Image Sparse Representation Research Articles

Ultrasound image restoration based on a learned dictionary and a higher-order MRF

The restoration of images degraded by blur and multiplicative noise is a critical preprocessing step in medical ultrasound images which exhibit clinical diagnostic features of interest. This paper proposes a novel non-smooth non-convex variational model for ultrasound images denoising and deblurring motivated by the successes of sparse representation of images and FoE based approaches. Dictionaries are well adapted to textures and extended to arbitrary image sizes by defining a global image prior, while FoE image prior explicitly characterizes the statistics properties of natural image. Following these ideas, the new model is composed of the data-fidelity term, the sparse and redundant representations via learned dictionaries, and the FoE image prior model. The iPiano algorithm can efficiently deal with this optimization problem. The new proposed model is applied to several simulated images and real ultrasound images. The experimental results of denoising and deblurring show that proposed method gives a better visual effect by efficiently removing noise and preserving details well compared with two state-of-the-art methods.

SAR Target Configuration Recognition via Product Sparse Representation.

Sparse representation (SR) has been verified to be an effective tool for pattern recognition. Considering the multiplicative speckle noise in synthetic aperture radar (SAR) images, a product sparse representation (PSR) algorithm is proposed to achieve SAR target configuration recognition. To extract the essential characteristics