representation for Robust Multimodal Biometrics Recognition
2014-01-01
cancelable iris recognition in [17]. Nagesh and Li [18] presented an expression-invariant face recognition method using distributed CS and joint sparsity...80 85 90 95 100 Sparsity parameter (λ 1 ) R ec og nt io n R at e Rank one recognition across sparsity All modality Fingerprint Iris Fig. 5... irises and (c) all modalities. Results for composite kernels using different techniques is shown in figure (d). Finger 1 Finger 2 Finger 3 Finger 4 Iris 1
Joint Sparse Representation for Robust Multimodal Biometrics Recognition
2012-01-01
occlusion and random pixel corruption. Pillai et al. extended this work for robust cancelable iris recognition in [11]. Nagesh and Li [12] presented... Biometrics Recognition Sumit Shekhar, Student Member, IEEE, Vishal M. Patel, Member, IEEE, Nasser M. Nasrabadi, Fellow, IEEE, and Rama Chellappa, Fellow...IEEE . Abstract—Traditional biometric recognition systems rely on a single biometric signature for authentication. While the advan- tage of using
Sparse Data Representation: The Role of Redundancy in Data Processing
2005-09-13
effectively utilized in feature extraction and denoising . These redundant families can be frames, dictionaries, or libraries of bases. On the other hand...contains two kind of bases, which are natural but may be antagonistic : the singular value decomposition basis ( SVD ), which allows explicit
Learning Multiscale Sparse Representations for Image and Video Restoration
2007-07-01
25], and more recently to video denoising [35]. In this paper, we extend the basic K- SVD work, providing a framework for learning multiscale and...The original K- SVD denoising algorithm [1], the extensions to color image denoising , non-homogeneous noise, and inpainting [25], and the K- SVD for...section, we briefly review these algorithms. 2.1. The grayscale image denoising K- SVD algorithm. We now briefly review the main ideas of the K- SVD
Sparse Representation Based Multiple Frame Video Super-Resolution.
Dai, Qiqin; Yoo, Seunghwan; Kappeler, Armin; Katsaggelos, Aggelos K
2016-11-22
In this paper, we propose two multiple-frame superresolution (SR) algorithms based on dictionary learning and motion estimation. First, we adopt the use of video bilevel dictionary learning which has been used for single-frame SR. It is extended to multiple frames by using motion estimation with subpixel accuracy. We propose a batch and a temporally recursive multi-frame SR algorithm, which improve over single frame SR. Finally, we propose a novel dictionary learning algorithm utilizing consecutive video frames, rather than still images or individual video frames, which further improves the performance of the video SR algorithms. Extensive experimental comparisons with state-of-the-art SR algorithms verify the effectiveness of our proposed multiple-frame video SR approach.
Sparse Distributed Representation and Hierarchy: Keys to Scalable Machine Intelligence
2016-04-01
Government. Report contains color. 14. ABSTRACT We developed and tested a cortically-inspired model of spatiotemporal pattern learning and recognition...analogs of the canonical cortical processing module known as macrocolumns. Sparsey differs from mainstream neural models, e.g., Deep Learning , in many...ways including: a) it uses single-trial, Hebbian learning rather than incremental, many-trial, gradient- based learning ; and b) it multiplicatively
Solving Boltzmann and Fokker-Planck Equations Using Sparse Representation
2011-05-31
material science. We have com- puted the electronic structure of 2D quantum dot system, and compared the efficiency with the benchmark software OCTOPUS . For...one self-consistent iteration step with 512 electrons, OCTOPUS costs 1091 sec, and selected inversion costs 9.76 sec. The algorithm exhibits
Learning doubly sparse transforms for images.
Ravishankar, Saiprasad; Bresler, Yoram
2013-12-01
The sparsity of images in a transform domain or dictionary has been exploited in many applications in image processing. For example, analytical sparsifying transforms, such as wavelets and discrete cosine transform (DCT), have been extensively used in compression standards. Recently, synthesis sparsifying dictionaries that are directly adapted to the data have become popular especially in applications such as image denoising. Following up on our recent research, where we introduced the idea of learning square sparsifying transforms, we propose here novel problem formulations for learning doubly sparse transforms for signals or image patches. These transforms are a product of a fixed, fast analytic transform such as the DCT, and an adaptive matrix constrained to be sparse. Such transforms can be learnt, stored, and implemented efficiently. We show the superior promise of our learnt transforms as compared with analytical sparsifying transforms such as the DCT for image representation. We also show promising performance in image denoising that compares favorably with approaches involving learnt synthesis dictionaries such as the K-SVD algorithm. The proposed approach is also much faster than K-SVD denoising.
Cellular Adaptation Facilitates Sparse and Reliable Coding in Sensory Pathways
Farkhooi, Farzad; Froese, Anja; Muller, Eilif; Menzel, Randolf; Nawrot, Martin P.
2013-01-01
Most neurons in peripheral sensory pathways initially respond vigorously when a preferred stimulus is presented, but adapt as stimulation continues. It is unclear how this phenomenon affects stimulus coding in the later stages of sensory processing. Here, we show that a temporally sparse and reliable stimulus representation develops naturally in sequential stages of a sensory network with adapting neurons. As a modeling framework we employ a mean-field approach together with an adaptive population density treatment, accompanied by numerical simulations of spiking neural networks. We find that cellular adaptation plays a critical role in the dynamic reduction of the trial-by-trial variability of cortical spike responses by transiently suppressing self-generated fast fluctuations in the cortical balanced network. This provides an explanation for a widespread cortical phenomenon by a simple mechanism. We further show that in the insect olfactory system cellular adaptation is sufficient to explain the emergence of the temporally sparse and reliable stimulus representation in the mushroom body. Our results reveal a generic, biophysically plausible mechanism that can explain the emergence of a temporally sparse and reliable stimulus representation within a sequential processing architecture. PMID:24098101
Sparse recovery of the multimodal and dispersive characteristics of Lamb waves.
Harley, Joel B; Moura, José M F
2013-05-01
Guided waves in plates, known as Lamb waves, are characterized by complex, multimodal, and frequency dispersive wave propagation, which distort signals and make their analysis difficult. Estimating these multimodal and dispersive characteristics from experimental data becomes a difficult, underdetermined inverse problem. To accurately and robustly recover these multimodal and dispersive properties, this paper presents a methodology referred to as sparse wavenumber analysis based on sparse recovery methods. By utilizing a general model for Lamb waves, waves propagating in a plate structure, and robust l1 optimization strategies, sparse wavenumber analysis accurately recovers the Lamb wave's frequency-wavenumber representation with a limited number of surface mounted transducers. This is demonstrated with both simulated and experimental data in the presence of multipath reflections. With accurate frequency-wavenumber representations, sparse wavenumber synthesis is then used to accurately remove multipath interference in each measurement and predict the responses between arbitrary points on a plate.
Normalized Neural Representations of Complex Odors
2016-01-01
The olfactory system removes correlations in natural odors using a network of inhibitory neurons in the olfactory bulb. It has been proposed that this network integrates the response from all olfactory receptors and inhibits them equally. However, how such global inhibition influences the neural representations of odors is unclear. Here, we study a simple statistical model of the processing in the olfactory bulb, which leads to concentration-invariant, sparse representations of the odor composition. We show that the inhibition strength can be tuned to obtain sparse representations that are still useful to discriminate odors that vary in relative concentration, size, and composition. The model reveals two generic consequences of global inhibition: (i) odors with many molecular species are more difficult to discriminate and (ii) receptor arrays with heterogeneous sensitivities perform badly. Comparing these predictions to experiments will help us to understand the role of global inhibition in shaping normalized odor representations in the olfactory bulb. PMID:27835696
Intrapartum fetal heart rate classification from trajectory in Sparse SVM feature space.
Spilka, J; Frecon, J; Leonarduzzi, R; Pustelnik, N; Abry, P; Doret, M
2015-01-01
Intrapartum fetal heart rate (FHR) constitutes a prominent source of information for the assessment of fetal reactions to stress events during delivery. Yet, early detection of fetal acidosis remains a challenging signal processing task. The originality of the present contribution are three-fold: multiscale representations and wavelet leader based multifractal analysis are used to quantify FHR variability ; Supervised classification is achieved by means of Sparse-SVM that aim jointly to achieve optimal detection performance and to select relevant features in a multivariate setting ; Trajectories in the feature space accounting for the evolution along time of features while labor progresses are involved in the construction of indices quantifying fetal health. The classification performance permitted by this combination of tools are quantified on a intrapartum FHR large database (≃ 1250 subjects) collected at a French academic public hospital.
A view of Kanerva's sparse distributed memory
NASA Technical Reports Server (NTRS)
Denning, P. J.
1986-01-01
Pentti Kanerva is working on a new class of computers, which are called pattern computers. Pattern computers may close the gap between capabilities of biological organisms to recognize and act on patterns (visual, auditory, tactile, or olfactory) and capabilities of modern computers. Combinations of numeric, symbolic, and pattern computers may one day be capable of sustaining robots. The overview of the requirements for a pattern computer, a summary of Kanerva's Sparse Distributed Memory (SDM), and examples of tasks this computer can be expected to perform well are given.
NASA Astrophysics Data System (ADS)
Kuvychko, Igor
2000-10-01
Vision is a part of a larger informational system that converts visual information into knowledge structures. These structures drive vision process, resolving ambiguity and uncertainty via feedback, and provide image understanding, that is an interpretation of visual information in terms of such knowledge models. The solution to Image Understanding problems is suggested in form of active multilevel hierarchical networks represented dually as discrete and continuous structures. Computational intelligence methods transform images into model-based knowledge representation. Certainty Dimension converts attractors in neural networks into fuzzy sets, preserving input-output relationships. Symbols naturally emerge in such networks. Symbolic Space is a dual structure that combines closed distributed space split by the set of fuzzy regions, and discrete set of symbols equivalent to the cores of regions represented as points in the Certainty dimension. Model Space carries knowledge in form of links and relations between the symbols, and supports graph, diagrammatic and topological operations. Composition of spaces works similar to M. Minsky frames and agents, Gerard Edelman's maps of maps, etc., combining machine learning, classification and analogy together with induction, deduction and other methods of higher level model-based reasoning. Based on such principles, an Image Understanding system can convert images into knowledge models, effectively resolving uncertainty and ambiguity via feedback projections and does not require supercomputers.
Sparse and powerful cortical spikes.
Wolfe, Jason; Houweling, Arthur R; Brecht, Michael
2010-06-01
Activity in cortical networks is heterogeneous, sparse and often precisely timed. The functional significance of sparseness and precise spike timing is debated, but our understanding of the developmental and synaptic mechanisms that shape neuronal discharge patterns has improved. Evidence for highly specialized, selective and abstract cortical response properties is accumulating. Singe-cell stimulation experiments demonstrate a high sensitivity of cortical networks to the action potentials of some, but not all, single neurons. It is unclear how this sensitivity of cortical networks to small perturbations comes about and whether it is a generic property of cortex. The unforeseen sensitivity to cortical spikes puts serious constraints on the nature of neural coding schemes.
Clutter Mitigation in Echocardiography Using Sparse Signal Separation
Turek, Javier S.; Elad, Michael; Yavneh, Irad
2015-01-01
In ultrasound imaging, clutter artifacts degrade images and may cause inaccurate diagnosis. In this paper, we apply a method called Morphological Component Analysis (MCA) for sparse signal separation with the objective of reducing such clutter artifacts. The MCA approach assumes that the two signals in the additive mix have each a sparse representation under some dictionary of atoms (a matrix), and separation is achieved by finding these sparse representations. In our work, an adaptive approach is used for learning the dictionary from the echo data. MCA is compared to Singular Value Filtering (SVF), a Principal Component Analysis- (PCA-) based filtering technique, and to a high-pass Finite Impulse Response (FIR) filter. Each filter is applied to a simulated hypoechoic lesion sequence, as well as experimental cardiac ultrasound data. MCA is demonstrated in both cases to outperform the FIR filter and obtain results comparable to the SVF method in terms of contrast-to-noise ratio (CNR). Furthermore, MCA shows a lower impact on tissue sections while removing the clutter artifacts. In experimental heart data, MCA obtains in our experiments clutter mitigation with an average CNR improvement of 1.33 dB. PMID:26199622
Sparse feature fidelity for perceptual image quality assessment.
Chang, Hua-Wen; Yang, Hua; Gan, Yong; Wang, Ming-Hui
2013-10-01
The prediction of an image quality metric (IQM) should be consistent with subjective human evaluation. As the human visual system (HVS) is critical to visual perception, modeling of the HVS is regarded as the most suitable way to achieve perceptual quality predictions. Sparse coding that is equivalent to independent component analysis (ICA) can provide a very good description of the receptive fields of simple cells in the primary visual cortex, which is the most important part of the HVS. With this inspiration, a quality metric called sparse feature fidelity (SFF) is proposed for full-reference image quality assessment (IQA) on the basis of transformation of images into sparse representations in the primary visual cortex. The proposed method is based on the sparse features that are acquired by a feature detector, which is trained on samples of natural images by an ICA algorithm. In addition, two strategies are designed to simulate the properties of the visual perception: 1) visual attention and 2) visual threshold. The computation of SFF has two stages: training and fidelity computation, in addition, the fidelity computation consists of two components: feature similarity and luminance correlation. The feature similarity measures the structure differences between the two images, whereas the luminance correlation evaluates brightness distortions. SFF also reflects the chromatic properties of the HVS, and it is very effective for color IQA. The experimental results on five image databases show that SFF has a better performance in matching subjective ratings compared with the leading IQMs.
Hyperspherical Sparse Approximation Techniques for High-Dimensional Discontinuity Detection
Zhang, Guannan; Webster, Clayton G.; Gunzburger, Max; Burkardt, John
2016-08-04
This work proposes a hyperspherical sparse approximation framework for detecting jump discontinuities in functions in high-dimensional spaces. The need for a novel approach results from the theoretical and computational inefficiencies of well-known approaches, such as adaptive sparse grids, for discontinuity detection. Our approach constructs the hyperspherical coordinate representation of the discontinuity surface of a function. Then sparse approximations of the transformed function are built in the hyperspherical coordinate system, with values at each point estimated by solving a one-dimensional discontinuity detection problem. Due to the smoothness of the hypersurface, the new technique can identify jump discontinuities with significantly reduced computational cost, compared to existing methods. Several approaches are used to approximate the transformed discontinuity surface in the hyperspherical system, including adaptive sparse grid and radial basis function interpolation, discrete least squares projection, and compressed sensing approximation. Moreover, hierarchical acceleration techniques are also incorporated to further reduce the overall complexity. In conclusion, rigorous complexity analyses of the new methods are provided, as are several numerical examples that illustrate the effectiveness of our approach.
Hyperspherical Sparse Approximation Techniques for High-Dimensional Discontinuity Detection
Zhang, Guannan; Webster, Clayton G.; Gunzburger, Max; ...
2016-08-04
This work proposes a hyperspherical sparse approximation framework for detecting jump discontinuities in functions in high-dimensional spaces. The need for a novel approach results from the theoretical and computational inefficiencies of well-known approaches, such as adaptive sparse grids, for discontinuity detection. Our approach constructs the hyperspherical coordinate representation of the discontinuity surface of a function. Then sparse approximations of the transformed function are built in the hyperspherical coordinate system, with values at each point estimated by solving a one-dimensional discontinuity detection problem. Due to the smoothness of the hypersurface, the new technique can identify jump discontinuities with significantly reduced computationalmore » cost, compared to existing methods. Several approaches are used to approximate the transformed discontinuity surface in the hyperspherical system, including adaptive sparse grid and radial basis function interpolation, discrete least squares projection, and compressed sensing approximation. Moreover, hierarchical acceleration techniques are also incorporated to further reduce the overall complexity. In conclusion, rigorous complexity analyses of the new methods are provided, as are several numerical examples that illustrate the effectiveness of our approach.« less
Adaptive sparse signal processing of on-orbit lightning data using learned dictionaries
NASA Astrophysics Data System (ADS)
Moody, D. I.; Hamlin, T.; Light, T. E.; Loveland, R. C.; Smith, D. A.; Suszcynsky, D. M.
2012-12-01
For the past two decades, there has been an ongoing research effort at Los Alamos National Laboratory (LANL) to learn more about the Earth's radiofrequency (RF) background utilizing satellite-based RF observations of terrestrial lightning. Arguably the richest satellite lightning database ever recorded is that from the Fast On-orbit Recording of Transient Events (FORTE) satellite, which returned at least five years of data from its two RF payloads after launch in 1997. While some classification work has been done previously on the LANL FORTE RF database, application of modern pattern recognition techniques may further lightning research in the scientific community and potentially improve on-orbit processing and event discrimination capabilities for future satellite payloads. We now develop and implement new event classification capability on the FORTE database using state-of-the-art adaptive signal processing combined with compressive sensing and machine learning techniques. The focus of our work is improved feature extraction using sparse representations in learned dictionaries. Extracting classification features from RF signals typically relies on knowledge of the application domain in order to find feature vectors unique to a signal class and robust against background noise. Conventional localized data representations for RF transients using analytical dictionaries, such as a short-time Fourier basis or wavelets, can be suitable for analyzing some types of signals, but not others. Instead, we learn RF dictionaries directly from data, without relying on analytical constraints or additional knowledge about the signal characteristics, using several established machine learning algorithms. Sparse classification features are extracted via matching pursuit search over the learned dictionaries, and used in conjunction with a statistical classifier to distinguish between lightning types. We present preliminary results of our work and discuss classification performance
NASA Astrophysics Data System (ADS)
Juhler, Martin Vogt
2016-08-01
Recent research, both internationally and in Norway, has clearly expressed concerns about missing connections between subject-matter knowledge, pedagogical competence and real-life practice in schools. This study addresses this problem within the domain of field practice in teacher education, studying pre-service teachers' planning of a Physics lesson. Two means of intervention were introduced. The first was lesson study, which is a method for planning, carrying out and reflecting on a research lesson in detail with a learner and content-centered focus. This was used in combination with a second means, content representations, which is a systematic tool that connects overall teaching aims with pedagogical prompts. Changes in teaching were assessed through the construct of pedagogical content knowledge (PCK). A deductive coding analysis was carried out for this purpose. Transcripts of pre-service teachers' planning of a Physics lesson were coded into four main PCK categories, which were thereafter divided into 16 PCK sub-categories. The results showed that the intervention affected the pre-service teachers' potential to start developing PCK. First, they focused much more on categories concerning the learners. Second, they focused far more uniformly in all of the four main categories comprising PCK. Consequently, these differences could affect their potential to start developing PCK.
Stochastic convex sparse principal component analysis.
Baytas, Inci M; Lin, Kaixiang; Wang, Fei; Jain, Anil K; Zhou, Jiayu
2016-12-01
Principal component analysis (PCA) is a dimensionality reduction and data analysis tool commonly used in many areas. The main idea of PCA is to represent high-dimensional data with a few representative components that capture most of the variance present in the data. However, there is an obvious disadvantage of traditional PCA when it is applied to analyze data where interpretability is important. In applications, where the features have some physical meanings, we lose the ability to interpret the principal components extracted by conventional PCA because each principal component is a linear combination of all the original features. For this reason, sparse PCA has been proposed to improve the interpretability of traditional PCA by introducing sparsity to the loading vectors of principal components. The sparse PCA can be formulated as an ℓ1 regularized optimization problem, which can be solved by proximal gradient methods. However, these methods do not scale well because computation of the exact gradient is generally required at each iteration. Stochastic gradient framework addresses this challenge by computing an expected gradient at each iteration. Nevertheless, stochastic approaches typically have low convergence rates due to the high variance. In this paper, we propose a convex sparse principal component analysis (Cvx-SPCA), which leverages a proximal variance reduced stochastic scheme to achieve a geometric convergence rate. We further show that the convergence analysis can be significantly simplified by using a weak condition which allows a broader class of objectives to be applied. The efficiency and effectiveness of the proposed method are demonstrated on a large-scale electronic medical record cohort.
Sparse Regression by Projection and Sparse Discriminant Analysis.
Qi, Xin; Luo, Ruiyan; Carroll, Raymond J; Zhao, Hongyu
2015-04-01
Recent years have seen active developments of various penalized regression methods, such as LASSO and elastic net, to analyze high dimensional data. In these approaches, the direction and length of the regression coefficients are determined simultaneously. Due to the introduction of penalties, the length of the estimates can be far from being optimal for accurate predictions. We introduce a new framework, regression by projection, and its sparse version to analyze high dimensional data. The unique nature of this framework is that the directions of the regression coefficients are inferred first, and the lengths and the tuning parameters are determined by a cross validation procedure to achieve the largest prediction accuracy. We provide a theoretical result for simultaneous model selection consistency and parameter estimation consistency of our method in high dimension. This new framework is then generalized such that it can be applied to principal components analysis, partial least squares and canonical correlation analysis. We also adapt this framework for discriminant analysis. Compared to the existing methods, where there is relatively little control of the dependency among the sparse components, our method can control the relationships among the components. We present efficient algorithms and related theory for solving the sparse regression by projection problem. Based on extensive simulations and real data analysis, we demonstrate that our method achieves good predictive performance and variable selection in the regression setting, and the ability to control relationships between the sparse components leads to more accurate classification. In supplemental materials available online, the details of the algorithms and theoretical proofs, and R codes for all simulation studies are provided.
Sparse grid techniques for particle-in-cell schemes
NASA Astrophysics Data System (ADS)
Ricketson, L. F.; Cerfon, A. J.
2017-02-01
We propose the use of sparse grids to accelerate particle-in-cell (PIC) schemes. By using the so-called ‘combination technique’ from the sparse grids literature, we are able to dramatically increase the size of the spatial cells in multi-dimensional PIC schemes while paying only a slight penalty in grid-based error. The resulting increase in cell size allows us to reduce the statistical noise in the simulation without increasing total particle number. We present initial proof-of-principle results from test cases in two and three dimensions that demonstrate the new scheme’s efficiency, both in terms of computation time and memory usage.
Structure of a single whisker representation in layer 2 of mouse somatosensory cortex.
Clancy, Kelly B; Schnepel, Philipp; Rao, Antara T; Feldman, Daniel E
2015-03-04
Layer (L)2 is a major output of primary sensory cortex that exhibits very sparse spiking, but the structure of sensory representation in L2 is not well understood. We combined two-photon calcium imaging with deflection of many whiskers to map whisker receptive fields, characterize sparse coding, and quantitatively define the point representation in L2 of mouse somatosensory cortex. Neurons within a column-sized imaging field showed surprisingly heterogeneous, salt-and-pepper tuning to many different whiskers. Single whisker deflection elicited low-probability spikes in highly distributed, shifting neural ensembles spanning multiple cortical columns. Whisker-evoked response probability correlated strongly with spontaneous firing rate, but weakly with tuning properties, indicating a spectrum of inherent responsiveness across pyramidal cells. L2 neurons projecting to motor and secondary somatosensory cortex differed in whisker tuning and responsiveness, and carried different amounts of information about columnar whisker deflection. From these data, we derive a quantitative, fine-scale picture of the distributed point representation in L2.
1982-10-27
sparse matrices as well as other areas. Contents 1. operations on Sparse Matrices .. . . . . . . . . . . . . . . . . . . . . . . . I 1.1 Multi...22 2.1.1 Nonsymmetric systems ............................................. 22 2.1.1.1 General sparse matrices ...46 2.1.2.1 General sparse matrices ......................................... 46 2.1.2.2 Band or profile forms
A Probabilistic Analysis of Sparse Coded Feature Pooling and Its Application for Image Retrieval
Zhang, Yunchao; Chen, Jing; Huang, Xiujie; Wang, Yongtian
2015-01-01
Feature coding and pooling as a key component of image retrieval have been widely studied over the past several years. Recently sparse coding with max-pooling is regarded as the state-of-the-art for image classification. However there is no comprehensive study concerning the application of sparse coding for image retrieval. In this paper, we first analyze the effects of different sampling strategies for image retrieval, then we discuss feature pooling strategies on image retrieval performance with a probabilistic explanation in the context of sparse coding framework, and propose a modified sum pooling procedure which can improve the retrieval accuracy significantly. Further we apply sparse coding method to aggregate multiple types of features for large-scale image retrieval. Extensive experiments on commonly-used evaluation datasets demonstrate that our final compact image representation improves the retrieval accuracy significantly. PMID:26132080
Sparse-based multispectral image encryption via ptychography
NASA Astrophysics Data System (ADS)
Rawat, Nitin; Shi, Yishi; Kim, Byoungho; Lee, Byung-Geun
2015-12-01
Recently, we proposed a model of securing a ptychography-based monochromatic image encryption system via the classical Photon-counting imaging (PCI) technique. In this study, we examine a single-channel multispectral sparse-based photon-counting ptychography imaging (SMPI)-based cryptosystem. A ptychography-based cryptosystem creates a complex object wave field, which can be reconstructed by a series of diffraction intensity patterns through an aperture movement. The PCI sensor records only a few complex Bayer patterned samples that have been utilized in the decryption process. Sparse sensing and nonlinear properties of the classical PCI system, together with the scanning probes, enlarge the key space, and such a combination therefore enhances the system's security. We demonstrate that the sparse samples have adequate information for image decryption, as well as information authentication by means of optical correlation.
A note on rank reduction in sparse multivariate regression.
Chen, Kun; Chan, Kung-Sik
A reduced-rank regression with sparse singular value decomposition (RSSVD) approach was proposed by Chen et al. for conducting variable selection in a reduced-rank model. To jointly model the multivariate response, the method efficiently constructs a prespecified number of latent variables as some sparse linear combinations of the predictors. Here, we generalize the method to also perform rank reduction, and enable its usage in reduced-rank vector autoregressive (VAR) modeling to perform automatic rank determination and order selection. We show that in the context of stationary time-series data, the generalized approach correctly identifies both the model rank and the sparse dependence structure between the multivariate response and the predictors, with probability one asymptotically. We demonstrate the efficacy of the proposed method by simulations and analyzing a macro-economical multivariate time series using a reduced-rank VAR model.
Sparse and accurate high resolution SAR imaging
NASA Astrophysics Data System (ADS)
Vu, Duc; Zhao, Kexin; Rowe, William; Li, Jian
2012-05-01
We investigate the usage of an adaptive method, the Iterative Adaptive Approach (IAA), in combination with a maximum a posteriori (MAP) estimate to reconstruct high resolution SAR images that are both sparse and accurate. IAA is a nonparametric weighted least squares algorithm that is robust and user parameter-free. IAA has been shown to reconstruct SAR images with excellent side lobes suppression and high resolution enhancement. We first reconstruct the SAR images using IAA, and then we enforce sparsity by using MAP with a sparsity inducing prior. By coupling these two methods, we can produce a sparse and accurate high resolution image that are conducive for feature extractions and target classification applications. In addition, we show how IAA can be made computationally efficient without sacrificing accuracies, a desirable property for SAR applications where the size of the problems is quite large. We demonstrate the success of our approach using the Air Force Research Lab's "Gotcha Volumetric SAR Data Set Version 1.0" challenge dataset. Via the widely used FFT, individual vehicles contained in the scene are barely recognizable due to the poor resolution and high side lobe nature of FFT. However with our approach clear edges, boundaries, and textures of the vehicles are obtained.
Highly parallel sparse Cholesky factorization
NASA Technical Reports Server (NTRS)
Gilbert, John R.; Schreiber, Robert
1990-01-01
Several fine grained parallel algorithms were developed and compared to compute the Cholesky factorization of a sparse matrix. The experimental implementations are on the Connection Machine, a distributed memory SIMD machine whose programming model conceptually supplies one processor per data element. In contrast to special purpose algorithms in which the matrix structure conforms to the connection structure of the machine, the focus is on matrices with arbitrary sparsity structure. The most promising algorithm is one whose inner loop performs several dense factorizations simultaneously on a 2-D grid of processors. Virtually any massively parallel dense factorization algorithm can be used as the key subroutine. The sparse code attains execution rates comparable to those of the dense subroutine. Although at present architectural limitations prevent the dense factorization from realizing its potential efficiency, it is concluded that a regular data parallel architecture can be used efficiently to solve arbitrarily structured sparse problems. A performance model is also presented and it is used to analyze the algorithms.
Clothed particle representation in quantum field theory: Mass renormalization
Korda, V.Yu.; Shebeko, A.V.
2004-10-15
We consider the neutral pion and nucleon fields interacting via the pseudoscalar (PS) Yukawa-type coupling. The method of unitary clothing transformations is used to handle the so-called clothed particle representation, where the total field Hamiltonian and the three boost operators in the instant form of relativistic dynamics take on the same sparse structure in the Hilbert space of hadronic states. In this approach the mass counterterms are cancelled (at least, partly) by commutators of the generators of clothing transformations and the field interaction operator. This allows the pion and nucleon mass shifts to be expressed through the corresponding three-dimensional integrals whose integrands depend on certain covariant combinations of the relevant three-momenta. The property provides the momentum independence of mass renormalization. The present results prove to be equivalent to the results obtained by Feynman techniques.
Sparse Matrices in MATLAB: Design and Implementation
NASA Technical Reports Server (NTRS)
Gilbert, John R.; Moler, Cleve; Schreiber, Robert
1992-01-01
The matrix computation language and environment MATLAB is extended to include sparse matrix storage and operations. The only change to the outward appearance of the MATLAB language is a pair of commands to create full or sparse matrices. Nearly all the operations of MATLAB now apply equally to full or sparse matrices, without any explicit action by the user. The sparse data structure represents a matrix in space proportional to the number of nonzero entries, and most of the operations compute sparse results in time proportional to the number of arithmetic operations on nonzeros.
Optimal parallel solution of sparse triangular systems
NASA Technical Reports Server (NTRS)
Alvarado, Fernando L.; Schreiber, Robert
1990-01-01
A method for the parallel solution of triangular sets of equations is described that is appropriate when there are many right-handed sides. By preprocessing, the method can reduce the number of parallel steps required to solve Lx = b compared to parallel forward or backsolve. Applications are to iterative solvers with triangular preconditioners, to structural analysis, or to power systems applications, where there may be many right-handed sides (not all available a priori). The inverse of L is represented as a product of sparse triangular factors. The problem is to find a factored representation of this inverse of L with the smallest number of factors (or partitions), subject to the requirement that no new nonzero elements be created in the formation of these inverse factors. A method from an earlier reference is shown to solve this problem. This method is improved upon by constructing a permutation of the rows and columns of L that preserves triangularity and allow for the best possible such partition. A number of practical examples and algorithmic details are presented. The parallelism attainable is illustrated by means of elimination trees and clique trees.
Analysis of image content recognition algorithm based on sparse coding and machine learning
NASA Astrophysics Data System (ADS)
Xiao, Yu
2017-03-01
This paper presents an image classification algorithm based on spatial sparse coding model and random forest. Firstly, SIFT feature extraction of the image; and then use the sparse encoding theory to generate visual vocabulary based on SIFT features, and using the visual vocabulary of SIFT features into a sparse vector; through the combination of regional integration and spatial sparse vector, the sparse vector gets a fixed dimension is used to represent the image; at last random forest classifier for image sparse vectors for training and testing, using the experimental data set for standard test Caltech-101 and Scene-15. The experimental results show that the proposed algorithm can effectively represent the features of the image and improve the classification accuracy. In this paper, we propose an innovative image recognition algorithm based on image segmentation, sparse coding and multi instance learning. This algorithm introduces the concept of multi instance learning, the image as a multi instance bag, sparse feature transformation by SIFT images as instances, sparse encoding model generation visual vocabulary as the feature space is mapped to the feature space through the statistics on the number of instances in bags, and then use the 1-norm SVM to classify images and generate sample weights to select important image features.
A strategy of car detection via sparse dictionary
NASA Astrophysics Data System (ADS)
Jin, Guo-Qing; Dong, Ying-Hui
2011-06-01
In recent years there is a growing interest in the study of sparse representation for object detection. These approaches heavily depend on local salient image patches, thus weakening the global contribution to the object identification of other less informative signals.Our generic approach not only employs the informative representation by linear transform, but also keeps all the spatial dependence implied among the objects. As an example,car images can be represented using parts from a vocabulary, along with spatial relations observed among them.Our approach is conducted with the quantitative measurement in developing the car detector at every stage. The theory underneath the optimal solution is the maximal mutual information carried out by the system. Our goal is to keep the maximal mutual information transmitted from stage to stage so that only the least uncertainty about the class identification remains based on the observation of classifier's output.
Sparse parallel transmission on randomly perturbed spiral k-space trajectory.
Pang, Yong; Jiang, Xiaohua; Zhang, Xiaoliang
2014-04-01
Combination of parallel transmission and sparse pulse is able to shorten the excitation by using both the coil sensitivity and sparse k-space, showing improved fast excitation capability over the use of parallel transmission alone. However, to design an optimal k-space trajectory for sparse parallel transmission is a challenging task. In this work, a randomly perturbed sparse k-space trajectory is designed by modifying the path of a spiral trajectory along the sparse k-space data, and the sparse parallel transmission RF pulses are subsequently designed based on this optimal trajectory. This method combines the parallel transmission and sparse spiral k-space trajectory, potentially to further reduce the RF transmission time. Bloch simulation of 90° excitation by using a four channel coil array is performed to demonstrate its feasibility. Excitation performance of the sparse parallel transmission technique at different reduction factors of 1, 2, and 4 is evaluated. For comparison, parallel excitation using regular spiral trajectory is performed. The passband errors of the excitation profiles of each transmission are calculated for quantitative assessment of the proposed excitation method.
Sparse Neural Network Models of Antimicrobial Peptide-Activity Relationships.
Müller, Alex T; Kaymaz, Aral C; Gabernet, Gisela; Posselt, Gernot; Wessler, Silja; Hiss, Jan A; Schneider, Gisbert
2016-12-01
We present an adaptive neural network model for chemical data classification. The method uses an evolutionary algorithm for optimizing the network structure by seeking sparsely connected architectures. The number of hidden layers, the number of neurons in each layer and their connectivity are free variables of the system. We used the method for predicting antimicrobial peptide activity from the amino acid sequence. Visualization of the evolved sparse network structures suggested a high charge density and a low aggregation potential in solution as beneficial for antimicrobial activity. However, different training data sets and peptide representations resulted in greatly varying network structures. Overall, the sparse network models turned out to be less accurate than fully-connected networks. In a prospective application, we synthesized and tested 10 de novo generated peptides that were predicted to either possess antimicrobial activity, or to be inactive. Two of the predicted antibacterial peptides showed cosiderable bacteriostatic effects against both Staphylococcus aureus and Escherichia coli. None of the predicted inactive peptides possessed antibacterial properties. Molecular dynamics simulations of selected peptide structures in water and TFE suggest a pronounced peptide helicity in a hydrophobic environment. The results of this study underscore the applicability of neural networks for guiding the computer-assisted design of new peptides with desired properties.
ERIC Educational Resources Information Center
Kuntz, Aaron M.
2010-01-01
What can be known and how to render what we know are perpetual quandaries met by qualitative research, complicated further by the understanding that the everyday discourses influencing our representations are often tacit, unspoken or heard so often that they seem to warrant little reflection. In this article, I offer analytic memos as a means for…
The Real-Valued Sparse Direction of Arrival (DOA) Estimation Based on the Khatri-Rao Product.
Chen, Tao; Wu, Huanxin; Zhao, Zhongkai
2016-05-14
There is a problem that complex operation which leads to a heavy calculation burden is required when the direction of arrival (DOA) of a sparse signal is estimated by using the array covariance matrix. The solution of the multiple measurement vectors (MMV) model is difficult. In this paper, a real-valued sparse DOA estimation algorithm based on the Khatri-Rao (KR) product called the L₁-RVSKR is proposed. The proposed algorithm is based on the sparse representation of the array covariance matrix. The array covariance matrix is transformed to a real-valued matrix via a unitary transformation so that a real-valued sparse model is achieved. The real-valued sparse model is vectorized for transforming to a single measurement vector (SMV) model, and a new virtual overcomplete dictionary is constructed according to the KR product's property. Finally, the sparse DOA estimation is solved by utilizing the idea of a sparse representation of array covariance vectors (SRACV). The simulation results demonstrate the superior performance and the low computational complexity of the proposed algorithm.
Cellular-resolution population imaging reveals robust sparse coding in the Drosophila Mushroom Body
Honegger, Kyle S.; Campbell, Robert A. A.; Turner, Glenn C.
2011-01-01
Sensory stimuli are represented in the brain by the activity of populations of neurons. In most biological systems, studying population coding is challenging since only a tiny proportion of cells can be recorded simultaneously. Here we used 2-photon imaging to record neural activity in the relatively simple Drosophila mushroom body (MB), an area involved in olfactory learning and memory. Using the highly sensitive calcium indicator, GCaMP3, we simultaneously monitored the activity of >100 MB neurons in vivo (about 5% of the total population). The MB is thought to encode odors in sparse patterns of activity, but the code has yet to be explored either on a population level or with a wide variety of stimuli. We therefore imaged responses to odors chosen to evaluate the robustness of sparse representations. Different odors activated distinct patterns of MB neurons, however we found no evidence for spatial organization of neurons by either response probability or odor tuning within the cell body layer. The degree of sparseness was consistent across a wide range of stimuli, from monomolecular odors to artificial blends and even complex natural smells. Sparseness was mainly invariant across concentrations, largely because of the influence of recent odor experience. Finally, in contrast to sensory processing in other systems, no response features distinguished natural stimuli from monomolecular odors. Our results indicate that the fundamental feature of odor processing in the MB is to create sparse stimulus representations in a format that facilitates arbitrary associations between odor and punishment or reward. PMID:21849538
NASA Astrophysics Data System (ADS)
Zhang, Han; Chen, Xuefeng; Du, Zhaohui; Li, Xiang; Yan, Ruqiang
2016-04-01
Fault information of aero-engine bearings presents two particular phenomena, i.e., waveform distortion and impulsive feature frequency band dispersion, which leads to a challenging problem for current techniques of bearing fault diagnosis. Moreover, although many progresses of sparse representation theory have been made in feature extraction of fault information, the theory also confronts inevitable performance degradation due to the fact that relatively weak fault information has not sufficiently prominent and sparse representations. Therefore, a novel nonlocal sparse model (coined NLSM) and its algorithm framework has been proposed in this paper, which goes beyond simple sparsity by introducing more intrinsic structures of feature information. This work adequately exploits the underlying prior information that feature information exhibits nonlocal self-similarity through clustering similar signal fragments and stacking them together into groups. Within this framework, the prior information is transformed into a regularization term and a sparse optimization problem, which could be solved through block coordinate descent method (BCD), is formulated. Additionally, the adaptive structural clustering sparse dictionary learning technique, which utilizes k-Nearest-Neighbor (kNN) clustering and principal component analysis (PCA) learning, is adopted to further enable sufficient sparsity of feature information. Moreover, the selection rule of regularization parameter and computational complexity are described in detail. The performance of the proposed framework is evaluated through numerical experiment and its superiority with respect to the state-of-the-art method in the field is demonstrated through the vibration signals of experimental rig of aircraft engine bearings.
Sparseness of vowel category structure: Evidence from English dialect comparison
Scharinger, Mathias; Idsardi, William J.
2014-01-01
Current models of speech perception tend to emphasize either fine-grained acoustic properties or coarse-grained abstract characteristics of speech sounds. We argue for a particular kind of 'sparse' vowel representations and provide new evidence that these representations account for the successful access of the corresponding categories. In an auditory semantic priming experiment, American English listeners made lexical decisions on targets (e.g. load) preceded by semantically related primes (e.g. pack). Changes of the prime vowel that crossed a vowel-category boundary (e.g. peck) were not treated as a tolerable variation, as assessed by a lack of priming, although the phonetic categories of the two different vowels considerably overlap in American English. Compared to the outcome of the same experiment with New Zealand English listeners, where such prime variations were tolerated, our experiment supports the view that phonological representations are important in guiding the mapping process from the acoustic signal to an abstract mental representation. Our findings are discussed with regard to current models of speech perception and recent findings from brain imaging research. PMID:24653528
Sparse Coding for Alpha Matting.
Johnson, Jubin; Varnousfaderani, Ehsan Shahrian; Cholakkal, Hisham; Rajan, Deepu
2016-07-01
Existing color sampling-based alpha matting methods use the compositing equation to estimate alpha at a pixel from the pairs of foreground ( F ) and background ( B ) samples. The quality of the matte depends on the selected ( F,B ) pairs. In this paper, the matting problem is reinterpreted as a sparse coding of pixel features, wherein the sum of the codes gives the estimate of the alpha matte from a set of unpaired F and B samples. A non-parametric probabilistic segmentation provides a certainty measure on the pixel belonging to foreground or background, based on which a dictionary is formed for use in sparse coding. By removing the restriction to conform to ( F,B ) pairs, this method allows for better alpha estimation from multiple F and B samples. The same framework is extended to videos, where the requirement of temporal coherence is handled effectively. Here, the dictionary is formed by samples from multiple frames. A multi-frame graph model, as opposed to a single image as for image matting, is proposed that can be solved efficiently in closed form. Quantitative and qualitative evaluations on a benchmark dataset are provided to show that the proposed method outperforms the current stateoftheart in image and video matting.
Sparse Coding for Alpha Matting.
Johnson, Jubin; Varnousfaderani, Ehsan; Cholakkal, Hisham; Rajan, Deepu
2016-04-21
Existing color sampling based alpha matting methods use the compositing equation to estimate alpha at a pixel from pairs of foreground (F) and background (B) samples. The quality of the matte depends on the selected (F,B) pairs. In this paper, the matting problem is reinterpreted as a sparse coding of pixel features, wherein the sum of the codes gives the estimate of the alpha matte from a set of unpaired F and B samples. A non-parametric probabilistic segmentation provides a certainty measure on the pixel belonging to foreground or background, based on which a dictionary is formed for use in sparse coding. By removing the restriction to conform to (F,B) pairs, this method allows for better alpha estimation from multiple F and B samples. The same framework is extended to videos, where the requirement of temporal coherence is handled effectively. Here, the dictionary is formed by samples from multiple frames. A multi-frame graph model, as opposed to a single image as for image matting, is proposed that can be solved efficiently in closed form. Quantitative and qualitative evaluations on a benchmark dataset are provided to show that the proposed method outperforms current state-of-the-art in image and video matting.
Universal Priors for Sparse Modeling(PREPRINT)
2009-08-01
UNIVERSAL PRIORS FOR SPARSE MODELING By Ignacio Ramı́rez Federico Lecumberry and Guillermo Sapiro IMA Preprint Series # 2276 ( August 2009...8-98) Prescribed by ANSI Std Z39-18 Universal Priors for Sparse Modeling (Invited Paper) Ignacio Ramı́rez#1, Federico Lecumberry ∗2, Guillermo Sapiro...I. Ramirez, F. Lecumberry , and G. Sapiro. Sparse modeling with univer- sal priors and learned incoherent dictionaries. Submitted to NIPS, 2009. [22
Dimensionality reduction of hyperspectral images based on sparse discriminant manifold embedding
NASA Astrophysics Data System (ADS)
Huang, Hong; Luo, Fulin; Liu, Jiamin; Yang, Yaqiong
2015-08-01
Sparse manifold clustering and embedding (SMCE) adaptively selects neighbor points from the same manifold and approximately spans a low-dimensional affine subspace, but it does not explicitly give a projection matrix and encounters the out-of-sample problem. To overcome this drawback, we propose a new dimensionality reduction method, called sparse manifold embedding (SME), based on graph embedding and sparse representation for hyperspectral image (HSI). It utilizes the sparse coefficients of affine subspace to construct a similarity graph and preserves this sparse similarity in embedding space. Furthermore, we try to make full use of the prior label information to design a novel supervised learning method termed sparse discriminant manifold embedding (SDME). SDME not only inherits the merits of the sparsity property of affine subspace but also boosts the compactness of intra-manifold, which achieves discriminating features and further improves the classification performance of HSI. Experiments on two real hyperspectral data sets (Indian Pines and PaviaU) show the benefits of the proposed SME and SDME methods.
A novel sparse boosting method for crater detection in the high resolution planetary image
NASA Astrophysics Data System (ADS)
Wang, Yan; Yang, Gang; Guo, Lei
2015-09-01
Impact craters distributed on planetary surface become one of the main barriers during the soft landing of planetary probes. In order to accelerate the crater detection, in this paper, we present a new sparse boosting (SparseBoost) method for automatic detection of sub-kilometer craters. The SparseBoost method integrates an improved sparse kernel density estimator (RSDE-WL1) into the Boost algorithm and the RSDE-WL1 estimator is achieved by introducing weighted l1 penalty term into the reduced set density estimator. An iterative algorithm is proposed to implement the RSDE-WL1. The SparseBoost algorithm has the advantage of fewer selected features and simpler representation of the weak classifiers compared with the Boost algorithm. Our SparseBoost based crater detection method is evaluated on a large and high resolution image of Martian surface. Experimental results demonstrate that the proposed method can achieve less computational complexity in comparison with other crater detection methods in terms of selected features.
A Data Type for Efficient Representation of Other Data Types
NASA Technical Reports Server (NTRS)
James, Mark
2008-01-01
A self-organizing, monomorphic data type denoted a sequence has been conceived to address certain concerns that arise in programming parallel computers. A sequence in the present sense can be regarded abstractly as a vector, set, bag, queue, or other construct. Heretofore, in programming a parallel computer, it has been necessary for the programmer to state explicitly, at the outset, what parts of the program and the underlying data structures must be represented in parallel form. Not only is this requirement not optimal from the perspective of implementation; it entails an additional requirement that the programmer have intimate understanding of the underlying parallel structure. The present sequence data type overcomes both the implementation and parallel structure obstacles. In so doing, the sequence data type provides unified means by which the programmer can represent a data structure for natural and automatic decomposition to a parallel computing architecture. Sequences exhibit the behavioral and structural characteristics of vectors, but the underlying representations are automatically synthesized from combinations of programmers advice and execution use metrics. Sequences can vary bidirectionally between sparseness and density, making them excellent choices for many kinds of algorithms. The novelty and benefit of this behavior lies in the fact that it can relieve programmers of the details of implementations. The creation of a sequence enables decoupling of a conceptual representation from an implementation. The underlying representation of a sequence is a hybrid of representations composed of vectors, linked lists, connected blocks, and hash tables. The internal structure of a sequence can automatically change from time to time on the basis of how it is being used. Those portions of a sequence where elements have not been added or removed can be as efficient as vectors. As elements are inserted and removed in a given portion, then different methods are
Sparse and stable Markowitz portfolios.
Brodie, Joshua; Daubechies, Ingrid; De Mol, Christine; Giannone, Domenico; Loris, Ignace
2009-07-28
We consider the problem of portfolio selection within the classical Markowitz mean-variance framework, reformulated as a constrained least-squares regression problem. We propose to add to the objective function a penalty proportional to the sum of the absolute values of the portfolio weights. This penalty regularizes (stabilizes) the optimization problem, encourages sparse portfolios (i.e., portfolios with only few active positions), and allows accounting for transaction costs. Our approach recovers as special cases the no-short-positions portfolios, but does allow for short positions in limited number. We implement this methodology on two benchmark data sets constructed by Fama and French. Using only a modest amount of training data, we construct portfolios whose out-of-sample performance, as measured by Sharpe ratio, is consistently and significantly better than that of the naïve evenly weighted portfolio.
Otazo, Ricardo; Kim, Daniel; Axel, Leon; Sodickson, Daniel K.
2010-01-01
First-pass cardiac perfusion MRI is a natural candidate for compressed sensing acceleration since its representation in the combined temporal Fourier and spatial domain is sparse and the required incoherence can be effectively accomplished by k-t random undersampling. However, the required number of samples in practice (three to five times the number of sparse coefficients) limits the acceleration for compressed sensing alone. Parallel imaging may also be used to accelerate cardiac perfusion MRI, with acceleration factors ultimately limited by noise amplification. In this work, compressed sensing and parallel imaging are combined by merging the k-t SPARSE technique with SENSE reconstruction to substantially increase the acceleration rate for perfusion imaging. We also present a new theoretical framework for understanding the combination of k-t SPARSE with SENSE based on distributed compressed sensing theory. This framework, which identifies parallel imaging as a distributed multisensor implementation of compressed sensing, enables an estimate of feasible acceleration for the combined approach. We demonstrate feasibility of 8-fold acceleration in vivo with whole-heart coverage and high spatial and temporal resolution using standard coil arrays. The method is relatively insensitive to respiratory motion artifacts and presents similar temporal fidelity and image quality when compared to GRAPPA with 2-fold acceleration. PMID:20535813
Approximate Orthogonal Sparse Embedding for Dimensionality Reduction.
Lai, Zhihui; Wong, Wai Keung; Xu, Yong; Yang, Jian; Zhang, David
2016-04-01
Locally linear embedding (LLE) is one of the most well-known manifold learning methods. As the representative linear extension of LLE, orthogonal neighborhood preserving projection (ONPP) has attracted widespread attention in the field of dimensionality reduction. In this paper, a unified sparse learning framework is proposed by introducing the sparsity or L1-norm learning, which further extends the LLE-based methods to sparse cases. Theoretical connections between the ONPP and the proposed sparse linear embedding are discovered. The optimal sparse embeddings derived from the proposed framework can be computed by iterating the modified elastic net and singular value decomposition. We also show that the proposed model can be viewed as a general model for sparse linear and nonlinear (kernel) subspace learning. Based on this general model, sparse kernel embedding is also proposed for nonlinear sparse feature extraction. Extensive experiments on five databases demonstrate that the proposed sparse learning framework performs better than the existing subspace learning algorithm, particularly in the cases of small sample sizes.
Jemcov, A.; Matovic, M.D.
1996-12-31
This paper examines the sparse representation and preconditioning of a discrete Steklov-Poincare operator which arises in domain decomposition methods. A non-overlapping domain decomposition method is applied to a second order self-adjoint elliptic operator (Poisson equation), with homogeneous boundary conditions, as a model problem. It is shown that the discrete Steklov-Poincare operator allows sparse representation with a bounded condition number in wavelet basis if the transformation is followed by thresholding and resealing. These two steps combined enable the effective use of Krylov subspace methods as an iterative solution procedure for the system of linear equations. Finding the solution of an interface problem in domain decomposition methods, known as a Schur complement problem, has been shown to be equivalent to the discrete form of Steklov-Poincare operator. A common way to obtain Schur complement matrix is by ordering the matrix of discrete differential operator in subdomain node groups then block eliminating interface nodes. The result is a dense matrix which corresponds to the interface problem. This is equivalent to reducing the original problem to several smaller differential problems and one boundary integral equation problem for the subdomain interface.
Large-scale sparse singular value computations
NASA Technical Reports Server (NTRS)
Berry, Michael W.
1992-01-01
Four numerical methods for computing the singular value decomposition (SVD) of large sparse matrices on a multiprocessor architecture are presented. Lanczos and subspace iteration-based methods for determining several of the largest singular triplets (singular values and corresponding left and right-singular vectors) for sparse matrices arising from two practical applications: information retrieval and seismic reflection tomography are emphasized. The target architectures for implementations are the CRAY-2S/4-128 and Alliant FX/80. The sparse SVD problem is well motivated by recent information-retrieval techniques in which dominant singular values and their corresponding singular vectors of large sparse term-document matrices are desired, and by nonlinear inverse problems from seismic tomography applications which require approximate pseudo-inverses of large sparse Jacobian matrices.
Approximate inverse preconditioners for general sparse matrices
Chow, E.; Saad, Y.
1994-12-31
Preconditioned Krylov subspace methods are often very efficient in solving sparse linear matrices that arise from the discretization of elliptic partial differential equations. However, for general sparse indifinite matrices, the usual ILU preconditioners fail, often because of the fact that the resulting factors L and U give rise to unstable forward and backward sweeps. In such cases, alternative preconditioners based on approximate inverses may be attractive. We are currently developing a number of such preconditioners based on iterating on each column to get the approximate inverse. For this approach to be efficient, the iteration must be done in sparse mode, i.e., we must use sparse-matrix by sparse-vector type operatoins. We will discuss a few options and compare their performance on standard problems from the Harwell-Boeing collection.
A Hyperspherical Adaptive Sparse-Grid Method for High-Dimensional Discontinuity Detection
Zhang, Guannan; Webster, Clayton G.; Gunzburger, Max D.; Burkardt, John V.
2015-06-24
This study proposes and analyzes a hyperspherical adaptive hierarchical sparse-grid method for detecting jump discontinuities of functions in high-dimensional spaces. The method is motivated by the theoretical and computational inefficiencies of well-known adaptive sparse-grid methods for discontinuity detection. Our novel approach constructs a function representation of the discontinuity hypersurface of an N-dimensional discontinuous quantity of interest, by virtue of a hyperspherical transformation. Then, a sparse-grid approximation of the transformed function is built in the hyperspherical coordinate system, whose value at each point is estimated by solving a one-dimensional discontinuity detection problem. Due to the smoothness of the hypersurface, the new technique can identify jump discontinuities with significantly reduced computational cost, compared to existing methods. In addition, hierarchical acceleration techniques are also incorporated to further reduce the overall complexity. Rigorous complexity analyses of the new method are provided as are several numerical examples that illustrate the effectiveness of the approach.
Sparse-view ultrasound diffraction tomography using compressed sensing with nonuniform FFT.
Hua, Shaoyan; Ding, Mingyue; Yuchi, Ming
2014-01-01
Accurate reconstruction of the object from sparse-view sampling data is an appealing issue for ultrasound diffraction tomography (UDT). In this paper, we present a reconstruction method based on compressed sensing framework for sparse-view UDT. Due to the piecewise uniform characteristics of anatomy structures, the total variation is introduced into the cost function to find a more faithful sparse representation of the object. The inverse problem of UDT is iteratively resolved by conjugate gradient with nonuniform fast Fourier transform. Simulation results show the effectiveness of the proposed method that the main characteristics of the object can be properly presented with only 16 views. Compared to interpolation and multiband method, the proposed method can provide higher resolution and lower artifacts with the same view number. The robustness to noise and the computation complexity are also discussed.
Social Representations of High School Students about Mathematics Assessment
ERIC Educational Resources Information Center
Martínez-Sierra, Gustavo; Valle-Zequeida, María E.; Miranda-Tirado, Marisa; Dolores-Flores, Crisólogo
2016-01-01
The perceptions of students about assessment in mathematics classes have been sparsely investigated. In order to fill this gap, this qualitative study aims to identify the social "representations" (understood as the system of values, ideas, and practices about a social object) of high school students regarding "assessment in…
Sparse-aperture adaptive optics
NASA Astrophysics Data System (ADS)
Tuthill, Peter; Lloyd, James; Ireland, Michael; Martinache, Frantz; Monnier, John; Woodruff, Henry; ten Brummelaar, Theo; Turner, Nils; Townes, Charles
2006-06-01
Aperture masking interferometry and Adaptive Optics (AO) are two of the competing technologies attempting to recover diffraction-limited performance from ground-based telescopes. However, there are good arguments that these techniques should be viewed as complementary, not competitive. Masking has been shown to deliver superior PSF calibration, rejection of atmospheric noise and robust recovery of phase information through the use of closure phases. However, this comes at the penalty of loss of flux at the mask, restricting the technique to bright targets. Adaptive optics, on the other hand, can reach a fainter class of objects but suffers from the difficulty of calibration of the PSF which can vary with observational parameters such as seeing, airmass and source brightness. Here we present results from a fusion of these two techniques: placing an aperture mask downstream of an AO system. The precision characterization of the PSF enabled by sparse-aperture interferometry can now be applied to deconvolution of AO images, recovering structure from the traditionally-difficult regime within the core of the AO-corrected transfer function. Results of this program from the Palomar and Keck adaptive optical systems are presented.
The sparse matrix transform for covariance estimation and analysis of high dimensional signals.
Cao, Guangzhi; Bachega, Leonardo R; Bouman, Charles A
2011-03-01
Covariance estimation for high dimensional signals is a classically difficult problem in statistical signal analysis and machine learning. In this paper, we propose a maximum likelihood (ML) approach to covariance estimation, which employs a novel non-linear sparsity constraint. More specifically, the covariance is constrained to have an eigen decomposition which can be represented as a sparse matrix transform (SMT). The SMT is formed by a product of pairwise coordinate rotations known as Givens rotations. Using this framework, the covariance can be efficiently estimated using greedy optimization of the log-likelihood function, and the number of Givens rotations can be efficiently computed using a cross-validation procedure. The resulting estimator is generally positive definite and well-conditioned, even when the sample size is limited. Experiments on a combination of simulated data, standard hyperspectral data, and face image sets show that the SMT-based covariance estimates are consistently more accurate than both traditional shrinkage estimates and recently proposed graphical lasso estimates for a variety of different classes and sample sizes. An important property of the new covariance estimate is that it naturally yields a fast implementation of the estimated eigen-transformation using the SMT representation. In fact, the SMT can be viewed as a generalization of the classical fast Fourier transform (FFT) in that it uses "butterflies" to represent an orthonormal transform. However, unlike the FFT, the SMT can be used for fast eigen-signal analysis of general non-stationary signals.
Interpretable exemplar-based shape classification using constrained sparse linear models
NASA Astrophysics Data System (ADS)
Sigurdsson, Gunnar A.; Yang, Zhen; Tran, Trac D.; Prince, Jerry L.
2015-03-01
Many types of diseases manifest themselves as observable changes in the shape of the affected organs. Using shape classification, we can look for signs of disease and discover relationships between diseases. We formulate the problem of shape classification in a holistic framework that utilizes a lossless scalar field representation and a non-parametric classification based on sparse recovery. This framework generalizes over certain classes of unseen shapes while using the full information of the shape, bypassing feature extraction. The output of the method is the class whose combination of exemplars most closely approximates the shape, and furthermore, the algorithm returns the most similar exemplars along with their similarity to the shape, which makes the result simple to interpret. Our results show that the method offers accurate classification between three cerebellar diseases and controls in a database of cerebellar ataxia patients. For reproducible comparison, promising results are presented on publicly available 2D datasets, including the ETH-80 dataset where the method achieves 88.4% classification accuracy.
Separation of seismic blended data by sparse inversion over dictionary learning
NASA Astrophysics Data System (ADS)
Zhou, Yanhui; Chen, Wenchao; Gao, Jinghuai
2014-07-01
Recent development of blended acquisition calls for the new procedure to process blended seismic measurements. Presently, deblending and reconstructing unblended data followed by conventional processing is the most practical processing workflow. We study seismic deblending by advanced sparse inversion with a learned dictionary in this paper. To make our method more effective, hybrid acquisition and time-dithering sequential shooting are introduced so that clean single-shot records can be used to train the dictionary to favor the sparser representation of data to be recovered. Deblending and dictionary learning with l1-norm based sparsity are combined to construct the corresponding problem with respect to unknown recovery, dictionary, and coefficient sets. A two-step optimization approach is introduced. In the step of dictionary learning, the clean single-shot data are selected as trained data to learn the dictionary. For deblending, we fix the dictionary and employ an alternating scheme to update the recovery and coefficients separately. Synthetic and real field data were used to verify the performance of our method. The outcome can be a significant reference in designing high-efficient and low-cost blended acquisition.
Active dictionary learning for image representation
NASA Astrophysics Data System (ADS)
Wu, Tong; Sarwate, Anand D.; Bajwa, Waheed U.
2015-05-01
Sparse representations of images in overcomplete bases (i.e., redundant dictionaries) have many applications in computer vision and image processing. Recent works have demonstrated improvements in image representations by learning a dictionary from training data instead of using a predefined one. But learning a sparsifying dictionary can be computationally expensive in the case of a massive training set. This paper proposes a new approach, termed active screening, to overcome this challenge. Active screening sequentially selects subsets of training samples using a simple heuristic and adds the selected samples to a "learning pool," which is then used to learn a newer dictionary for improved representation performance. The performance of the proposed active dictionary learning approach is evaluated through numerical experiments on real-world image data; the results of these experiments demonstrate the effectiveness of the proposed method.
Sparse High Dimensional Models in Economics
Fan, Jianqing; Lv, Jinchi; Qi, Lei
2010-01-01
This paper reviews the literature on sparse high dimensional models and discusses some applications in economics and finance. Recent developments of theory, methods, and implementations in penalized least squares and penalized likelihood methods are highlighted. These variable selection methods are proved to be effective in high dimensional sparse modeling. The limits of dimensionality that regularization methods can handle, the role of penalty functions, and their statistical properties are detailed. Some recent advances in ultra-high dimensional sparse modeling are also briefly discussed. PMID:22022635
NASA Astrophysics Data System (ADS)
Lee, O. A.; Eicken, H.; Weyapuk, W., Jr.; Adams, B.; Mohoney, A. R.
2015-12-01
The significance of highly dispersed, remnant Arctic sea ice as a platform for marine mammals and indigenous hunters in spring and summer may have increased disproportionately with changes in the ice cover. As dispersed remnant ice becomes more common in the future it will be increasingly important to understand its ecological role for upper trophic levels such as marine mammals and its role for supporting primary productivity of ice-associated algae. Potential sparse ice habitat at sea ice concentrations below 15% is difficult to detect using remote sensing data alone. A combination of high resolution satellite imagery (including Synthetic Aperture Radar), data from the Barrow sea ice radar, and local observations from indigenous sea ice experts was used to detect sparse sea ice in the Alaska Arctic. Traditional knowledge on sea ice use by marine mammals was used to delimit the scales where sparse ice could still be used as habitat for seals and walrus. Potential sparse ice habitat was quantified with respect to overall spatial extent, size of ice floes, and density of floes. Sparse ice persistence offshore did not prevent the occurrence of large coastal walrus haul outs, but the lack of sparse ice and early sea ice retreat coincided with local observations of ringed seal pup mortality. Observations from indigenous hunters will continue to be an important source of information for validating remote sensing detections of sparse ice, and improving understanding of marine mammal adaptations to sea ice change.
Social biases determine spatiotemporal sparseness of ciliate mating heuristics
2012-01-01
Ciliates become highly social, even displaying animal-like qualities, in the joint presence of aroused conspecifics and nonself mating pheromones. Pheromone detection putatively helps trigger instinctual and learned courtship and dominance displays from which social judgments are made about the availability, compatibility, and fitness representativeness or likelihood of prospective mates and rivals. In earlier studies, I demonstrated the heterotrich Spirostomum ambiguum improves mating competence by effecting preconjugal strategies and inferences in mock social trials via behavioral heuristics built from Hebbian-like associative learning. Heuristics embody serial patterns of socially relevant action that evolve into ordered, topologically invariant computational networks supporting intra- and intermate selection. S. ambiguum employs heuristics to acquire, store, plan, compare, modify, select, and execute sets of mating propaganda. One major adaptive constraint over formation and use of heuristics involves a ciliate’s initial subjective bias, responsiveness, or preparedness, as defined by Stevens’ Law of subjective stimulus intensity, for perceiving the meaningfulness of mechanical pressures accompanying cell-cell contacts and additional perimating events. This bias controls durations and valences of nonassociative learning, search rates for appropriate mating strategies, potential net reproductive payoffs, levels of social honesty and deception, successful error diagnosis and correction of mating signals, use of insight or analysis to solve mating dilemmas, bioenergetics expenditures, and governance of mating decisions by classical or quantum statistical mechanics. I now report this same social bias also differentially affects the spatiotemporal sparseness, as measured with metric entropy, of ciliate heuristics. Sparseness plays an important role in neural systems through optimizing the specificity, efficiency, and capacity of memory representations. The
Takemura, Kazuhiro; Guo, Hao; Sakuraba, Shun; Matubayasi, Nobuyuki; Kitao, Akio
2012-12-07
We propose a method to evaluate binding free energy differences among distinct protein-protein complex model structures through all-atom molecular dynamics simulations in explicit water using the solution theory in the energy representation. Complex model structures are generated from a pair of monomeric structures using the rigid-body docking program ZDOCK. After structure refinement by side chain optimization and all-atom molecular dynamics simulations in explicit water, complex models are evaluated based on the sum of their conformational and solvation free energies, the latter calculated from the energy distribution functions obtained from relatively short molecular dynamics simulations of the complex in water and of pure water based on the solution theory in the energy representation. We examined protein-protein complex model structures of two protein-protein complex systems, bovine trypsin/CMTI-1 squash inhibitor (PDB ID: 1PPE) and RNase SA/barstar (PDB ID: 1AY7), for which both complex and monomer structures were determined experimentally. For each system, we calculated the energies for the crystal complex structure and twelve generated model structures including the model most similar to the crystal structure and very different from it. In both systems, the sum of the conformational and solvation free energies tended to be lower for the structure similar to the crystal. We concluded that our energy calculation method is useful for selecting low energy complex models similar to the crystal structure from among a set of generated models.
Sparse Superpixel Unmixing for Hyperspectral Image Analysis
NASA Technical Reports Server (NTRS)
Castano, Rebecca; Thompson, David R.; Gilmore, Martha
2010-01-01
Software was developed that automatically detects minerals that are present in each pixel of a hyperspectral image. An algorithm based on sparse spectral unmixing with Bayesian Positive Source Separation is used to produce mineral abundance maps from hyperspectral images. A superpixel segmentation strategy enables efficient unmixing in an interactive session. The algorithm computes statistically likely combinations of constituents based on a set of possible constituent minerals whose abundances are uncertain. A library of source spectra from laboratory experiments or previous remote observations is used. A superpixel segmentation strategy improves analysis time by orders of magnitude, permitting incorporation into an interactive user session (see figure). Mineralogical search strategies can be categorized as supervised or unsupervised. Supervised methods use a detection function, developed on previous data by hand or statistical techniques, to identify one or more specific target signals. Purely unsupervised results are not always physically meaningful, and may ignore subtle or localized mineralogy since they aim to minimize reconstruction error over the entire image. This algorithm offers advantages of both methods, providing meaningful physical interpretations and sensitivity to subtle or unexpected minerals.
Reconstructing cortical current density by exploring sparseness in the transform domain.
Ding, Lei
2009-05-07
In the present study, we have developed a novel electromagnetic source imaging approach to reconstruct extended cortical sources by means of cortical current density (CCD) modeling and a novel EEG imaging algorithm which explores sparseness in cortical source representations through the use of L1-norm in objective functions. The new sparse cortical current density (SCCD) imaging algorithm is unique since it reconstructs cortical sources by attaining sparseness in a transform domain (the variation map of cortical source distributions). While large variations are expected to occur along boundaries (sparseness) between active and inactive cortical regions, cortical sources can be reconstructed and their spatial extents can be estimated by locating these boundaries. We studied the SCCD algorithm using numerous simulations to investigate its capability in reconstructing cortical sources with different extents and in reconstructing multiple cortical sources with different extent contrasts. The SCCD algorithm was compared with two L2-norm solutions, i.e. weighted minimum norm estimate (wMNE) and cortical LORETA. Our simulation data from the comparison study show that the proposed sparse source imaging algorithm is able to accurately and efficiently recover extended cortical sources and is promising to provide high-accuracy estimation of cortical source extents.
Time-frequency signature sparse reconstruction using chirp dictionary
NASA Astrophysics Data System (ADS)
Nguyen, Yen T. H.; Amin, Moeness G.; Ghogho, Mounir; McLernon, Des
2015-05-01
This paper considers local sparse reconstruction of time-frequency signatures of windowed non-stationary radar returns. These signals can be considered instantaneously narrow-band, thus the local time-frequency behavior can be recovered accurately with incomplete observations. The typically employed sinusoidal dictionary induces competing requirements on window length. It confronts converse requests on the number of measurements for exact recovery, and sparsity. In this paper, we use chirp dictionary for each window position to determine the signal instantaneous frequency laws. This approach can considerably mitigate the problems of sinusoidal dictionary, and enable the utilization of longer windows for accurate time-frequency representations. It also reduces the picket fence by introducing a new factor, the chirp rate α. Simulation examples are provided, demonstrating the superior performance of local chirp dictionary over its sinusoidal counterpart.
Adaptive dictionary learning in sparse gradient domain for image recovery.
Liu, Qiegen; Wang, Shanshan; Ying, Leslie; Peng, Xi; Zhu, Yanjie; Liang, Dong
2013-12-01
Image recovery from undersampled data has always been challenging due to its implicit ill-posed nature but becomes fascinating with the emerging compressed sensing (CS) theory. This paper proposes a novel gradient based dictionary learning method for image recovery, which effectively integrates the popular total variation (TV) and dictionary learning technique into the same framework. Specifically, we first train dictionaries from the horizontal and vertical gradients of the image and then reconstruct the desired image using the sparse representations of both derivatives. The proposed method enables local features in the gradient images to be captured effectively, and can be viewed as an adaptive extension of the TV regularization. The results of various experiments on MR images consistently demonstrate that the proposed algorithm efficiently recovers images and presents advantages over the current leading CS reconstruction approaches.
Sparse Downscaling and Adaptive Fusion of Multi-sensor Precipitation
NASA Astrophysics Data System (ADS)
Ebtehaj, M.; Foufoula, E.
2011-12-01
The past decades have witnessed a remarkable emergence of new sources of multiscale multi-sensor precipitation data including data from global spaceborne active and passive sensors, regional ground based weather surveillance radars and local rain-gauges. Resolution enhancement of remotely sensed rainfall and optimal integration of multi-sensor data promise a posteriori estimates of precipitation fluxes with increased accuracy and resolution to be used in hydro-meteorological applications. In this context, new frameworks are proposed for resolution enhancement and multiscale multi-sensor precipitation data fusion, which capitalize on two main observations: (1) sparseness of remotely sensed precipitation fields in appropriately chosen transformed domains, (e.g., in wavelet space) which promotes the use of the newly emerged theory of sparse representation and compressive sensing for resolution enhancement; (2) a conditionally Gaussian Scale Mixture (GSM) parameterization in the wavelet domain which allows exploiting the efficient linear estimation methodologies, while capturing the non-Gaussian data structure of rainfall. The proposed methodologies are demonstrated using a data set of coincidental observations of precipitation reflectivity images by the spaceborne precipitation radar (PR) aboard the Tropical Rainfall Measurement Mission (TRMM) satellite and ground-based NEXRAD weather surveillance Doppler radars. Uniqueness and stability of the solution, capturing non-Gaussian singular structure of rainfall, reduced uncertainty of estimation and efficiency of computation are the main advantages of the proposed methodologies over the commonly used standard Gaussian techniques.
Brain at work: time, sparseness and superposition principles.
Molotchnikoff, Stephane; Rouat, Jean
2012-01-01
Many studies explored mechanisms through which the brain encodes sensory inputs allowing a coherent behavior. The brain could identify stimuli via a hierarchical stream of activity leading to a cardinal neuron responsive to one particular object. The opportunity to record from numerous neurons offered investigators the capability of examining simultaneously the functioning of many cells. These approaches suggested encoding processes that are parallel rather than serial. Binding the many features of a stimulus may be accomplished through an induced synchronization of cell's action potentials. These interpretations are supported by experimental data and offer many advantages but also several shortcomings. We argue for a coding mechanism based on a sparse synchronization paradigm. We show that synchronization of spikes is a fast and efficient mode to encode the representation of objects based on feature bindings. We introduce the view that sparse synchronization coding presents an interesting venue in probing brain encoding mechanisms as it allows the functional establishment of multi-layered and time-conditioned neuronal networks or multislice networks. We propose a model based on integrate-and-fire spiking neurons.
Robust feature point matching with sparse model.
Jiang, Bo; Tang, Jin; Luo, Bin; Lin, Liang
2014-12-01
Feature point matching that incorporates pairwise constraints can be cast as an integer quadratic programming (IQP) problem. Since it is NP-hard, approximate methods are required. The optimal solution for IQP matching problem is discrete, binary, and thus sparse in nature. This motivates us to use sparse model for feature point matching problem. The main advantage of the proposed sparse feature point matching (SPM) method is that it generates sparse solution and thus naturally imposes the discrete mapping constraints approximately in the optimization process. Therefore, it can optimize the IQP matching problem in an approximate discrete domain. In addition, an efficient algorithm can be derived to solve SPM problem. Promising experimental results on both synthetic points sets matching and real-world image feature sets matching tasks show the effectiveness of the proposed feature point matching method.
NASA Astrophysics Data System (ADS)
Xie, Qing; Cheng, Shuyi; Lü, Fangcheng; Li, Yanqing
2014-03-01
The acoustic detecting performance of a partial discharge (PD) ultrasonic sensor array can be improved by increasing the number of array elements. However, it will increase the complexity and cost of the PD detection system. Therefore, a sparse sensor with an optimization design can be chosen to ensure good acoustic performance. In this paper, first, a quantitative method is proposed for evaluating the acoustic performance of a square PD ultrasonic array sensor. Second, a method of sparse design is presented to combine the evaluation method with the chaotic monkey algorithm. Third, an optimal sparse structure of a 3 × 3 square PD ultrasonic array sensor is deduced. It is found that, under different sparseness and sparse structure, the main beam width of the directivity function shows a small variation, while the sidelobe amplitude shows a bigger variation. For a specific sparseness, the acoustic performance under the optimal sparse structure is close to that using a full array. Finally, some simulations based on the above method show that, for certain sparseness, the sensor with the optimal sparse structure exhibits superior positioning accuracy compared to that with a stochastic one. The sensor array structure may be chosen according to the actual requirements for an actual engineering application.
Sparse CSEM inversion driven by seismic coherence
NASA Astrophysics Data System (ADS)
Guo, Zhenwei; Dong, Hefeng; Kristensen, Åge
2016-12-01
Marine controlled source electromagnetic (CSEM) data inversion for hydrocarbon exploration is often challenging due to high computational cost, physical memory requirement and low resolution of the obtained resistivity map. This paper aims to enhance both the speed and resolution of CSEM inversion by introducing structural geological information in the inversion algorithm. A coarse mesh is generated for Occam’s inversion, where the parameters are fewer than in the fine regular mesh. This sparse mesh is defined as a coherence-based irregular (IC) sparse mesh, which is based on vertices extracted from available geological information. Inversion results on synthetic data illustrate that the IC sparse mesh has a smaller inversion computational cost compared to the regular dense (RD) mesh. It also has a higher resolution than with a regular sparse (RS) mesh for the same number of estimated parameters. In order to study how the IC sparse mesh reduces the computational time, four different meshes are generated for Occam’s inversion. As a result, an IC sparse mesh can reduce the computational cost while it keeps the resolution as good as a fine regular mesh. The IC sparse mesh reduces the computational cost of the matrix operation for model updates. When the number of estimated parameters reduces to a limited value, the computational cost is independent of the number of parameters. For a testing model with two resistive layers, the inversion result using an IC sparse mesh has higher resolution in both horizontal and vertical directions. Overall, the model representing significant geological information in the IC mesh can improve the resolution of the resistivity models obtained from inversion of CSEM data.
Sparse Extreme Learning Machine for Classification
Bai, Zuo; Huang, Guang-Bin; Wang, Danwei; Wang, Han; Westover, M. Brandon
2016-01-01
Extreme learning machine (ELM) was initially proposed for single-hidden-layer feedforward neural networks (SLFNs). In the hidden layer (feature mapping), nodes are randomly generated independently of training data. Furthermore, a unified ELM was proposed, providing a single framework to simplify and unify different learning methods, such as SLFNs, least square support vector machines, proximal support vector machines, and so on. However, the solution of unified ELM is dense, and thus, usually plenty of storage space and testing time are required for large-scale applications. In this paper, a sparse ELM is proposed as an alternative solution for classification, reducing storage space and testing time. In addition, unified ELM obtains the solution by matrix inversion, whose computational complexity is between quadratic and cubic with respect to the training size. It still requires plenty of training time for large-scale problems, even though it is much faster than many other traditional methods. In this paper, an efficient training algorithm is specifically developed for sparse ELM. The quadratic programming problem involved in sparse ELM is divided into a series of smallest possible sub-problems, each of which are solved analytically. Compared with SVM, sparse ELM obtains better generalization performance with much faster training speed. Compared with unified ELM, sparse ELM achieves similar generalization performance for binary classification applications, and when dealing with large-scale binary classification problems, sparse ELM realizes even faster training speed than unified ELM. PMID:25222727
Finding Nonoverlapping Substructures of a Sparse Matrix
Pinar, Ali; Vassilevska, Virginia
2005-08-11
Many applications of scientific computing rely on computations on sparse matrices. The design of efficient implementations of sparse matrix kernels is crucial for the overall efficiency of these applications. Due to the high compute-to-memory ratio and irregular memory access patterns, the performance of sparse matrix kernels is often far away from the peak performance on a modern processor. Alternative data structures have been proposed, which split the original matrix A into A{sub d} and A{sub s}, so that A{sub d} contains all dense blocks of a specified size in the matrix, and A{sub s} contains the remaining entries. This enables the use of dense matrix kernels on the entries of A{sub d} producing better memory performance. In this work, we study the problem of finding a maximum number of nonoverlapping dense blocks in a sparse matrix, which is previously not studied in the sparse matrix community. We show that the maximum nonoverlapping dense blocks problem is NP-complete by using a reduction from the maximum independent set problem on cubic planar graphs. We also propose a 2/3-approximation algorithm that runs in linear time in the number of nonzeros in the matrix. This extended abstract focuses on our results for 2x2 dense blocks. However we show that our results can be generalized to arbitrary sized dense blocks, and many other oriented substructures, which can be exploited to improve the memory performance of sparse matrix operations.
Sparse extreme learning machine for classification.
Bai, Zuo; Huang, Guang-Bin; Wang, Danwei; Wang, Han; Westover, M Brandon
2014-10-01
Extreme learning machine (ELM) was initially proposed for single-hidden-layer feedforward neural networks (SLFNs). In the hidden layer (feature mapping), nodes are randomly generated independently of training data. Furthermore, a unified ELM was proposed, providing a single framework to simplify and unify different learning methods, such as SLFNs, least square support vector machines, proximal support vector machines, and so on. However, the solution of unified ELM is dense, and thus, usually plenty of storage space and testing time are required for large-scale applications. In this paper, a sparse ELM is proposed as an alternative solution for classification, reducing storage space and testing time. In addition, unified ELM obtains the solution by matrix inversion, whose computational complexity is between quadratic and cubic with respect to the training size. It still requires plenty of training time for large-scale problems, even though it is much faster than many other traditional methods. In this paper, an efficient training algorithm is specifically developed for sparse ELM. The quadratic programming problem involved in sparse ELM is divided into a series of smallest possible sub-problems, each of which are solved analytically. Compared with SVM, sparse ELM obtains better generalization performance with much faster training speed. Compared with unified ELM, sparse ELM achieves similar generalization performance for binary classification applications, and when dealing with large-scale binary classification problems, sparse ELM realizes even faster training speed than unified ELM.
Haider, Bilal; Krause, Matthew R.; Duque, Alvaro; Yu, Yuguo; Touryan, Jonathan; Mazer, James A.; McCormick, David A.
2011-01-01
SUMMARY During natural vision, the entire visual field is stimulated by images rich in spatiotemporal structure. Although many visual system studies restrict stimuli to the classical receptive field (CRF), it is known that costimulation of the CRF and the surrounding nonclassical receptive field (nCRF) increases neuronal response sparseness. The cellular and network mechanisms underlying increased response sparseness remain largely unexplored. Here we show that combined CRF + nCRF stimulation increases the sparseness, reliability, and precision of spiking and membrane potential responses in classical regular spiking (RSC) pyramidal neurons of cat primary visual cortex. Conversely, fast-spiking interneurons exhibit increased activity and decreased selectivity during CRF + nCRF stimulation. The increased sparseness and reliability of RSC neuron spiking is associated with increased inhibitory barrages and narrower visually evoked synaptic potentials. Our experimental observations were replicated with a simple computational model, suggesting that network interactions among neuronal subtypes ultimately sharpen recurrent excitation, producing specific and reliable visual responses. PMID:20152117
Sparse partial least squares regression for simultaneous dimension reduction and variable selection.
Chun, Hyonho; Keleş, Sündüz
2010-01-01
Partial least squares regression has been an alternative to ordinary least squares for handling multicollinearity in several areas of scientific research since the 1960s. It has recently gained much attention in the analysis of high dimensional genomic data. We show that known asymptotic consistency of the partial least squares estimator for a univariate response does not hold with the very large p and small n paradigm. We derive a similar result for a multivariate response regression with partial least squares. We then propose a sparse partial least squares formulation which aims simultaneously to achieve good predictive performance and variable selection by producing sparse linear combinations of the original predictors. We provide an efficient implementation of sparse partial least squares regression and compare it with well-known variable selection and dimension reduction approaches via simulation experiments. We illustrate the practical utility of sparse partial least squares regression in a joint analysis of gene expression and genomewide binding data.
Efficient, sparse biological network determination
August, Elias; Papachristodoulou, Antonis
2009-01-01
Background Determining the interaction topology of biological systems is a topic that currently attracts significant research interest. Typical models for such systems take the form of differential equations that involve polynomial and rational functions. Such nonlinear models make the problem of determining the connectivity of biochemical networks from time-series experimental data much harder. The use of linear dynamics and linearization techniques that have been proposed in the past can circumvent this, but the general problem of developing efficient algorithms for models that provide more accurate system descriptions remains open. Results We present a network determination algorithm that can treat model descriptions with polynomial and rational functions and which does not make use of linearization. For this purpose, we make use of the observation that biochemical networks are in general 'sparse' and minimize the 1-norm of the decision variables (sum of weighted network connections) while constraints keep the error between data and the network dynamics small. The emphasis of our methodology is on determining the interconnection topology rather than the specific reaction constants and it takes into account the necessary properties that a chemical reaction network should have – something that techniques based on linearization can not. The problem can be formulated as a Linear Program, a convex optimization problem, for which efficient algorithms are available that can treat large data sets efficiently and uncertainties in data or model parameters. Conclusion The presented methodology is able to predict with accuracy and efficiency the connectivity structure of a chemical reaction network with mass action kinetics and of a gene regulatory network from simulation data even if the dynamics of these systems are non-polynomial (rational) and uncertainties in the data are taken into account. It also produces a network structure that can explain the real experimental
Sajjad, Muhammad; Mehmood, Irfan; Baik, Sung Wook
2015-01-01
Image super-resolution (SR) plays a vital role in medical imaging that allows a more efficient and effective diagnosis process. Usually, diagnosing is difficult and inaccurate from low-resolution (LR) and noisy images. Resolution enhancement through conventional interpolation methods strongly affects the precision of consequent processing steps, such as segmentation and registration. Therefore, we propose an efficient sparse coded image SR reconstruction technique using a trained dictionary. We apply a simple and efficient regularized version of orthogonal matching pursuit (ROMP) to seek the coefficients of sparse representation. ROMP has the transparency and greediness of OMP and the robustness of the L1-minization that enhance the dictionary learning process to capture feature descriptors such as oriented edges and contours from complex images like brain MRIs. The sparse coding part of the K-SVD dictionary training procedure is modified by substituting OMP with ROMP. The dictionary update stage allows simultaneously updating an arbitrary number of atoms and vectors of sparse coefficients. In SR reconstruction, ROMP is used to determine the vector of sparse coefficients for the underlying patch. The recovered representations are then applied to the trained dictionary, and finally, an optimization leads to high-resolution output of high-quality. Experimental results demonstrate that the super-resolution reconstruction quality of the proposed scheme is comparatively better than other state-of-the-art schemes.
Finding nonoverlapping substructures of a sparse matrix
Pinar, Ali; Vassilevska, Virginia
2004-08-09
Many applications of scientific computing rely on computations on sparse matrices, thus the design of efficient implementations of sparse matrix kernels is crucial for the overall efficiency of these applications. Due to the high compute-to-memory ratio and irregular memory access patterns, the performance of sparse matrix kernels is often far away from the peak performance on a modern processor. Alternative data structures have been proposed, which split the original matrix A into A{sub d} and A{sub s}, so that A{sub d} contains all dense blocks of a specified size in the matrix, and A{sub s} contains the remaining entries. This enables the use of dense matrix kernels on the entries of A{sub d} producing better memory performance. In this work, we study the problem of finding a maximum number of non overlapping rectangular dense blocks in a sparse matrix, which has not been studied in the sparse matrix community. We show that the maximum non overlapping dense blocks problem is NP-complete by using a reduction from the maximum independent set problem on cubic planar graphs. We also propose a 2/3-approximation algorithm for 2 times 2 blocks that runs in linear time in the number of nonzeros in the matrix. We discuss alternatives to rectangular blocks such as diagonal blocks and cross blocks and present complexity analysis and approximation algorithms.
Integrative analysis of multiple diverse omics datasets by sparse group multitask regression
Lin, Dongdong; Zhang, Jigang; Li, Jingyao; He, Hao; Deng, Hong-Wen; Wang, Yu-Ping
2014-01-01
A variety of high throughput genome-wide assays enable the exploration of genetic risk factors underlying complex traits. Although these studies have remarkable impact on identifying susceptible biomarkers, they suffer from issues such as limited sample size and low reproducibility. Combining individual studies of different genetic levels/platforms has the promise to improve the power and consistency of biomarker identification. In this paper, we propose a novel integrative method, namely sparse group multitask regression, for integrating diverse omics datasets, platforms, and populations to identify risk genes/factors of complex diseases. This method combines multitask learning with sparse group regularization, which will: (1) treat the biomarker identification in each single study as a task and then combine them by multitask learning; (2) group variables from all studies for identifying significant genes; (3) enforce sparse constraint on groups of variables to overcome the “small sample, but large variables” problem. We introduce two sparse group penalties: sparse group lasso and sparse group ridge in our multitask model, and provide an effective algorithm for each model. In addition, we propose a significance test for the identification of potential risk genes. Two simulation studies are performed to evaluate the performance of our integrative method by comparing it with conventional meta-analysis method. The results show that our sparse group multitask method outperforms meta-analysis method significantly. In an application to our osteoporosis studies, 7 genes are identified as significant genes by our method and are found to have significant effects in other three independent studies for validation. The most significant gene SOD2 has been identified in our previous osteoporosis study involving the same expression dataset. Several other genes such as TREML2, HTR1E, and GLO1 are shown to be novel susceptible genes for osteoporosis, as confirmed from other
Semi-implicit Integration Factor Methods on Sparse Grids for High-Dimensional Systems
Wang, Dongyong; Chen, Weitao; Nie, Qing
2015-01-01
Numerical methods for partial differential equations in high-dimensional spaces are often limited by the curse of dimensionality. Though the sparse grid technique, based on a one-dimensional hierarchical basis through tensor products, is popular for handling challenges such as those associated with spatial discretization, the stability conditions on time step size due to temporal discretization, such as those associated with high-order derivatives in space and stiff reactions, remain. Here, we incorporate the sparse grids with the implicit integration factor method (IIF) that is advantageous in terms of stability conditions for systems containing stiff reactions and diffusions. We combine IIF, in which the reaction is treated implicitly and the diffusion is treated explicitly and exactly, with various sparse grid techniques based on the finite element and finite difference methods and a multi-level combination approach. The overall method is found to be efficient in terms of both storage and computational time for solving a wide range of PDEs in high dimensions. In particular, the IIF with the sparse grid combination technique is flexible and effective in solving systems that may include cross-derivatives and non-constant diffusion coefficients. Extensive numerical simulations in both linear and nonlinear systems in high dimensions, along with applications of diffusive logistic equations and Fokker-Planck equations, demonstrate the accuracy, efficiency, and robustness of the new methods, indicating potential broad applications of the sparse grid-based integration factor method. PMID:25897178
NASA Astrophysics Data System (ADS)
Kunlun, Qi; Xiaochun, Zhang; Baiyan, Wu; Huayi, Wu
2016-10-01
High-resolution remote-sensing images are increasingly applied in land-use classification problems. Land-use scenes are often very complex and difficult to represent. Subsequently, the recognition of multiple land-cover classes is a continuing research question. We propose a classification framework based on a sparse coding-based correlaton (termed sparse correlaton) model to solve this challenge. Specifically, a general mapping strategy is presented to label visual words and generate sparse coding-based correlograms, which can exploit the spatial co-occurrences of visual words. A compact spatial representation without loss discrimination is achieved through adaptive vector quantization of correlogram in land-use scene classification. Moreover, instead of using K-means for visual word encoding in the original correlaton model, our proposed sparse correlaton model uses sparse coding to achieve lower reconstruction error. Experiments on a public ground truth image dataset of 21 land-use classes demonstrate that our sparse coding-based correlaton method can improve the performance of land-use scene classification and outperform many existing bag-of-visual-words-based methods.
Fast wavelet based sparse approximate inverse preconditioner
Wan, W.L.
1996-12-31
Incomplete LU factorization is a robust preconditioner for both general and PDE problems but unfortunately not easy to parallelize. Recent study of Huckle and Grote and Chow and Saad showed that sparse approximate inverse could be a potential alternative while readily parallelizable. However, for special class of matrix A that comes from elliptic PDE problems, their preconditioners are not optimal in the sense that independent of mesh size. A reason may be that no good sparse approximate inverse exists for the dense inverse matrix. Our observation is that for this kind of matrices, its inverse entries typically have piecewise smooth changes. We can take advantage of this fact and use wavelet compression techniques to construct a better sparse approximate inverse preconditioner. We shall show numerically that our approach is effective for this kind of matrices.
Sparsey™: event recognition via deep hierarchical sparse distributed codes
Rinkus, Gerard J.
2014-01-01
The visual cortex's hierarchical, multi-level organization is captured in many biologically inspired computational vision models, the general idea being that progressively larger scale (spatially/temporally) and more complex visual features are represented in progressively higher areas. However, most earlier models use localist representations (codes) in each representational field (which we equate with the cortical macrocolumn, “mac”), at each level. In localism, each represented feature/concept/event (hereinafter “item”) is coded by a single unit. The model we describe, Sparsey, is hierarchical as well but crucially, it uses sparse distributed coding (SDC) in every mac in all levels. In SDC, each represented item is coded by a small subset of the mac's units. The SDCs of different items can overlap and the size of overlap between items can be used to represent their similarity. The difference between localism and SDC is crucial because SDC allows the two essential operations of associative memory, storing a new item and retrieving the best-matching stored item, to be done in fixed time for the life of the model. Since the model's core algorithm, which does both storage and retrieval (inference), makes a single pass over all macs on each time step, the overall model's storage/retrieval operation is also fixed-time, a criterion we consider essential for scalability to the huge (“Big Data”) problems. A 2010 paper described a nonhierarchical version of this model in the context of purely spatial pattern processing. Here, we elaborate a fully hierarchical model (arbitrary numbers of levels and macs per level), describing novel model principles like progressive critical periods, dynamic modulation of principal cells' activation functions based on a mac-level familiarity measure, representation of multiple simultaneously active hypotheses, a novel method of time warp invariant recognition, and we report results showing learning/recognition of spatiotemporal
Sparsey™: event recognition via deep hierarchical sparse distributed codes.
Rinkus, Gerard J
2014-01-01
The visual cortex's hierarchical, multi-level organization is captured in many biologically inspired computational vision models, the general idea being that progressively larger scale (spatially/temporally) and more complex visual features are represented in progressively higher areas. However, most earlier models use localist representations (codes) in each representational field (which we equate with the cortical macrocolumn, "mac"), at each level. In localism, each represented feature/concept/event (hereinafter "item") is coded by a single unit. The model we describe, Sparsey, is hierarchical as well but crucially, it uses sparse distributed coding (SDC) in every mac in all levels. In SDC, each represented item is coded by a small subset of the mac's units. The SDCs of different items can overlap and the size of overlap between items can be used to represent their similarity. The difference between localism and SDC is crucial because SDC allows the two essential operations of associative memory, storing a new item and retrieving the best-matching stored item, to be done in fixed time for the life of the model. Since the model's core algorithm, which does both storage and retrieval (inference), makes a single pass over all macs on each time step, the overall model's storage/retrieval operation is also fixed-time, a criterion we consider essential for scalability to the huge ("Big Data") problems. A 2010 paper described a nonhierarchical version of this model in the context of purely spatial pattern processing. Here, we elaborate a fully hierarchical model (arbitrary numbers of levels and macs per level), describing novel model principles like progressive critical periods, dynamic modulation of principal cells' activation functions based on a mac-level familiarity measure, representation of multiple simultaneously active hypotheses, a novel method of time warp invariant recognition, and we report results showing learning/recognition of spatiotemporal patterns.
Protein family classification using sparse Markov transducers.
Eskin, E; Grundy, W N; Singer, Y
2000-01-01
In this paper we present a method for classifying proteins into families using sparse Markov transducers (SMTs). Sparse Markov transducers, similar to probabilistic suffix trees, estimate a probability distribution conditioned on an input sequence. SMTs generalize probabilistic suffix trees by allowing for wild-cards in the conditioning sequences. Because substitutions of amino acids are common in protein families, incorporating wildcards into the model significantly improves classification performance. We present two models for building protein family classifiers using SMTs. We also present efficient data structures to improve the memory usage of the models. We evaluate SMTs by building protein family classifiers using the Pfam database and compare our results to previously published results.
Tensor methods for large, sparse unconstrained optimization
Bouaricha, A.
1996-11-01
Tensor methods for unconstrained optimization were first introduced by Schnabel and Chow [SIAM J. Optimization, 1 (1991), pp. 293-315], who describe these methods for small to moderate size problems. This paper extends these methods to large, sparse unconstrained optimization problems. This requires an entirely new way of solving the tensor model that makes the methods suitable for solving large, sparse optimization problems efficiently. We present test results for sets of problems where the Hessian at the minimizer is nonsingular and where it is singular. These results show that tensor methods are significantly more efficient and more reliable than standard methods based on Newton`s method.
Dense encoding of natural odorants by ensembles of sparsely activated neurons in the olfactory bulb
Gschwend, Olivier; Beroud, Jonathan; Vincis, Roberto; Rodriguez, Ivan; Carleton, Alan
2016-01-01
Sensory information undergoes substantial transformation along sensory pathways, usually encompassing sparsening of activity. In the olfactory bulb, though natural odorants evoke dense glomerular input maps, mitral and tufted (M/T) cells tuning is considered to be sparse because of highly odor-specific firing rate change. However, experiments used to draw this conclusion were either based on recordings performed in anesthetized preparations or used monomolecular odorants presented at arbitrary concentrations. In this study, we evaluated the lifetime and population sparseness evoked by natural odorants by capturing spike temporal patterning of neuronal assemblies instead of individual M/T tonic activity. Using functional imaging and tetrode recordings in awake mice, we show that natural odorants at their native concentrations are encoded by broad assemblies of M/T cells. While reducing odorant concentrations, we observed a reduced number of activated glomeruli representations and consequently a narrowing of M/T tuning curves. We conclude that natural odorants at their native concentrations recruit M/T cells with phasic rather than tonic activity. When encoding odorants in assemblies, M/T cells carry information about a vast number of odorants (lifetime sparseness). In addition, each natural odorant activates a broad M/T cell assembly (population sparseness). PMID:27824096
Feature Modeling in Underwater Environments Using Sparse Linear Combinations
2010-01-01
waters . Optics Express, 16(13), 2008. 2, 3 [9] J. Jaflfe. Monte carlo modeling of underwate-image forma- tion: Validity of the linear and small-angle... turbid water , etc), we would like to determine if these two images contain the same (or similar) object(s). One approach is as follows: 1. Detect...nearest neighbor methods on extracted feature descriptors This methodology works well for clean, out-of- water images, however, when imaging underwater
VIM-based dynamic sparse grid approach to partial differential equations.
Mei, Shu-Li
2014-01-01
Combining the variational iteration method (VIM) with the sparse grid theory, a dynamic sparse grid approach for nonlinear PDEs is proposed in this paper. In this method, a multilevel interpolation operator is constructed based on the sparse grids theory firstly. The operator is based on the linear combination of the basic functions and independent of them. Second, by means of the precise integration method (PIM), the VIM is developed to solve the nonlinear system of ODEs which is obtained from the discretization of the PDEs. In addition, a dynamic choice scheme on both of the inner and external grid points is proposed. It is different from the traditional interval wavelet collocation method in which the choice of both of the inner and external grid points is dynamic. The numerical experiments show that our method is better than the traditional wavelet collocation method, especially in solving the PDEs with the Nuemann boundary conditions.
Second SIAM conference on sparse matrices: Abstracts. Final technical report
1996-12-31
This report contains abstracts on the following topics: invited and long presentations (IP1 & LP1); sparse matrix reordering & graph theory I; sparse matrix tools & environments I; eigenvalue computations I; iterative methods & acceleration techniques I; applications I; parallel algorithms I; sparse matrix reordering & graphy theory II; sparse matrix tool & environments II; least squares & optimization I; iterative methods & acceleration techniques II; applications II; eigenvalue computations II; least squares & optimization II; parallel algorithms II; sparse direct methods; iterative methods & acceleration techniques III; eigenvalue computations III; and sparse matrix reordering & graph theory III.
Sparse imaging of cortical electrical current densities via wavelet transforms
NASA Astrophysics Data System (ADS)
Liao, Ke; Zhu, Min; Ding, Lei; Valette, Sébastien; Zhang, Wenbo; Dickens, Deanna
2012-11-01
While the cerebral cortex in the human brain is of functional importance, functions defined on this structure are difficult to analyze spatially due to its highly convoluted irregular geometry. This study developed a novel L1-norm regularization method using a newly proposed multi-resolution face-based wavelet method to estimate cortical electrical activities in electroencephalography (EEG) and magnetoencephalography (MEG) inverse problems. The proposed wavelets were developed based on multi-resolution models built from irregular cortical surface meshes, which were realized in this study too. The multi-resolution wavelet analysis was used to seek sparse representation of cortical current densities in transformed domains, which was expected due to the compressibility of wavelets, and evaluated using Monte Carlo simulations. The EEG/MEG inverse problems were solved with the use of the novel L1-norm regularization method exploring the sparseness in the wavelet domain. The inverse solutions obtained from the new method using MEG data were evaluated by Monte Carlo simulations too. The present results indicated that cortical current densities could be efficiently compressed using the proposed face-based wavelet method, which exhibited better performance than the vertex-based wavelet method. In both simulations and auditory experimental data analysis, the proposed L1-norm regularization method showed better source detection accuracy and less estimation errors than other two classic methods, i.e. weighted minimum norm (wMNE) and cortical low-resolution electromagnetic tomography (cLORETA). This study suggests that the L1-norm regularization method with the use of face-based wavelets is a promising tool for studying functional activations of the human brain.
Time-frequency manifold sparse reconstruction: A novel method for bearing fault feature extraction
NASA Astrophysics Data System (ADS)
Ding, Xiaoxi; He, Qingbo
2016-12-01
In this paper, a novel transient signal reconstruction method, called time-frequency manifold (TFM) sparse reconstruction, is proposed for bearing fault feature extraction. This method introduces image sparse reconstruction into the TFM analysis framework. According to the excellent denoising performance of TFM, a more effective time-frequency (TF) dictionary can be learned from the TFM signature by image sparse decomposition based on orthogonal matching pursuit (OMP). Then, the TF distribution (TFD) of the raw signal in a reconstructed phase space would be re-expressed with the sum of learned TF atoms multiplied by corresponding coefficients. Finally, one-dimensional signal can be achieved again by the inverse process of TF analysis (TFA). Meanwhile, the amplitude information of the raw signal would be well reconstructed. The proposed technique combines the merits of the TFM in denoising and the atomic decomposition in image sparse reconstruction. Moreover, the combination makes it possible to express the nonlinear signal processing results explicitly in theory. The effectiveness of the proposed TFM sparse reconstruction method is verified by experimental analysis for bearing fault feature extraction.
Sparse matrix orderings for factorized inverse preconditioners
Benzi, M.; Tuama, M.
1998-09-01
The effect of reorderings on the performance of factorized sparse approximate inverse preconditioners is considered. It is shown that certain reorderings can be very beneficial both in the preconditioner construction phase and in terms of the rate of convergence of the preconditioned iteration.
Multilevel sparse functional principal component analysis.
Di, Chongzhi; Crainiceanu, Ciprian M; Jank, Wolfgang S
2014-01-29
We consider analysis of sparsely sampled multilevel functional data, where the basic observational unit is a function and data have a natural hierarchy of basic units. An example is when functions are recorded at multiple visits for each subject. Multilevel functional principal component analysis (MFPCA; Di et al. 2009) was proposed for such data when functions are densely recorded. Here we consider the case when functions are sparsely sampled and may contain only a few observations per function. We exploit the multilevel structure of covariance operators and achieve data reduction by principal component decompositions at both between and within subject levels. We address inherent methodological differences in the sparse sampling context to: 1) estimate the covariance operators; 2) estimate the functional principal component scores; 3) predict the underlying curves. Through simulations the proposed method is able to discover dominating modes of variations and reconstruct underlying curves well even in sparse settings. Our approach is illustrated by two applications, the Sleep Heart Health Study and eBay auctions.
Sparse models for correlative and integrative analysis of imaging and genetic data
Lin, Dongdong; Cao, Hongbao; Calhoun, Vince D.
2014-01-01
The development of advanced medical imaging technologies and high-throughput genomic measurements has enhanced our ability to understand their interplay as well as their relationship with human behavior by integrating these two types of datasets. However, the high dimensionality and heterogeneity of these datasets presents a challenge to conventional statistical methods; there is a high demand for the development of both correlative and integrative analysis approaches. Here, we review our recent work on developing sparse representation based approaches to address this challenge. We show how sparse models are applied to the correlation and integration of imaging and genetic data for biomarker identification. We present examples on how these approaches are used for the detection of risk genes and classification of complex diseases such as schizophrenia. Finally, we discuss future directions on the integration of multiple imaging and genomic datasets including their interactions such as epistasis. PMID:25218561
NASA Astrophysics Data System (ADS)
Moody, D. I.; Smith, D. A.; Heavner, M.; Hamlin, T.
2014-12-01
Ongoing research at Los Alamos National Laboratory studies the Earth's radiofrequency (RF) background utilizing satellite-based RF observations of terrestrial lightning. The Fast On-orbit Recording of Transient Events (FORTE) satellite, launched in 1997, provided a rich RF lightning database. Application of modern pattern recognition techniques to this dataset may further lightning research in the scientific community, and potentially improve on-orbit processing and event discrimination capabilities for future satellite payloads. We extend sparse signal processing techniques to radiofrequency (RF) transient signals, and specifically focus on improved signature extraction using sparse representations in data-adaptive dictionaries. We present various processing options and classification results for on-board discharges, and discuss robustness and potential for capability development.
A Low-Complexity Transceiver Design in Sparse Multipath Massive MIMO Channels
NASA Astrophysics Data System (ADS)
Yu, Yuehua; Wang, Peng; Chen, He; Li, Yonghui; Vucetic, Branka
2016-10-01
In this letter, we develop a low-complexity transceiver design, referred to as semi-random beam pairing (SRBP), for sparse multipath massive MIMO channels. By exploring a sparse representation of the MIMO channel in the virtual angular domain, we generate a set of transmit-receive beam pairs in a semi-random way to support the simultaneous transmission of multiple data streams. These data streams can be easily separated at the receiver via a successive interference cancelation (SIC) technique, and the power allocation among them are optimized based on the classical waterfilling principle. The achieved degree of freedom (DoF) and capacity of the proposed approach are analyzed. Simulation results show that, compared to the conventional singular value decomposition (SVD)-based method, the proposed transceiver design can achieve near-optimal DoF and capacity with a significantly lower computational complexity.
A sparse equivalent source method for near-field acoustic holography.
Fernandez-Grande, Efren; Xenaki, Angeliki; Gerstoft, Peter
2017-01-01
This study examines a near-field acoustic holography method consisting of a sparse formulation of the equivalent source method, based on the compressive sensing (CS) framework. The method, denoted Compressive-Equivalent Source Method (C-ESM), encourages spatially sparse solutions (based on the superposition of few waves) that are accurate when the acoustic sources are spatially localized. The importance of obtaining a non-redundant representation, i.e., a sensing matrix with low column coherence, and the inherent ill-conditioning of near-field reconstruction problems is addressed. Numerical and experimental results on a classical guitar and on a highly reactive dipole-like source are presented. C-ESM is valid beyond the conventional sampling limits, making wide-band reconstruction possible. Spatially extended sources can also be addressed with C-ESM, although in this case the obtained solution does not recover the spatial extent of the source.
An add-on video compression codec based on content-adaptive sparse super-resolution reconstructions
NASA Astrophysics Data System (ADS)
Yang, Shu; Jiang, Jianmin
2017-02-01
In this paper, we introduce an idea of content-adaptive sparse reconstruction to achieve optimized magnification quality for those down sampled video frames, to which two stages of pruning are applied to select the closest correlated images for construction of an over-complete dictionary and drive the sparse representation of its enlarged frame. In this way, not only the sampling and dictionary training process is accelerated and optimized in accordance with the input frame content, but also an add-on video compression codec can be further developed by applying such scheme as a preprocessor to any standard video compression algorithm. Our extensive experiments illustrate that (i) the proposed content-adaptive sparse reconstruction outperforms the existing benchmark in terms of super-resolution quality; (ii) When applied to H.264, one of the international video compression standards, the proposed add-on video codec can achieve three times more compression while maintaining competitive decoding quality.
Deformable MR Prostate Segmentation via Deep Feature Learning and Sparse Patch Matching.
Guo, Yanrong; Gao, Yaozong; Shen, Dinggang
2016-04-01
Automatic and reliable segmentation of the prostate is an important but difficult task for various clinical applications such as prostate cancer radiotherapy. The main challenges for accurate MR prostate localization lie in two aspects: (1) inhomogeneous and inconsistent appearance around prostate boundary, and (2) the large shape variation across different patients. To tackle these two problems, we propose a new deformable MR prostate segmentation method by unifying deep feature learning with the sparse patch matching. First, instead of directly using handcrafted features, we propose to learn the latent feature representation from prostate MR images by the stacked sparse auto-encoder (SSAE). Since the deep learning algorithm learns the feature hierarchy from the data, the learned features are often more concise and effective than the handcrafted features in describing the underlying data. To improve the discriminability of learned features, we further refine the feature representation in a supervised fashion. Second, based on the learned features, a sparse patch matching method is proposed to infer a prostate likelihood map by transferring the prostate labels from multiple atlases to the new prostate MR image. Finally, a deformable segmentation is used to integrate a sparse shape model with the prostate likelihood map for achieving the final segmentation. The proposed method has been extensively evaluated on the dataset that contains 66 T2-wighted prostate MR images. Experimental results show that the deep-learned features are more effective than the handcrafted features in guiding MR prostate segmentation. Moreover, our method shows superior performance than other state-of-the-art segmentation methods.
An ultra-sparse code underliesthe generation of neural sequences in a songbird
NASA Astrophysics Data System (ADS)
Hahnloser, Richard H. R.; Kozhevnikov, Alexay A.; Fee, Michale S.
2002-09-01
Sequences of motor activity are encoded in many vertebrate brains by complex spatio-temporal patterns of neural activity; however, the neural circuit mechanisms underlying the generation of these pre-motor patterns are poorly understood. In songbirds, one prominent site of pre-motor activity is the forebrain robust nucleus of the archistriatum (RA), which generates stereotyped sequences of spike bursts during song and recapitulates these sequences during sleep. We show that the stereotyped sequences in RA are driven from nucleus HVC (high vocal centre), the principal pre-motor input to RA. Recordings of identified HVC neurons in sleeping and singing birds show that individual HVC neurons projecting onto RA neurons produce bursts sparsely, at a single, precise time during the RA sequence. These HVC neurons burst sequentially with respect to one another. We suggest that at each time in the RA sequence, the ensemble of active RA neurons is driven by a subpopulation of RA-projecting HVC neurons that is active only at that time. As a population, these HVC neurons may form an explicit representation of time in the sequence. Such a sparse representation, a temporal analogue of the `grandmother cell' concept for object recognition, eliminates the problem of temporal interference during sequence generation and learning attributed to more distributed representations.
Segmentation of high angular resolution diffusion MRI using sparse riemannian manifold clustering.
Çetingül, H Ertan; Wright, Margaret J; Thompson, Paul M; Vidal, René
2014-02-01
We address the problem of segmenting high angular resolution diffusion imaging (HARDI) data into multiple regions (or fiber tracts) with distinct diffusion properties. We use the orientation distribution function (ODF) to model diffusion and cast the ODF segmentation problem as a clustering problem in the space of ODFs. Our approach integrates tools from sparse representation theory and Riemannian geometry into a graph theoretic segmentation framework. By exploiting the Riemannian properties of the space of ODFs, we learn a sparse representation for each ODF and infer the segmentation by applying spectral clustering to a similarity matrix built from these representations. In cases where regions with similar (resp. distinct) diffusion properties belong to different (resp. same) fiber tracts, we obtain the segmentation by incorporating spatial and user-specified pairwise relationships into the formulation. Experiments on synthetic data evaluate the sensitivity of our method to image noise and to the concentration parameters, and show its superior performance compared to alternative methods when analyzing complex fiber configurations. Experiments on phantom and real data demonstrate the accuracy of the proposed method in segmenting simulated fibers and white matter fiber tracts of clinical importance.
A path-oriented matrix-based knowledge representation system
NASA Technical Reports Server (NTRS)
Feyock, Stefan; Karamouzis, Stamos T.
1993-01-01
Experience has shown that designing a good representation is often the key to turning hard problems into simple ones. Most AI (Artificial Intelligence) search/representation techniques are oriented toward an infinite domain of objects and arbitrary relations among them. In reality much of what needs to be represented in AI can be expressed using a finite domain and unary or binary predicates. Well-known vector- and matrix-based representations can efficiently represent finite domains and unary/binary predicates, and allow effective extraction of path information by generalized transitive closure/path matrix computations. In order to avoid space limitations a set of abstract sparse matrix data types was developed along with a set of operations on them. This representation forms the basis of an intelligent information system for representing and manipulating relational data.
Sparse and compositionally robust inference of microbial ecological networks.
Kurtz, Zachary D; Müller, Christian L; Miraldi, Emily R; Littman, Dan R; Blaser, Martin J; Bonneau, Richard A
2015-05-01
16S ribosomal RNA (rRNA) gene and other environmental sequencing techniques provide snapshots of microbial communities, revealing phylogeny and the abundances of microbial populations across diverse ecosystems. While changes in microbial community structure are demonstrably associated with certain environmental conditions (from metabolic and immunological health in mammals to ecological stability in soils and oceans), identification of underlying mechanisms requires new statistical tools, as these datasets present several technical challenges. First, the abundances of microbial operational taxonomic units (OTUs) from amplicon-based datasets are compositional. Counts are normalized to the total number of counts in the sample. Thus, microbial abundances are not independent, and traditional statistical metrics (e.g., correlation) for the detection of OTU-OTU relationships can lead to spurious results. Secondly, microbial sequencing-based studies typically measure hundreds of OTUs on only tens to hundreds of samples; thus, inference of OTU-OTU association networks is severely under-powered, and additional information (or assumptions) are required for accurate inference. Here, we present SPIEC-EASI (SParse InversE Covariance Estimation for Ecological Association Inference), a statistical method for the inference of microbial ecological networks from amplicon sequencing datasets that addresses both of these issues. SPIEC-EASI combines data transformations developed for compositional data analysis with a graphical model inference framework that assumes the underlying ecological association network is sparse. To reconstruct the network, SPIEC-EASI relies on algorithms for sparse neighborhood and inverse covariance selection. To provide a synthetic benchmark in the absence of an experimentally validated gold-standard network, SPIEC-EASI is accompanied by a set of computational tools to generate OTU count data from a set of diverse underlying network topologies. SPIEC
Sparse and Compositionally Robust Inference of Microbial Ecological Networks
Kurtz, Zachary D.; Müller, Christian L.; Miraldi, Emily R.; Littman, Dan R.; Blaser, Martin J.; Bonneau, Richard A.
2015-01-01
16S ribosomal RNA (rRNA) gene and other environmental sequencing techniques provide snapshots of microbial communities, revealing phylogeny and the abundances of microbial populations across diverse ecosystems. While changes in microbial community structure are demonstrably associated with certain environmental conditions (from metabolic and immunological health in mammals to ecological stability in soils and oceans), identification of underlying mechanisms requires new statistical tools, as these datasets present several technical challenges. First, the abundances of microbial operational taxonomic units (OTUs) from amplicon-based datasets are compositional. Counts are normalized to the total number of counts in the sample. Thus, microbial abundances are not independent, and traditional statistical metrics (e.g., correlation) for the detection of OTU-OTU relationships can lead to spurious results. Secondly, microbial sequencing-based studies typically measure hundreds of OTUs on only tens to hundreds of samples; thus, inference of OTU-OTU association networks is severely under-powered, and additional information (or assumptions) are required for accurate inference. Here, we present SPIEC-EASI (SParse InversE Covariance Estimation for Ecological Association Inference), a statistical method for the inference of microbial ecological networks from amplicon sequencing datasets that addresses both of these issues. SPIEC-EASI combines data transformations developed for compositional data analysis with a graphical model inference framework that assumes the underlying ecological association network is sparse. To reconstruct the network, SPIEC-EASI relies on algorithms for sparse neighborhood and inverse covariance selection. To provide a synthetic benchmark in the absence of an experimentally validated gold-standard network, SPIEC-EASI is accompanied by a set of computational tools to generate OTU count data from a set of diverse underlying network topologies. SPIEC
A Comparison of Methods for Ocean Reconstruction from Sparse Observations
NASA Astrophysics Data System (ADS)
Streletz, G. J.; Kronenberger, M.; Weber, C.; Gebbie, G.; Hagen, H.; Garth, C.; Hamann, B.; Kreylos, O.; Kellogg, L. H.; Spero, H. J.
2014-12-01
We present a comparison of two methods for developing reconstructions of oceanic scalar property fields from sparse scattered observations. Observed data from deep sea core samples provide valuable information regarding the properties of oceans in the past. However, because the locations of sample sites are distributed on the ocean floor in a sparse and irregular manner, developing a global ocean reconstruction is a difficult task. Our methods include a flow-based and a moving least squares -based approximation method. The flow-based method augments the process of interpolating or approximating scattered scalar data by incorporating known flow information. The scheme exploits this additional knowledge to define a non-Euclidean distance measure between points in the spatial domain. This distance measure is used to create a reconstruction of the desired scalar field on the spatial domain. The resulting reconstruction thus incorporates information from both the scattered samples and the known flow field. The second method does not assume a known flow field, but rather works solely with the observed scattered samples. It is based on a modification of the moving least squares approach, a weighted least squares approximation method that blends local approximations into a global result. The modifications target the selection of data used for these local approximations and the construction of the weighting function. The definition of distance used in the weighting function is crucial for this method, so we use a machine learning approach to determine a set of near-optimal parameters for the weighting. We have implemented both of the reconstruction methods and have tested them using several sparse oceanographic datasets. Based upon these studies, we discuss the advantages and disadvantages of each method and suggest possible ways to combine aspects of both methods in order to achieve an overall high-quality reconstruction.
Simultaneous sparsity model for histopathological image representation and classification.
Srinivas, Umamahesh; Mousavi, Hojjat Seyed; Monga, Vishal; Hattel, Arthur; Jayarao, Bhushan
2014-05-01
The multi-channel nature of digital histopathological images presents an opportunity to exploit the correlated color channel information for better image modeling. Inspired by recent work in sparsity for single channel image classification, we propose a new simultaneous sparsity model for multi-channel histopathological image representation and classification (SHIRC). Essentially, we represent a histopathological image as a sparse linear combination of training examples under suitable channel-wise constraints. Classification is performed by solving a newly formulated simultaneous sparsity-based optimization problem. A practical challenge is the correspondence of image objects (cellular and nuclear structures) at different spatial locations in the image. We propose a robust locally adaptive variant of SHIRC (LA-SHIRC) to tackle this issue. Experiments on two challenging real-world image data sets: 1) mammalian tissue images acquired by pathologists of the animal diagnostics lab (ADL) at Pennsylvania State University, and 2) human intraductal breast lesions, reveal the merits of our proposal over state-of-the-art alternatives. Further, we demonstrate that LA-SHIRC exhibits a more graceful decay in classification accuracy against the number of training images which is highly desirable in practice where generous training per class is often not available.
Perception of biological motion from size-invariant body representations
Lappe, Markus; Wittinghofer, Karin; de Lussanet, Marc H. E.
2015-01-01
The visual recognition of action is one of the socially most important and computationally demanding capacities of the human visual system. It combines visual shape recognition with complex non-rigid motion perception. Action presented as a point-light animation is a striking visual experience for anyone who sees it for the first time. Information about the shape and posture of the human body is sparse in point-light animations, but it is essential for action recognition. In the posturo-temporal filter model of biological motion perception posture information is picked up by visual neurons tuned to the form of the human body before body motion is calculated. We tested whether point-light stimuli are processed through posture recognition of the human body form by using a typical feature of form recognition, namely size invariance. We constructed a point-light stimulus that can only be perceived through a size-invariant mechanism. This stimulus changes rapidly in size from one image to the next. It thus disrupts continuity of early visuo-spatial properties but maintains continuity of the body posture representation. Despite this massive manipulation at the visuo-spatial level, size-changing point-light figures are spontaneously recognized by naive observers, and support discrimination of human body motion. PMID:25852505
A survey of visual preprocessing and shape representation techniques
NASA Technical Reports Server (NTRS)
Olshausen, Bruno A.
1988-01-01
Many recent theories and methods proposed for visual preprocessing and shape representation are summarized. The survey brings together research from the fields of biology, psychology, computer science, electrical engineering, and most recently, neural networks. It was motivated by the need to preprocess images for a sparse distributed memory (SDM), but the techniques presented may also prove useful for applying other associative memories to visual pattern recognition. The material of this survey is divided into three sections: an overview of biological visual processing; methods of preprocessing (extracting parts of shape, texture, motion, and depth); and shape representation and recognition (form invariance, primitives and structural descriptions, and theories of attention).
Dense and Sparse Matrix Operations on the Cell Processor
Williams, Samuel W.; Shalf, John; Oliker, Leonid; Husbands,Parry; Yelick, Katherine
2005-05-01
The slowing pace of commodity microprocessor performance improvements combined with ever-increasing chip power demands has become of utmost concern to computational scientists. Therefore, the high performance computing community is examining alternative architectures that address the limitations of modern superscalar designs. In this work, we examine STI's forthcoming Cell processor: a novel, low-power architecture that combines a PowerPC core with eight independent SIMD processing units coupled with a software-controlled memory to offer high FLOP/s/Watt. Since neither Cell hardware nor cycle-accurate simulators are currently publicly available, we develop an analytic framework to predict Cell performance on dense and sparse matrix operations, using a variety of algorithmic approaches. Results demonstrate Cell's potential to deliver more than an order of magnitude better GFLOP/s per watt performance, when compared with the Intel Itanium2 and Cray X1 processors.
Understanding Representation in Design.
ERIC Educational Resources Information Center
Bodker, Susanne
1998-01-01
Discusses the design of computer applications, focusing on understanding design representations--what makes design representations work, and how, in different contexts. Examines the place of various types of representation (e.g., formal notations, models, prototypes, scenarios, and mock-ups) in design and the role of formalisms and representations…
Cognitive Dissonance as an Instructional Tool for Understanding Chemical Representations
ERIC Educational Resources Information Center
Corradi, David; Clarebout, Geraldine; Elen, Jan
2015-01-01
Previous research on multiple external representations (MER) indicates that sequencing representations (compared with presenting them as a whole) can, in some cases, increase conceptual understanding if there is interference between internal and external representations. We tested this mechanism by sequencing different combinations of scientific…
NASA Astrophysics Data System (ADS)
Lu, Pei; Xu, Zhiyong; Yu, Huapeng; Chang, Yongxin; Fu, Chengyu; Shao, Jianxin
2012-11-01
According to models of object recognition in cortex, the brain uses a hierarchical approach in which simple, low-level features having high position and scale specificity are pooled and combined into more complex, higher-level features having greater location invariance. At higher levels, spatial structure becomes implicitly encoded into the features themselves, which may overlap, while explicit spatial information is coded more coarsely. In this paper, the importance of sparsity and localized patch features in a hierarchical model inspired by visual cortex is investigated. As in the model of Serre, Wolf, and Poggio, we first apply Gabor filters at all positions and scales; feature complexity and position/scale invariance are then built up by alternating template matching and max pooling operations. In order to improve generalization performance, the sparsity is proposed and data dimension is reduced by means of compressive sensing theory and sparse representation algorithm. Similarly, within computational neuroscience, adding the sparsity on the number of feature inputs and feature selection is critical for learning biologically model from the statistics of natural images. Then, a redundancy dictionary of patch-based features that could distinguish object class from other categories is designed and then object recognition is implemented by the process of iterative optimization. The method is test on the UIUC car database. The success of this approach suggests a proof for the object class recognition in visual cortex.
A New Approach to Model Pitch Perception Using Sparse Coding
Furst, Miriam; Barak, Omri
2017-01-01
Our acoustical environment abounds with repetitive sounds, some of which are related to pitch perception. It is still unknown how the auditory system, in processing these sounds, relates a physical stimulus and its percept. Since, in mammals, all auditory stimuli are conveyed into the nervous system through the auditory nerve (AN) fibers, a model should explain the perception of pitch as a function of this particular input. However, pitch perception is invariant to certain features of the physical stimulus. For example, a missing fundamental stimulus with resolved or unresolved harmonics, or a low and high-level amplitude stimulus with the same spectral content–these all give rise to the same percept of pitch. In contrast, the AN representations for these different stimuli are not invariant to these effects. In fact, due to saturation and non-linearity of both cochlear and inner hair cells responses, these differences are enhanced by the AN fibers. Thus there is a difficulty in explaining how pitch percept arises from the activity of the AN fibers. We introduce a novel approach for extracting pitch cues from the AN population activity for a given arbitrary stimulus. The method is based on a technique known as sparse coding (SC). It is the representation of pitch cues by a few spatiotemporal atoms (templates) from among a large set of possible ones (a dictionary). The amount of activity of each atom is represented by a non-zero coefficient, analogous to an active neuron. Such a technique has been successfully applied to other modalities, particularly vision. The model is composed of a cochlear model, an SC processing unit, and a harmonic sieve. We show that the model copes with different pitch phenomena: extracting resolved and non-resolved harmonics, missing fundamental pitches, stimuli with both high and low amplitudes, iterated rippled noises, and recorded musical instruments. PMID:28099436
Solving large sparse eigenvalue problems on supercomputers
NASA Technical Reports Server (NTRS)
Philippe, Bernard; Saad, Youcef
1988-01-01
An important problem in scientific computing consists in finding a few eigenvalues and corresponding eigenvectors of a very large and sparse matrix. The most popular methods to solve these problems are based on projection techniques on appropriate subspaces. The main attraction of these methods is that they only require the use of the matrix in the form of matrix by vector multiplications. The implementations on supercomputers of two such methods for symmetric matrices, namely Lanczos' method and Davidson's method are compared. Since one of the most important operations in these two methods is the multiplication of vectors by the sparse matrix, methods of performing this operation efficiently are discussed. The advantages and the disadvantages of each method are compared and implementation aspects are discussed. Numerical experiments on a one processor CRAY 2 and CRAY X-MP are reported. Possible parallel implementations are also discussed.
Feature selection using sparse Bayesian inference
NASA Astrophysics Data System (ADS)
Brandes, T. Scott; Baxter, James R.; Woodworth, Jonathan
2014-06-01
A process for selecting a sparse subset of features that maximize discrimination between target classes is described in a Bayesian framework. Demonstrated on high range resolution radar (HRR) signature data, this has the effect of selecting the most informative range bins for a classification task. The sparse Bayesian classifier (SBC) model is directly compared against Fisher's linear discriminant analysis (LDA), showing a clear performance gain with the Bayesian framework using HRRs from the publicly available MSTAR data set. The discriminative power of the selected features from the SBC is shown to be particularly dominant over LDA when only a few features are selected or when there is a shift in training and testing data sets, as demonstrated by training on a specific target type and testing on a slightly different target type.
Sparse brain network using penalized linear regression
NASA Astrophysics Data System (ADS)
Lee, Hyekyoung; Lee, Dong Soo; Kang, Hyejin; Kim, Boong-Nyun; Chung, Moo K.
2011-03-01
Sparse partial correlation is a useful connectivity measure for brain networks when it is difficult to compute the exact partial correlation in the small-n large-p setting. In this paper, we formulate the problem of estimating partial correlation as a sparse linear regression with a l1-norm penalty. The method is applied to brain network consisting of parcellated regions of interest (ROIs), which are obtained from FDG-PET images of the autism spectrum disorder (ASD) children and the pediatric control (PedCon) subjects. To validate the results, we check their reproducibilities of the obtained brain networks by the leave-one-out cross validation and compare the clustered structures derived from the brain networks of ASD and PedCon.
Statistical prediction with Kanerva's sparse distributed memory
NASA Technical Reports Server (NTRS)
Rogers, David
1989-01-01
A new viewpoint of the processing performed by Kanerva's sparse distributed memory (SDM) is presented. In conditions of near- or over-capacity, where the associative-memory behavior of the model breaks down, the processing performed by the model can be interpreted as that of a statistical predictor. Mathematical results are presented which serve as the framework for a new statistical viewpoint of sparse distributed memory and for which the standard formulation of SDM is a special case. This viewpoint suggests possible enhancements to the SDM model, including a procedure for improving the predictiveness of the system based on Holland's work with genetic algorithms, and a method for improving the capacity of SDM even when used as an associative memory.
Causal Network Inference Via Group Sparse Regularization
Bolstad, Andrew; Van Veen, Barry D.; Nowak, Robert
2011-01-01
This paper addresses the problem of inferring sparse causal networks modeled by multivariate autoregressive (MAR) processes. Conditions are derived under which the Group Lasso (gLasso) procedure consistently estimates sparse network structure. The key condition involves a “false connection score” ψ. In particular, we show that consistent recovery is possible even when the number of observations of the network is far less than the number of parameters describing the network, provided that ψ < 1. The false connection score is also demonstrated to be a useful metric of recovery in nonasymptotic regimes. The conditions suggest a modified gLasso procedure which tends to improve the false connection score and reduce the chances of reversing the direction of causal influence. Computational experiments and a real network based electrocorticogram (ECoG) simulation study demonstrate the effectiveness of the approach. PMID:21918591
The Cortex Transform as an image preprocessor for sparse distributed memory: An initial study
NASA Technical Reports Server (NTRS)
Olshausen, Bruno; Watson, Andrew
1990-01-01
An experiment is described which was designed to evaluate the use of the Cortex Transform as an image processor for Sparse Distributed Memory (SDM). In the experiment, a set of images were injected with Gaussian noise, preprocessed with the Cortex Transform, and then encoded into bit patterns. The various spatial frequency bands of the Cortex Transform were encoded separately so that they could be evaluated based on their ability to properly cluster patterns belonging to the same class. The results of this study indicate that by simply encoding the low pass band of the Cortex Transform, a very suitable input representation for the SDM can be achieved.
SPARSE INTEGRATIVE CLUSTERING OF MULTIPLE OMICS DATA SETS
Wang, Sijian; Mo, Qianxing
2012-01-01
High resolution microarrays and second-generation sequencing platforms are powerful tools to investigate genome-wide alterations in DNA copy number, methylation, and gene expression associated with a disease. An integrated genomic profiling approach measuring multiple omics data types simultaneously in the same set of biological samples would render an integrated data resolution that would not be available with any single data type. In this study, we use penalized latent variable regression methods for joint modeling of multiple omics data types to identify common latent variables that can be used to cluster patient samples into biologically and clinically relevant disease subtypes. We consider lasso (Tibshirani, 1996), elastic net (Zou and Hastie, 2005), and fused lasso (Tibshirani et al., 2005) methods to induce sparsity in the coefficient vectors, revealing important genomic features that have significant contributions to the latent variables. An iterative ridge regression is used to compute the sparse coefficient vectors. In model selection, a uniform design (Fang and Wang, 1994) is used to seek “experimental” points that scattered uniformly across the search domain for efficient sampling of tuning parameter combinations. We compared our method to sparse singular value decomposition (SVD) and penalized Gaussian mixture model (GMM) using both real and simulated data sets. The proposed method is applied to integrate genomic, epigenomic, and transcriptomic data for subtype analysis in breast and lung cancer data sets. PMID:24587839
Dictionary learning and sparse recovery for electrodermal activity analysis
NASA Astrophysics Data System (ADS)
Kelsey, Malia; Dallal, Ahmed; Eldeeb, Safaa; Akcakaya, Murat; Kleckner, Ian; Gerard, Christophe; Quigley, Karen S.; Goodwin, Matthew S.
2016-05-01
Measures of electrodermal activity (EDA) have advanced research in a wide variety of areas including psychophysiology; however, the majority of this research is typically undertaken in laboratory settings. To extend the ecological validity of laboratory assessments, researchers are taking advantage of advances in wireless biosensors to gather EDA data in ambulatory settings, such as in school classrooms. While measuring EDA in naturalistic contexts may enhance ecological validity, it also introduces analytical challenges that current techniques cannot address. One limitation is the limited efficiency and automation of analysis techniques. Many groups either analyze their data by hand, reviewing each individual record, or use computationally inefficient software that limits timely analysis of large data sets. To address this limitation, we developed a method to accurately and automatically identify SCRs using curve fitting methods. Curve fitting has been shown to improve the accuracy of SCR amplitude and location estimations, but have not yet been used to reduce computational complexity. In this paper, sparse recovery and dictionary learning methods are combined to improve computational efficiency of analysis and decrease run time, while maintaining a high degree of accuracy in detecting SCRs. Here, a dictionary is first created using curve fitting methods for a standard SCR shape. Then, orthogonal matching pursuit (OMP) is used to detect SCRs within a dataset using the dictionary to complete sparse recovery. Evaluation of our method, including a comparison to for speed and accuracy with existing software, showed an accuracy of 80% and a reduced run time.
Sparse Gamma Rhythms Arising through Clustering in Adapting Neuronal Networks
Kilpatrick, Zachary P.; Ermentrout, Bard
2011-01-01
Gamma rhythms (30–100 Hz) are an extensively studied synchronous brain state responsible for a number of sensory, memory, and motor processes. Experimental evidence suggests that fast-spiking interneurons are responsible for carrying the high frequency components of the rhythm, while regular-spiking pyramidal neurons fire sparsely. We propose that a combination of spike frequency adaptation and global inhibition may be responsible for this behavior. Excitatory neurons form several clusters that fire every few cycles of the fast oscillation. This is first shown in a detailed biophysical network model and then analyzed thoroughly in an idealized model. We exploit the fact that the timescale of adaptation is much slower than that of the other variables. Singular perturbation theory is used to derive an approximate periodic solution for a single spiking unit. This is then used to predict the relationship between the number of clusters arising spontaneously in the network as it relates to the adaptation time constant. We compare this to a complementary analysis that employs a weak coupling assumption to predict the first Fourier mode to destabilize from the incoherent state of an associated phase model as the external noise is reduced. Both approaches predict the same scaling of cluster number with respect to the adaptation time constant, which is corroborated in numerical simulations of the full system. Thus, we develop several testable predictions regarding the formation and characteristics of gamma rhythms with sparsely firing excitatory neurons. PMID:22125486
Unsupervised analysis of polyphonic music by sparse coding.
Abdallah, Samer A; Plumbley, Mark D
2006-01-01
We investigate a data-driven approach to the analysis and transcription of polyphonic music, using a probabilistic model which is able to find sparse linear decompositions of a sequence of short-term Fourier spectra. The resulting system represents each input spectrum as a weighted sum of a small number of "atomic" spectra chosen from a larger dictionary; this dictionary is, in turn, learned from the data in such a way as to represent the given training set in an (information theoretically) efficient way. When exposed to examples of polyphonic music, most of the dictionary elements take on the spectral characteristics of individual notes in the music, so that the sparse decomposition can be used to identify the notes in a polyphonic mixture. Our approach differs from other methods of polyphonic analysis based on spectral decomposition by combining all of the following: (a) a formulation in terms of an explicitly given probabilistic model, in which the process estimating which notes are present corresponds naturally with the inference of latent variables in the model; (b) a particularly simple generative model, motivated by very general considerations about efficient coding, that makes very few assumptions about the musical origins of the signals being processed; and (c) the ability to learn a dictionary of atomic spectra (most of which converge to harmonic spectral profiles associated with specific notes) from polyphonic examples alone-no separate training on monophonic examples is required.
Notes on implementation of sparsely distributed memory
NASA Technical Reports Server (NTRS)
Keeler, J. D.; Denning, P. J.
1986-01-01
The Sparsely Distributed Memory (SDM) developed by Kanerva is an unconventional memory design with very interesting and desirable properties. The memory works in a manner that is closely related to modern theories of human memory. The SDM model is discussed in terms of its implementation in hardware. Two appendices discuss the unconventional approaches of the SDM: Appendix A treats a resistive circuit for fast, parallel address decoding; and Appendix B treats a systolic array for high throughput read and write operations.
Sparseness- and continuity-constrained seismic imaging
NASA Astrophysics Data System (ADS)
Herrmann, Felix J.
2005-04-01
Non-linear solution strategies to the least-squares seismic inverse-scattering problem with sparseness and continuity constraints are proposed. Our approach is designed to (i) deal with substantial amounts of additive noise (SNR < 0 dB); (ii) use the sparseness and locality (both in position and angle) of directional basis functions (such as curvelets and contourlets) on the model: the reflectivity; and (iii) exploit the near invariance of these basis functions under the normal operator, i.e., the scattering-followed-by-imaging operator. Signal-to-noise ratio and the continuity along the imaged reflectors are significantly enhanced by formulating the solution of the seismic inverse problem in terms of an optimization problem. During the optimization, sparseness on the basis and continuity along the reflectors are imposed by jointly minimizing the l1- and anisotropic diffusion/total-variation norms on the coefficients and reflectivity, respectively. [Joint work with Peyman P. Moghaddam was carried out as part of the SINBAD project, with financial support secured through ITF (the Industry Technology Facilitator) from the following organizations: BG Group, BP, ExxonMobil, and SHELL. Additional funding came from the NSERC Discovery Grants 22R81254.
Imaging black holes with sparse modeling
NASA Astrophysics Data System (ADS)
Honma, Mareki; Akiyama, Kazunori; Tazaki, Fumie; Kuramochi, Kazuki; Ikeda, Shiro; Hada, Kazuhiro; Uemura, Makoto
2016-03-01
We introduce a new imaging method for radio interferometry based on sparse- modeling. The direct observables in radio interferometry are visibilities, which are Fourier transformation of an astronomical image on the sky-plane, and incomplete sampling of visibilities in the spatial frequency domain results in an under-determined problem, which has been usually solved with 0 filling to un-sampled grids. In this paper we propose to directly solve this under-determined problem using sparse modeling without 0 filling, which realizes super resolution, i.e., resolution higher than the standard refraction limit. We show simulation results of sparse modeling for the Event Horizon Telescope (EHT) observations of super-massive black holes and demonstrate that our approach has significant merit in observations of black hole shadows expected to be realized in near future. We also present some results with the method applied to real data, and also discuss more advanced techniques for practical observations such as imaging with closure phase as well as treating the effect of interstellar scattering effect.
Stacked Predictive Sparse Decomposition for Classification of Histology Sections.
Chang, Hang; Zhou, Yin; Borowsky, Alexander; Barner, Kenneth; Spellman, Paul; Parvin, Bahram
2015-05-01
Image-based classification of histology sections, in terms of distinct components (e.g., tumor, stroma, normal), provides a series of indices for histology composition (e.g., the percentage of each distinct components in histology sections), and enables the study of nuclear properties within each component. Furthermore, the study of these indices, constructed from each whole slide image in a large cohort, has the potential to provide predictive models of clinical outcome. For example, correlations can be established between the constructed indices and the patients' survival information at cohort level, which is a fundamental step towards personalized medicine. However, performance of the existing techniques is hindered as a result of large technical variations (e.g., variations of color/textures in tissue images due to non-standard experimental protocols) and biological heterogeneities (e.g., cell type, cell state) that are always present in a large cohort. We propose a system that automatically learns a series of dictionary elements for representing the underlying spatial distribution using stacked predictive sparse decomposition. The learned representation is then fed into the spatial pyramid matching framework with a linear support vector machine classifier. The system has been evaluated for classification of distinct histological components for two cohorts of tumor types. Throughput has been increased by using of graphical processing unit (GPU), and evaluation indicates a superior performance results, compared with previous research.
Stacked Predictive Sparse Decomposition for Classification of Histology Sections
Zhou, Yin; Borowsky, Alexander; Barner, Kenneth; Spellman, Paul
2016-01-01
Image-based classification of histology sections, in terms of distinct components (e.g., tumor, stroma, normal), provides a series of indices for histology composition (e.g., the percentage of each distinct components in histology sections), and enables the study of nuclear properties within each component. Furthermore, the study of these indices, constructed from each whole slide image in a large cohort, has the potential to provide predictive models of clinical outcome. For example, correlations can be established between the constructed indices and the patients’ survival information at cohort level, which is a fundamental step towards personalized medicine. However, performance of the existing techniques is hindered as a result of large technical variations (e.g., variations of color/textures in tissue images due to non-standard experimental protocols) and biological heterogeneities (e.g., cell type, cell state) that are always present in a large cohort. We propose a system that automatically learns a series of dictionary elements for representing the underlying spatial distribution using stacked predictive sparse decomposition. The learned representation is then fed into the spatial pyramid matching framework with a linear support vector machine classifier. The system has been evaluated for classification of distinct histological components for two cohorts of tumor types. Throughput has been increased by using of graphical processing unit (GPU), and evaluation indicates a superior performance results, compared with previous research. PMID:27721567
A fast algorithm for reordering sparse matrices for parallel factorization
Lewis, J.G.; Peyton, B.W.; Pothen, A.
1989-01-01
Jess and Kees introduced a method for ordering a sparse symmetric matrix A for efficient parallel factorization. The parallel ordering is computed in two steps. First, the matrix A is ordered by some fill-reducing ordering. Second, a parallel ordering of A is computed from the filled graph that results from factoring A using the initial fill-reducing ordering. Among all orderings whose fill lies in the filled graph, this parallel ordering achieves the minimum number of parallel steps in the factorization of A. Jess and Kees did not specify the implementation details of an algorithm for either step of this scheme. Liu and Mirzaian (1987) designed an algorithm implementing the second step, but it has time and space requirements higher than the cost of computing common fill-reducing orderings. We present here a new fast algorithm that implements the parallel ordering step by exploiting the clique tree representation of a chordal graph. We succeed in reducing the cost of the parallel ordering step well below that of the fill-reducing step. Our algorithm has time and space complexity linear in the number of compressed subscripts of L, i.e., the sum of the sizes of the maximal cliques of the filled graph. Empirically we demonstrate running times nearly identical to Liu's heuristic Composite Rotations algorithm that approximates the minimum number of parallel steps. 21 refs., 3 figs., 4 tabs.
Sparse modeling applied to patient identification for safety in medical physics applications
NASA Astrophysics Data System (ADS)
Lewkowitz, Stephanie
Every scheduled treatment at a radiation therapy clinic involves a series of safety protocol to ensure the utmost patient care. Despite safety protocol, on a rare occasion an entirely preventable medical event, an accident, may occur. Delivering a treatment plan to the wrong patient is preventable, yet still is a clinically documented error. This research describes a computational method to identify patients with a novel machine learning technique to combat misadministration. The patient identification program stores face and fingerprint data for each patient. New, unlabeled data from those patients are categorized according to the library. The categorization of data by this face-fingerprint detector is accomplished with new machine learning algorithms based on Sparse Modeling that have already begun transforming the foundation of Computer Vision. Previous patient recognition software required special subroutines for faces and different tailored subroutines for fingerprints. In this research, the same exact model is used for both fingerprints and faces, without any additional subroutines and even without adjusting the two hyperparameters. Sparse modeling is a powerful tool, already shown utility in the areas of super-resolution, denoising, inpainting, demosaicing, and sub-nyquist sampling, i.e. compressed sensing. Sparse Modeling is possible because natural images are inherently sparse in some bases, due to their inherent structure. This research chooses datasets of face and fingerprint images to test the patient identification model. The model stores the images of each dataset as a basis (library). One image at a time is removed from the library, and is classified by a sparse code in terms of the remaining library. The Locally Competitive Algorithm, a truly neural inspired Artificial Neural Network, solves the computationally difficult task of finding the sparse code for the test image. The components of the sparse representation vector are summed by ℓ1 pooling
NASA Astrophysics Data System (ADS)
Masmoudi, Atef; Zouari, Sonia; Ghribi, Abdelaziz
2015-11-01
We propose a new adaptive block-wise lossless image compression algorithm, which is based on the so-called alphabet reduction scheme combined with an adaptive arithmetic coding (AC). This new encoding algorithm is particularly efficient for lossless compression of images with sparse and locally sparse histograms. AC is a very efficient technique for lossless data compression and produces a rate that is close to the entropy; however, a compression performance loss occurs when encoding images or blocks with a limited number of active symbols by comparison with the number of symbols in the nominal alphabet, which consists in the amplification of the zero frequency problem. Generally, most methods add one to the frequency count of each symbol from the nominal alphabet, which leads to a statistical model distortion, and therefore reduces the efficiency of the AC. The aim of this work is to overcome this drawback by assigning to each image block the smallest possible set including all the existing symbols called active symbols. This is an alternative of using the nominal alphabet when applying the conventional arithmetic encoders. We show experimentally that the proposed method outperforms several lossless image compression encoders and standards including the conventional arithmetic encoders, JPEG2000, and JPEG-LS.
Noor, Amina; Ahmad, Aitzaz; Serpedin, Erchin
2015-10-27
Network component analysis (NCA) is an important method for inferring transcriptional regulatory networks (TRNs) and recovering transcription factor activities (TFAs) using gene expression data, and the prior information about the connectivity matrix. The algorithms currently available crucially depend on the completeness of this prior information. However, inaccuracies in the measurement process may render incompleteness in the available knowledge about the connectivity matrix. Hence, computationally efficient algorithms are needed to overcome the possible incompleteness in the available data. We present a sparse network component analysis algorithm (sparseNCA), which incorporates the effect of incompleteness in the estimation of TRNs by imposing an additional sparsity constraint using the `1 norm, which results in a greater estimation accuracy. In order to improve the computational efficiency, an iterative re-weighted `2 method is proposed for the NCA problem which not only promotes sparsity but is hundreds of times faster than the `1 norm based solution. The performance of sparseNCA is rigorously compared to that of FastNCA and NINCA using synthetic data as well as real data. It is shown that sparseNCA outperforms the existing state-of-the-art algorithms both in terms of estimation accuracy and consistency with the added advantage of low computational complexity. The performance of sparseNCA compared to its predecessors is particularly pronounced in case of incomplete prior information about the sparsity of the network. Subnetwork analysis is performed on the E.coli data which reiterates the superior consistency of the proposed algorithm.
Computer aided surface representation
Barnhill, R.E.
1991-04-02
Modern computing resources permit the generation of large amounts of numerical data. These large data sets, if left in numerical form, can be overwhelming. Such large data sets are usually discrete points from some underlying physical phenomenon. Because we need to evaluate the phenomenon at places where we don't have data, a continuous representation (a surface'') is required. A simple example is a weather map obtained from a discrete set of weather stations. (For more examples including multi-dimensional ones, see the article by Dr. Rosemary Chang in the enclosed IRIS Universe). In order to create a scientific structure encompassing the data, we construct an interpolating mathematical surface which can evaluate at arbitrary locations. We can also display and analyze the results via interactive computer graphics. In our research we construct a very wide variety of surfaces for applied geometry problems that have sound theoretical foundations. However, our surfaces have the distinguishing feature that they are constructed to solve short or long term practical problems. This DOE-funded project has developed the premiere research team in the subject of constructing surfaces (3D and higher dimensional) that provide smooth representations of real scientific and engineering information, including state of the art computer graphics visualizations. However, our main contribution is in the development of fundamental constructive mathematical methods and visualization techniques which can be incorporated into a wide variety of applications. This project combines constructive mathematics, algorithms, and computer graphics, all applied to real problems. The project is a unique resource, considered by our peers to be a de facto national center for this type of research.
Slowness and Sparseness Lead to Place, Head-Direction, and Spatial-View Cells
Franzius, Mathias; Sprekeler, Henning; Wiskott, Laurenz
2007-01-01
We present a model for the self-organized formation of place cells, head-direction cells, and spatial-view cells in the hippocampal formation based on unsupervised learning on quasi-natural visual stimuli. The model comprises a hierarchy of Slow Feature Analysis (SFA) nodes, which were recently shown to reproduce many properties of complex cells in the early visual system [1]. The system extracts a distributed grid-like representation of position and orientation, which is transcoded into a localized place-field, head-direction, or view representation, by sparse coding. The type of cells that develops depends solely on the relevant input statistics, i.e., the movement pattern of the simulated animal. The numerical simulations are complemented by a mathematical analysis that allows us to accurately predict the output of the top SFA layer. PMID:17784780
Sparse Parallel MRI Based on Accelerated Operator Splitting Schemes
Xie, Weisi; Su, Zhenghang
2016-01-01
Recently, the sparsity which is implicit in MR images has been successfully exploited for fast MR imaging with incomplete acquisitions. In this paper, two novel algorithms are proposed to solve the sparse parallel MR imaging problem, which consists of l1 regularization and fidelity terms. The two algorithms combine forward-backward operator splitting and Barzilai-Borwein schemes. Theoretically, the presented algorithms overcome the nondifferentiable property in l1 regularization term. Meanwhile, they are able to treat a general matrix operator that may not be diagonalized by fast Fourier transform and to ensure that a well-conditioned optimization system of equations is simply solved. In addition, we build connections between the proposed algorithms and the state-of-the-art existing methods and prove their convergence with a constant stepsize in Appendix. Numerical results and comparisons with the advanced methods demonstrate the efficiency of proposed algorithms. PMID:27746824
Mental Representations Formed From Educational Website Formats
Elizabeth T. Cady; Kimberly R. Raddatz; Tuan Q. Tran; Bernardo de la Garza; Peter D. Elgin
2006-10-01
The increasing popularity of web-based distance education places high demand on distance educators to format web pages to facilitate learning. However, limited guidelines exist regarding appropriate writing styles for web-based distance education. This study investigated the effect of four different writing styles on reader’s mental representation of hypertext. Participants studied hypertext written in one of four web-writing styles (e.g., concise, scannable, objective, and combined) and were then administered a cued association task intended to measure their mental representations of the hypertext. It is hypothesized that the scannable and combined styles will bias readers to scan rather than elaborately read, which may result in less dense mental representations (as identified through Pathfinder analysis) relative to the objective and concise writing styles. Further, the use of more descriptors in the objective writing style will lead to better integration of ideas and more dense mental representations than the concise writing style.
Online learning control using adaptive critic designs with sparse kernel machines.
Xu, Xin; Hou, Zhongsheng; Lian, Chuanqiang; He, Haibo
2013-05-01
In the past decade, adaptive critic designs (ACDs), including heuristic dynamic programming (HDP), dual heuristic programming (DHP), and their action-dependent ones, have been widely studied to realize online learning control of dynamical systems. However, because neural networks with manually designed features are commonly used to deal with continuous state and action spaces, the generalization capability and learning efficiency of previous ACDs still need to be improved. In this paper, a novel framework of ACDs with sparse kernel machines is presented by integrating kernel methods into the critic of ACDs. To improve the generalization capability as well as the computational efficiency of kernel machines, a sparsification method based on the approximately linear dependence analysis is used. Using the sparse kernel machines, two kernel-based ACD algorithms, that is, kernel HDP (KHDP) and kernel DHP (KDHP), are proposed and their performance is analyzed both theoretically and empirically. Because of the representation learning and generalization capability of sparse kernel machines, KHDP and KDHP can obtain much better performance than previous HDP and DHP with manually designed neural networks. Simulation and experimental results of two nonlinear control problems, that is, a continuous-action inverted pendulum problem and a ball and plate control problem, demonstrate the effectiveness of the proposed kernel ACD methods.
Smoothed l0 Norm Regularization for Sparse-View X-Ray CT Reconstruction
Li, Ming; Peng, Chengtao; Guan, Yihui; Xu, Pin
2016-01-01
Low-dose computed tomography (CT) reconstruction is a challenging problem in medical imaging. To complement the standard filtered back-projection (FBP) reconstruction, sparse regularization reconstruction gains more and more research attention, as it promises to reduce radiation dose, suppress artifacts, and improve noise properties. In this work, we present an iterative reconstruction approach using improved smoothed l0 (SL0) norm regularization which is used to approximate l0 norm by a family of continuous functions to fully exploit the sparseness of the image gradient. Due to the excellent sparse representation of the reconstruction signal, the desired tissue details are preserved in the resulting images. To evaluate the performance of the proposed SL0 regularization method, we reconstruct the simulated dataset acquired from the Shepp-Logan phantom and clinical head slice image. Additional experimental verification is also performed with two real datasets from scanned animal experiment. Compared to the referenced FBP reconstruction and the total variation (TV) regularization reconstruction, the results clearly reveal that the presented method has characteristic strengths. In particular, it improves reconstruction quality via reducing noise while preserving anatomical features. PMID:27725935
A Hyperspherical Adaptive Sparse-Grid Method for High-Dimensional Discontinuity Detection
Zhang, Guannan; Webster, Clayton G.; Gunzburger, Max D.; ...
2015-06-24
This study proposes and analyzes a hyperspherical adaptive hierarchical sparse-grid method for detecting jump discontinuities of functions in high-dimensional spaces. The method is motivated by the theoretical and computational inefficiencies of well-known adaptive sparse-grid methods for discontinuity detection. Our novel approach constructs a function representation of the discontinuity hypersurface of an N-dimensional discontinuous quantity of interest, by virtue of a hyperspherical transformation. Then, a sparse-grid approximation of the transformed function is built in the hyperspherical coordinate system, whose value at each point is estimated by solving a one-dimensional discontinuity detection problem. Due to the smoothness of the hypersurface, the newmore » technique can identify jump discontinuities with significantly reduced computational cost, compared to existing methods. In addition, hierarchical acceleration techniques are also incorporated to further reduce the overall complexity. Rigorous complexity analyses of the new method are provided as are several numerical examples that illustrate the effectiveness of the approach.« less
A hyper-spherical adaptive sparse-grid method for high-dimensional discontinuity detection
Zhang, Guannan; Webster, Clayton G; Gunzburger, Max D; Burkardt, John V
2014-03-01
This work proposes and analyzes a hyper-spherical adaptive hi- erarchical sparse-grid method for detecting jump discontinuities of functions in high-dimensional spaces is proposed. The method is motivated by the the- oretical and computational inefficiencies of well-known adaptive sparse-grid methods for discontinuity detection. Our novel approach constructs a func- tion representation of the discontinuity hyper-surface of an N-dimensional dis- continuous quantity of interest, by virtue of a hyper-spherical transformation. Then, a sparse-grid approximation of the transformed function is built in the hyper-spherical coordinate system, whose value at each point is estimated by solving a one-dimensional discontinuity detection problem. Due to the smooth- ness of the hyper-surface, the new technique can identify jump discontinuities with significantly reduced computational cost, compared to existing methods. Moreover, hierarchical acceleration techniques are also incorporated to further reduce the overall complexity. Rigorous error estimates and complexity anal- yses of the new method are provided as are several numerical examples that illustrate the effectiveness of the approach.
Analysis of Feedback Mechanisms with Unknown Delay Using Sparse Multivariate Autoregressive Method
Ip, Edward H.; Zhang, Qiang; Sowinski, Tomasz; Simpson, Sean L.
2015-01-01
This paper discusses the study of two interacting processes in which a feedback mechanism exists between the processes. The study was motivated by problems such as the circadian oscillation of gene expression where two interacting protein transcriptions form both negative and positive feedback loops with long delays to equilibrium. Traditionally, data of this type could be examined using autoregressive analysis. However, in circadian oscillation the order of an autoregressive model cannot be determined a priori. We propose a sparse multivariate autoregressive method that incorporates mixed linear effects into regression analysis, and uses a forward-backward greedy search algorithm to select non-zero entries in the regression coefficients, the number of which is constrained not to exceed a pre-specified number. A small simulation study provides preliminary evidence of the validity of the method. Besides the circadian oscillation example, an additional example of blood pressure variations using data from an intervention study is used to illustrate the method and the interpretation of the results obtained from the sparse matrix method. These applications demonstrate how sparse representation can be used for handling high dimensional variables that feature dynamic, reciprocal relationships. PMID:26252637
Edge-preserving traveltime tomography with a sparse multiscale imaging constraint
NASA Astrophysics Data System (ADS)
Sun, Mengyao; Zhang, Jie
2016-08-01
Solving the near-surface statics problem is often the first step in land or shallow marine seismic data processing. Near-surface velocity structures can be very complex, with large velocity contrasts within a small depth range. First-arrival traveltime tomography is a common approach for near-surface imaging. However, first-arrival traveltime tomography generally produces smooth model solutions due to the Tikhonov regularization, which constrains the model for minimum structures. Failing to resolve high velocity contrasts may result in inaccurate static values for reflection imaging. In this study, we develop a sparse multiscale imaging constraint for traveltime tomography to address this issue. In this method, we assume that the velocity model is sparse under a known wavelet basis. According to the model sparse representation, we first obtain the low wavenumber velocity structures, followed by the finer features, by alternately solving two sets of inversion problems. The synthetic tests and two real data applications show that this method exhibits better performance in reconstructing near-surface models with high velocity contrasts.
R. KELSEY
2001-02-01
For focused applications with limited user and use application communities, XML can be the right choice for representation. It is easy to use, maintain, and extend and enjoys wide support in commercial and research sectors. When the knowledge and information to be represented is object-based and use of that knowledge and information is a high priority, then XML-based representation should be considered. This paper discusses some of the issues involved in using XML-based representation and presents an example application that successfully uses an XML-based representation.
Space-Time Approximation with Sparse Grids
Griebel, M; Oeltz, D; Vassilevski, P S
2005-04-14
In this article we introduce approximation spaces for parabolic problems which are based on the tensor product construction of a multiscale basis in space and a multiscale basis in time. Proper truncation then leads to so-called space-time sparse grid spaces. For a uniform discretization of the spatial space of dimension d with O(N{sup d}) degrees of freedom, these spaces involve for d > 1 also only O(N{sup d}) degrees of freedom for the discretization of the whole space-time problem. But they provide the same approximation rate as classical space-time Finite Element spaces which need O(N{sup d+1}) degrees of freedoms. This makes these approximation spaces well suited for conventional parabolic and for time-dependent optimization problems. We analyze the approximation properties and the dimension of these sparse grid space-time spaces for general stable multiscale bases. We then restrict ourselves to an interpolatory multiscale basis, i.e. a hierarchical basis. Here, to be able to handle also complicated spatial domains {Omega}, we construct the hierarchical basis from a given spatial Finite Element basis as follows: First we determine coarse grid points recursively over the levels by the coarsening step of the algebraic multigrid method. Then, we derive interpolatory prolongation operators between the respective coarse and fine grid points by a least squares approach. This way we obtain an algebraic hierarchical basis for the spatial domain which we then use in our space-time sparse grid approach. We give numerical results on the convergence rate of the interpolation error of these spaces for various space-time problems with two spatial dimensions. Also implementational issues, data structures and questions of adaptivity are addressed to some extent.
Improving EEG-Based Driver Fatigue Classification Using Sparse-Deep Belief Networks
Chai, Rifai; Ling, Sai Ho; San, Phyo Phyo; Naik, Ganesh R.; Nguyen, Tuan N.; Tran, Yvonne; Craig, Ashley; Nguyen, Hung T.
2017-01-01
This paper presents an improvement of classification performance for electroencephalography (EEG)-based driver fatigue classification between fatigue and alert states with the data collected from 43 participants. The system employs autoregressive (AR) modeling as the features extraction algorithm, and sparse-deep belief networks (sparse-DBN) as the classification algorithm. Compared to other classifiers, sparse-DBN is a semi supervised learning method which combines unsupervised learning for modeling features in the pre-training layer and supervised learning for classification in the following layer. The sparsity in sparse-DBN is achieved with a regularization term that penalizes a deviation of the expected activation of hidden units from a fixed low-level prevents the network from overfitting and is able to learn low-level structures as well as high-level structures. For comparison, the artificial neural networks (ANN), Bayesian neural networks (BNN), and original deep belief networks (DBN) classifiers are used. The classification results show that using AR feature extractor and DBN classifiers, the classification performance achieves an improved classification performance with a of sensitivity of 90.8%, a specificity of 90.4%, an accuracy of 90.6%, and an area under the receiver operating curve (AUROC) of 0.94 compared to ANN (sensitivity at 80.8%, specificity at 77.8%, accuracy at 79.3% with AUC-ROC of 0.83) and BNN classifiers (sensitivity at 84.3%, specificity at 83%, accuracy at 83.6% with AUROC of 0.87). Using the sparse-DBN classifier, the classification performance improved further with sensitivity of 93.9%, a specificity of 92.3%, and an accuracy of 93.1% with AUROC of 0.96. Overall, the sparse-DBN classifier improved accuracy by 13.8, 9.5, and 2.5% over ANN, BNN, and DBN classifiers, respectively. PMID:28326009
Improving EEG-Based Driver Fatigue Classification Using Sparse-Deep Belief Networks.
Chai, Rifai; Ling, Sai Ho; San, Phyo Phyo; Naik, Ganesh R; Nguyen, Tuan N; Tran, Yvonne; Craig, Ashley; Nguyen, Hung T
2017-01-01
This paper presents an improvement of classification performance for electroencephalography (EEG)-based driver fatigue classification between fatigue and alert states with the data collected from 43 participants. The system employs autoregressive (AR) modeling as the features extraction algorithm, and sparse-deep belief networks (sparse-DBN) as the classification algorithm. Compared to other classifiers, sparse-DBN is a semi supervised learning method which combines unsupervised learning for modeling features in the pre-training layer and supervised learning for classification in the following layer. The sparsity in sparse-DBN is achieved with a regularization term that penalizes a deviation of the expected activation of hidden units from a fixed low-level prevents the network from overfitting and is able to learn low-level structures as well as high-level structures. For comparison, the artificial neural networks (ANN), Bayesian neural networks (BNN), and original deep belief networks (DBN) classifiers are used. The classification results show that using AR feature extractor and DBN classifiers, the classification performance achieves an improved classification performance with a of sensitivity of 90.8%, a specificity of 90.4%, an accuracy of 90.6%, and an area under the receiver operating curve (AUROC) of 0.94 compared to ANN (sensitivity at 80.8%, specificity at 77.8%, accuracy at 79.3% with AUC-ROC of 0.83) and BNN classifiers (sensitivity at 84.3%, specificity at 83%, accuracy at 83.6% with AUROC of 0.87). Using the sparse-DBN classifier, the classification performance improved further with sensitivity of 93.9%, a specificity of 92.3%, and an accuracy of 93.1% with AUROC of 0.96. Overall, the sparse-DBN classifier improved accuracy by 13.8, 9.5, and 2.5% over ANN, BNN, and DBN classifiers, respectively.
Properties of Artifact Representations for Evolutionary Design
NASA Technical Reports Server (NTRS)
Hornby, Gregory S.
2004-01-01
To achieve evolutionary design systems that scale to the levels achieved by man-made artifacts we can look to their characteristics of modularity, hierarchy and regularity to guide us. For this we focus on design representations, since they strongly determine the ability of evolutionary design systems to evolve artifacts with these characteristics. We identify three properties of design representations - combination, control-flow and abstraction - and discuss how they relate to hierarchy, modularity and regularity.
Learning modulation of odor representations: new findings from Arc-indexed networks
Yuan, Qi; Harley, Carolyn W.
2014-01-01
We first review our understanding of odor representations in rodent olfactory bulb (OB) and anterior piriform cortex (APC). We then consider learning-induced representation changes. Finally we describe the perspective on network representations gained from examining Arc-indexed odor networks of awake rats. Arc-indexed networks are sparse and distributed, consistent with current views. However Arc provides representations of repeated odors. Arc-indexed repeated odor representations are quite variable. Sparse representations are assumed to be compact and reliable memory codes. Arc suggests this is not necessarily the case. The variability seen is consistent with electrophysiology in awake animals and may reflect top-down cortical modulation of context. Arc-indexing shows that distinct odors share larger than predicted neuron pools. These may be low-threshold neuronal subsets. Learning’s effect on Arc-indexed representations is to increase the stable or overlapping component of rewarded odor representations. This component can decrease for similar odors when their discrimination is rewarded. The learning effects seen are supported by electrophysiology, but mechanisms remain to be elucidated. PMID:25565958
Fast sparse Raman spectral unmixing for chemical fingerprinting and quantification
NASA Astrophysics Data System (ADS)
Yaghoobi, Mehrdad; Wu, Di; Clewes, Rhea J.; Davies, Mike E.
2016-10-01
Raman spectroscopy is a well-established spectroscopic method for the detection of condensed phase chemicals. It is based on scattered light from exposure of a target material to a narrowband laser beam. The information generated enables presumptive identification from measuring correlation with library spectra. Whilst this approach is successful in identification of chemical information of samples with one component, it is more difficult to apply to spectral mixtures. The capability of handling spectral mixtures is crucial for defence and security applications as hazardous materials may be present as mixtures due to the presence of degradation, interferents or precursors. A novel method for spectral unmixing is proposed here. Most modern decomposition techniques are based on the sparse decomposition of mixture and the application of extra constraints to preserve the sum of concentrations. These methods have often been proposed for passive spectroscopy, where spectral baseline correction is not required. Most successful methods are computationally expensive, e.g. convex optimisation and Bayesian approaches. We present a novel low complexity sparsity based method to decompose the spectra using a reference library of spectra. It can be implemented on a hand-held spectrometer in near to real-time. The algorithm is based on iteratively subtracting the contribution of selected spectra and updating the contribution of each spectrum. The core algorithm is called fast non-negative orthogonal matching pursuit, which has been proposed by the authors in the context of nonnegative sparse representations. The iteration terminates when the maximum number of expected chemicals has been found or the residual spectrum has a negligible energy, i.e. in the order of the noise level. A backtracking step removes the least contributing spectrum from the list of detected chemicals and reports it as an alternative component. This feature is particularly useful in detection of chemicals
All scale-free networks are sparse.
Del Genio, Charo I; Gross, Thilo; Bassler, Kevin E
2011-10-21
We study the realizability of scale-free networks with a given degree sequence, showing that the fraction of realizable sequences undergoes two first-order transitions at the values 0 and 2 of the power-law exponent. We substantiate this finding by analytical reasoning and by a numerical method, proposed here, based on extreme value arguments, which can be applied to any given degree distribution. Our results reveal a fundamental reason why large scale-free networks without constraints on minimum and maximum degree must be sparse.
Effective dimension reduction for sparse functional data.
Yao, F; Lei, E; Wu, Y
2015-06-01
We propose a method of effective dimension reduction for functional data, emphasizing the sparse design where one observes only a few noisy and irregular measurements for some or all of the subjects. The proposed method borrows strength across the entire sample and provides a way to characterize the effective dimension reduction space, via functional cumulative slicing. Our theoretical study reveals a bias-variance trade-off associated with the regularizing truncation and decaying structures of the predictor process and the effective dimension reduction space. A simulation study and an application illustrate the superior finite-sample performance of the method.
Effective dimension reduction for sparse functional data
YAO, F.; LEI, E.; WU, Y.
2015-01-01
Summary We propose a method of effective dimension reduction for functional data, emphasizing the sparse design where one observes only a few noisy and irregular measurements for some or all of the subjects. The proposed method borrows strength across the entire sample and provides a way to characterize the effective dimension reduction space, via functional cumulative slicing. Our theoretical study reveals a bias-variance trade-off associated with the regularizing truncation and decaying structures of the predictor process and the effective dimension reduction space. A simulation study and an application illustrate the superior finite-sample performance of the method. PMID:26566293
Parallel preconditioning techniques for sparse CG solvers
Basermann, A.; Reichel, B.; Schelthoff, C.
1996-12-31
Conjugate gradient (CG) methods to solve sparse systems of linear equations play an important role in numerical methods for solving discretized partial differential equations. The large size and the condition of many technical or physical applications in this area result in the need for efficient parallelization and preconditioning techniques of the CG method. In particular for very ill-conditioned matrices, sophisticated preconditioner are necessary to obtain both acceptable convergence and accuracy of CG. Here, we investigate variants of polynomial and incomplete Cholesky preconditioners that markedly reduce the iterations of the simply diagonally scaled CG and are shown to be well suited for massively parallel machines.
Partitioning sparse rectangular matrices for parallel processing
Kolda, T.G.
1998-05-01
The authors are interested in partitioning sparse rectangular matrices for parallel processing. The partitioning problem has been well-studied in the square symmetric case, but the rectangular problem has received very little attention. They will formalize the rectangular matrix partitioning problem and discuss several methods for solving it. They will extend the spectral partitioning method for symmetric matrices to the rectangular case and compare this method to three new methods -- the alternating partitioning method and two hybrid methods. The hybrid methods will be shown to be best.
Distributed memory compiler design for sparse problems
NASA Technical Reports Server (NTRS)
Wu, Janet; Saltz, Joel; Berryman, Harry; Hiranandani, Seema
1991-01-01
A compiler and runtime support mechanism is described and demonstrated. The methods presented are capable of solving a wide range of sparse and unstructured problems in scientific computing. The compiler takes as input a FORTRAN 77 program enhanced with specifications for distributing data, and the compiler outputs a message passing program that runs on a distributed memory computer. The runtime support for this compiler is a library of primitives designed to efficiently support irregular patterns of distributed array accesses and irregular distributed array partitions. A variety of Intel iPSC/860 performance results obtained through the use of this compiler are presented.
Sparse dynamics for partial differential equations
Schaeffer, Hayden; Caflisch, Russel; Hauck, Cory D.; Osher, Stanley
2013-01-01
We investigate the approximate dynamics of several differential equations when the solutions are restricted to a sparse subset of a given basis. The restriction is enforced at every time step by simply applying soft thresholding to the coefficients of the basis approximation. By reducing or compressing the information needed to represent the solution at every step, only the essential dynamics are represented. In many cases, there are natural bases derived from the differential equations, which promote sparsity. We find that our method successfully reduces the dynamics of convection equations, diffusion equations, weak shocks, and vorticity equations with high-frequency source terms. PMID:23533273
Establishing point correspondence of 3D faces via sparse facial deformable model.
Pan, Gang; Zhang, Xiaobo; Wang, Yueming; Hu, Zhenfang; Zheng, Xiaoxiang; Wu, Zhaohui
2013-11-01
Establishing a dense vertex-to-vertex anthropometric correspondence between 3D faces is an important and fundamental problem in 3D face research, which can contribute to most applications of 3D faces. This paper proposes a sparse facial deformable model to automatically achieve this task. For an input 3D face, the basic idea is to generate a new 3D face that has the same mesh topology as a reference face and the highly similar shape to the input face, and whose vertices correspond to those of the reference face in an anthropometric sense. Two constraints: 1) the shape constraint and 2) correspondence constraint are modeled in our method to satisfy the three requirements. The shape constraint is solved by a novel face deformation approach in which a normal-ray scheme is integrated to the closest-vertex scheme to keep high-curvature shapes in deformation. The correspondence constraint is based on an assumption that if the vertices on 3D faces are corresponded, their shape signals lie on a manifold and each face signal can be represented sparsely by a few typical items in a dictionary. The dictionary can be well learnt and contains the distribution information of the corresponded vertices. The correspondence information can be conveyed to the sparse representation of the generated 3D face. Thus, a patch-based sparse representation is proposed as the correspondence constraint. By solving the correspondence constraint iteratively, the vertices of the generated face can be adjusted to correspondence positions gradually. At the early iteration steps, smaller sparsity thresholds are set that yield larger representation errors but better globally corresponded vertices. At the later steps, relatively larger sparsity thresholds are used to encode local shapes. By this method, the vertices in the new face approach the right positions progressively until the final global correspondence is reached. Our method is automatic, and the manual work is needed only in training procedure
Regression-based adaptive sparse polynomial dimensional decomposition for sensitivity analysis
NASA Astrophysics Data System (ADS)
Tang, Kunkun; Congedo, Pietro; Abgrall, Remi
2014-11-01
Polynomial dimensional decomposition (PDD) is employed in this work for global sensitivity analysis and uncertainty quantification of stochastic systems subject to a large number of random input variables. Due to the intimate structure between PDD and Analysis-of-Variance, PDD is able to provide simpler and more direct evaluation of the Sobol' sensitivity indices, when compared to polynomial chaos (PC). Unfortunately, the number of PDD terms grows exponentially with respect to the size of the input random vector, which makes the computational cost of the standard method unaffordable for real engineering applications. In order to address this problem of curse of dimensionality, this work proposes a variance-based adaptive strategy aiming to build a cheap meta-model by sparse-PDD with PDD coefficients computed by regression. During this adaptive procedure, the model representation by PDD only contains few terms, so that the cost to resolve repeatedly the linear system of the least-square regression problem is negligible. The size of the final sparse-PDD representation is much smaller than the full PDD, since only significant terms are eventually retained. Consequently, a much less number of calls to the deterministic model is required to compute the final PDD coefficients.
Reweighted mass center based object-oriented sparse subspace clustering for hyperspectral images
NASA Astrophysics Data System (ADS)
Zhai, Han; Zhang, Hongyan; Zhang, Liangpei; Li, Pingxiang
2016-10-01
Considering the inevitable obstacles faced by the pixel-based clustering methods, such as salt-and-pepper noise, high computational complexity, and the lack of spatial information, a reweighted mass center based object-oriented sparse subspace clustering (RMC-OOSSC) algorithm for hyperspectral images (HSIs) is proposed. First, the mean-shift segmentation method is utilized to oversegment the HSI to obtain meaningful objects. Second, a distance reweighted mass center learning model is presented to extract the representative and discriminative features for each object. Third, assuming that all the objects are sampled from a union of subspaces, it is natural to apply the SSC algorithm to the HSI. Faced with the high correlation among the hyperspectral objects, a weighting scheme is adopted to ensure that the highly correlated objects are preferred in the procedure of sparse representation, to reduce the representation errors. Two widely used hyperspectral datasets were utilized to test the performance of the proposed RMC-OOSSC algorithm, obtaining high clustering accuracies (overall accuracy) of 71.98% and 89.57%, respectively. The experimental results show that the proposed method clearly improves the clustering performance with respect to the other state-of-the-art clustering methods, and it significantly reduces the computational time.
Atomic library optimization for pulse ultrasonic sparse signal decomposition and reconstruction
NASA Astrophysics Data System (ADS)
Song, Shoupeng; Li, Yingxue; Dogandžić, Aleksandar
2016-02-01
Compressive sampling of pulse ultrasonic NDE signals could bring significant savings in the data acquisition process. Sparse representation of these signals using an atomic library is key to their interpretation and reconstruction from compressive samples. However, the obstacles to practical applicability of such representations are: large size of the atomic library and computational complexity of the sparse decomposition and reconstruction. To help solve these problems, we develop a method for optimizing the ranges of parameters of traditional Gabor-atom library to match a real pulse ultrasonic signal in terms of correlation. As a result of atomic-library optimization, the number of the atoms is greatly reduced. Numerical simulations compare the proposed approach with the traditional method. Simulation results show that both the time efficiency and signal reconstruction energy error are superior to the traditional one even with small-scale atomic library. The performance of the proposed method is also explored under different noise levels. Finally, we apply the proposed method to real pipeline ultrasonic testing data, and the results indicate that our reduced atomic library outperforms the traditional library.
Locality-constrained Subcluster Representation Ensemble for lung image classification.
Song, Yang; Cai, Weidong; Huang, Heng; Zhou, Yun; Wang, Yue; Feng, David Dagan
2015-05-01
In this paper, we propose a new Locality-constrained Subcluster Representation Ensemble (LSRE) model, to classify high-resolution computed tomography (HRCT) images of interstitial lung diseases (ILDs). Medical images normally exhibit large intra-class variation and inter-class ambiguity in the feature space. Modelling of feature space separation between different classes is thus problematic and this affects the classification performance. Our LSRE model tackles this issue in an ensemble classification construct. The image set is first partitioned into subclusters based on spectral clustering with approximation-based affinity matrix. Basis representations of the test image are then generated with sparse approximation from the subclusters. These basis representations are finally fused with approximation- and distribution-based weights to classify the test image. Our experimental results on a large HRCT database show good performance improvement over existing popular classifiers.
Multilayer Extreme Learning Machine With Subnetwork Nodes for Representation Learning.
Yang, Yimin; Wu, Q M Jonathan
2015-10-09
The extreme learning machine (ELM), which was originally proposed for ''generalized'' single-hidden layer feedforward neural networks, provides efficient unified learning solutions for the applications of clustering, regression, and classification. It presents competitive accuracy with superb efficiency in many applications. However, ELM with subnetwork nodes architecture has not attracted much research attentions. Recently, many methods have been proposed for supervised/unsupervised dimension reduction or representation learning, but these methods normally only work for one type of problem. This paper studies the general architecture of multilayer ELM (ML-ELM) with subnetwork nodes, showing that: 1) the proposed method provides a representation learning platform with unsupervised/supervised and compressed/sparse representation learning and 2) experimental results on ten image datasets and 16 classification datasets show that, compared to other conventional feature learning methods, the proposed ML-ELM with subnetwork nodes performs competitively or much better than other feature learning methods.
Multilayer Extreme Learning Machine With Subnetwork Nodes for Representation Learning.
Yang, Yimin; Wu, Q M Jonathan
2016-11-01
The extreme learning machine (ELM), which was originally proposed for "generalized" single-hidden layer feedforward neural networks, provides efficient unified learning solutions for the applications of clustering, regression, and classification. It presents competitive accuracy with superb efficiency in many applications. However, ELM with subnetwork nodes architecture has not attracted much research attentions. Recently, many methods have been proposed for supervised/unsupervised dimension reduction or representation learning, but these methods normally only work for one type of problem. This paper studies the general architecture of multilayer ELM (ML-ELM) with subnetwork nodes, showing that: 1) the proposed method provides a representation learning platform with unsupervised/supervised and compressed/sparse representation learning and 2) experimental results on ten image datasets and 16 classification datasets show that, compared to other conventional feature learning methods, the proposed ML-ELM with subnetwork nodes performs competitively or much better than other feature learning methods.
OSKI: A Library of Automatically Tuned Sparse Matrix Kernels
Vuduc, R; Demmel, J W; Yelick, K A
2005-07-19
The Optimized Sparse Kernel Interface (OSKI) is a collection of low-level primitives that provide automatically tuned computational kernels on sparse matrices, for use by solver libraries and applications. These kernels include sparse matrix-vector multiply and sparse triangular solve, among others. The primary aim of this interface is to hide the complex decision-making process needed to tune the performance of a kernel implementation for a particular user's sparse matrix and machine, while also exposing the steps and potentially non-trivial costs of tuning at run-time. This paper provides an overview of OSKI, which is based on our research on automatically tuned sparse kernels for modern cache-based superscalar machines.
Inscriptions Becoming Representations in Representational Practices
ERIC Educational Resources Information Center
Medina, Richard; Suthers, Daniel
2013-01-01
We analyze the interaction of 3 students working on mathematics problems over several days in a virtual math team. Our analysis traces out how successful collaboration in a later session is contingent upon the work of prior sessions and shows how the development of representational practices is an important aspect of these participants' problem…
Embedded Data Representations.
Willett, Wesley; Jansen, Yvonne; Dragicevic, Pierre
2017-01-01
We introduce embedded data representations, the use of visual and physical representations of data that are deeply integrated with the physical spaces, objects, and entities to which the data refers. Technologies like lightweight wireless displays, mixed reality hardware, and autonomous vehicles are making it increasingly easier to display data in-context. While researchers and artists have already begun to create embedded data representations, the benefits, trade-offs, and even the language necessary to describe and compare these approaches remain unexplored. In this paper, we formalize the notion of physical data referents - the real-world entities and spaces to which data corresponds - and examine the relationship between referents and the visual and physical representations of their data. We differentiate situated representations, which display data in proximity to data referents, and embedded representations, which display data so that it spatially coincides with data referents. Drawing on examples from visualization, ubiquitous computing, and art, we explore the role of spatial indirection, scale, and interaction for embedded representations. We also examine the tradeoffs between non-situated, situated, and embedded data displays, including both visualizations and physicalizations. Based on our observations, we identify a variety of design challenges for embedded data representation, and suggest opportunities for future research and applications.
Protein family classification using sparse markov transducers.
Eskin, Eleazar; Noble, William Stafford; Singer, Yoram
2003-01-01
We present a method for classifying proteins into families based on short subsequences of amino acids using a new probabilistic model called sparse Markov transducers (SMT). We classify a protein by estimating probability distributions over subsequences of amino acids from the protein. Sparse Markov transducers, similar to probabilistic suffix trees, estimate a probability distribution conditioned on an input sequence. SMTs generalize probabilistic suffix trees by allowing for wild-cards in the conditioning sequences. Since substitutions of amino acids are common in protein families, incorporating wild-cards into the model significantly improves classification performance. We present two models for building protein family classifiers using SMTs. As protein databases become larger, data driven learning algorithms for probabilistic models such as SMTs will require vast amounts of memory. We therefore describe and use efficient data structures to improve the memory usage of SMTs. We evaluate SMTs by building protein family classifiers using the Pfam and SCOP databases and compare our results to previously published results and state-of-the-art protein homology detection methods. SMTs outperform previous probabilistic suffix tree methods and under certain conditions perform comparably to state-of-the-art protein homology methods.
Sparse spectrum model for a turbulent phase.
Charnotskii, Mikhail
2013-03-01
Monte Carlo (MC) simulation of phase front perturbations by atmospheric turbulence finds numerous applications for design and modeling of the adaptive optics systems, laser beam propagation simulations, and evaluating the performance of the various optical systems operating in the open air environment. Accurate generation of two-dimensional random fields of turbulent phase is complicated by the enormous diversity of scales that can reach five orders of magnitude in each coordinate. In addition there is a need for generation of the long "ribbons" of turbulent phase that are used to represent the time evolution of the wave front. This makes it unfeasible to use the standard discrete Fourier transform-based technique as a basis for the MC simulation algorithm. We propose a new model for turbulent phase: the sparse spectrum (SS) random field. The principal assumption of the SS model is that each realization of the random field has a discrete random spectral support. Statistics of the random amplitudes and wave vectors of the SS model are arranged to provide the required spectral and correlation properties of the random field. The SS-based MC model offers substantial reduction of computer costs for simulation of the wide-band random fields and processes, and is capable of generating long aperiodic phase "ribbons." We report the results of model trials that determine the number of sparse components, and the range of wavenumbers that is necessary to accurately reproduce the random field with a power-law spectrum.
Image reconstruction from photon sparse data
Mertens, Lena; Sonnleitner, Matthias; Leach, Jonathan; Agnew, Megan; Padgett, Miles J.
2017-01-01
We report an algorithm for reconstructing images when the average number of photons recorded per pixel is of order unity, i.e. photon-sparse data. The image optimisation algorithm minimises a cost function incorporating both a Poissonian log-likelihood term based on the deviation of the reconstructed image from the measured data and a regularization-term based upon the sum of the moduli of the second spatial derivatives of the reconstructed image pixel intensities. The balance between these two terms is set by a bootstrapping technique where the target value of the log-likelihood term is deduced from a smoothed version of the original data. When compared to the original data, the processed images exhibit lower residuals with respect to the true object. We use photon-sparse data from two different experimental systems, one system based on a single-photon, avalanche photo-diode array and the other system on a time-gated, intensified camera. However, this same processing technique could most likely be applied to any low photon-number image irrespective of how the data is collected. PMID:28169363
Image reconstruction from photon sparse data
NASA Astrophysics Data System (ADS)
Mertens, Lena; Sonnleitner, Matthias; Leach, Jonathan; Agnew, Megan; Padgett, Miles J.
2017-02-01
We report an algorithm for reconstructing images when the average number of photons recorded per pixel is of order unity, i.e. photon-sparse data. The image optimisation algorithm minimises a cost function incorporating both a Poissonian log-likelihood term based on the deviation of the reconstructed image from the measured data and a regularization-term based upon the sum of the moduli of the second spatial derivatives of the reconstructed image pixel intensities. The balance between these two terms is set by a bootstrapping technique where the target value of the log-likelihood term is deduced from a smoothed version of the original data. When compared to the original data, the processed images exhibit lower residuals with respect to the true object. We use photon-sparse data from two different experimental systems, one system based on a single-photon, avalanche photo-diode array and the other system on a time-gated, intensified camera. However, this same processing technique could most likely be applied to any low photon-number image irrespective of how the data is collected.
Sparse estimation of a covariance matrix.
Bien, Jacob; Tibshirani, Robert J
2011-12-01
We suggest a method for estimating a covariance matrix on the basis of a sample of vectors drawn from a multivariate normal distribution. In particular, we penalize the likelihood with a lasso penalty on the entries of the covariance matrix. This penalty plays two important roles: it reduces the effective number of parameters, which is important even when the dimension of the vectors is smaller than the sample size since the number of parameters grows quadratically in the number of variables, and it produces an estimate which is sparse. In contrast to sparse inverse covariance estimation, our method's close relative, the sparsity attained here is in the covariance matrix itself rather than in the inverse matrix. Zeros in the covariance matrix correspond to marginal independencies; thus, our method performs model selection while providing a positive definite estimate of the covariance. The proposed penalized maximum likelihood problem is not convex, so we use a majorize-minimize approach in which we iteratively solve convex approximations to the original nonconvex problem. We discuss tuning parameter selection and demonstrate on a flow-cytometry dataset how our method produces an interpretable graphical display of the relationship between variables. We perform simulations that suggest that simple elementwise thresholding of the empirical covariance matrix is competitive with our method for identifying the sparsity structure. Additionally, we show how our method can be used to solve a previously studied special case in which a desired sparsity pattern is prespecified.
Sparse Identification of Nonlinear Dynamics (SINDy)
NASA Astrophysics Data System (ADS)
Brunton, Steven; Proctor, Joshua; Kutz, Nathan
2016-11-01
This work develops a general new framework to discover the governing equations underlying a dynamical system simply from data measurements, leveraging advances in sparsity techniques and machine learning. The so-called sparse identification of nonlinear dynamics (SINDy) method results in models that are parsimonious, balancing model complexity with descriptive ability while avoiding over fitting. The only assumption about the structure of the model is that there are only a few important terms that govern the dynamics, so that the equations are sparse in the space of possible functions; this assumption holds for many physical systems in an appropriate basis. We demonstrate the algorithm on a wide range of problems, from simple canonical systems, including the chaotic Lorenz system, to the canonical fluid vortex shedding behind an circular cylinder at Re=100. We also show that this method generalizes to parameterized systems and systems that are time-varying or have external forcing. With abundant data and elusive laws, data-driven discovery of dynamics will continue to play an increasingly important role in the characterization and control of fluid dynamics.
Inferring sparse networks for noisy transient processes
NASA Astrophysics Data System (ADS)
Tran, Hoang M.; Bukkapatnam, Satish T. S.
2016-02-01
Inferring causal structures of real world complex networks from measured time series signals remains an open issue. The current approaches are inadequate to discern between direct versus indirect influences (i.e., the presence or absence of a directed arc connecting two nodes) in the presence of noise, sparse interactions, as well as nonlinear and transient dynamics of real world processes. We report a sparse regression (referred to as the -min) approach with theoretical bounds on the constraints on the allowable perturbation to recover the network structure that guarantees sparsity and robustness to noise. We also introduce averaging and perturbation procedures to further enhance prediction scores (i.e., reduce inference errors), and the numerical stability of -min approach. Extensive investigations have been conducted with multiple benchmark simulated genetic regulatory network and Michaelis-Menten dynamics, as well as real world data sets from DREAM5 challenge. These investigations suggest that our approach can significantly improve, oftentimes by 5 orders of magnitude over the methods reported previously for inferring the structure of dynamic networks, such as Bayesian network, network deconvolution, silencing and modular response analysis methods based on optimizing for sparsity, transients, noise and high dimensionality issues.
Fast Sparse Level Sets on Graphics Hardware.
Jalba, Andrei C; van der Laan, Wladimir J; Roerdink, Jos B T M
2013-01-01
The level-set method is one of the most popular techniques for capturing and tracking deformable interfaces. Although level sets have demonstrated great potential in visualization and computer graphics applications, such as surface editing and physically based modeling, their use for interactive simulations has been limited due to the high computational demands involved. In this paper, we address this computational challenge by leveraging the increased computing power of graphics processors, to achieve fast simulations based on level sets. Our efficient, sparse GPU level-set method is substantially faster than other state-of-the-art, parallel approaches on both CPU and GPU hardware. We further investigate its performance through a method for surface reconstruction, based on GPU level sets. Our novel multiresolution method for surface reconstruction from unorganized point clouds compares favorably with recent, existing techniques and other parallel implementations. Finally, we point out that both level-set computations and rendering of level-set surfaces can be performed at interactive rates, even on large volumetric grids. Therefore, many applications based on level sets can benefit from our sparse level-set method.
Image reconstruction from photon sparse data.
Mertens, Lena; Sonnleitner, Matthias; Leach, Jonathan; Agnew, Megan; Padgett, Miles J
2017-02-07
We report an algorithm for reconstructing images when the average number of photons recorded per pixel is of order unity, i.e. photon-sparse data. The image optimisation algorithm minimises a cost function incorporating both a Poissonian log-likelihood term based on the deviation of the reconstructed image from the measured data and a regularization-term based upon the sum of the moduli of the second spatial derivatives of the reconstructed image pixel intensities. The balance between these two terms is set by a bootstrapping technique where the target value of the log-likelihood term is deduced from a smoothed version of the original data. When compared to the original data, the processed images exhibit lower residuals with respect to the true object. We use photon-sparse data from two different experimental systems, one system based on a single-photon, avalanche photo-diode array and the other system on a time-gated, intensified camera. However, this same processing technique could most likely be applied to any low photon-number image irrespective of how the data is collected.
NASA Astrophysics Data System (ADS)
Takahashi, Hideaki; Tanabe, Kohsuke; Aketa, Masataka; Kishi, Ryohei; Furukawa, Shin-ichi; Nakano, Masayoshi
2007-02-01
The Beckmann rearrangement of acetone oxime promoted by proton transfers in the supercritical water has been investigated by means of the hybrid quantum mechanical/molecular mechanical approach combined with the theory of energy representation (QM/MM-ER) recently developed. The transition state (TS) structures have been explored by ab initio calculations for the reaction of hydrated acetone oxime on the assumption that the reaction is catalyzed by proton transfers along the hydrogen bonds connecting the solute and the solvent water molecules. Up to two water molecules have been considered as reactants that take part in the proton transfers. As a result of the density functional theory calculations with B3LYP functional and aug-cc-pVDZ basis set, it has been found that participation of two water molecules in the reaction reduces the activation free energy by -12.3kcal/mol. Furthermore, the QM/MM-ER simulations have revealed that the TS is more stabilized than the reactant state in the supercritical water by 2.7kcal/mol when two water molecules are involved in the reaction. Solvation free energies of the reactant and the TS have been decomposed into terms due to the electronic polarization of the solute, electron density fluctuation, and others to elucidate the origin of the stabilization of the TS as compared with the reactant. It has been revealed that the promotion of the chemical reaction due to the hydration mainly originates from the interaction between the nonpolarized solute and the solvent water molecules at the supercritical state.
Song, Rui; Kosorok, Michael; Zeng, Donglin; Zhao, Yingqi; Laber, Eric; Yuan, Ming
2015-01-01
As a new strategy for treatment which takes indi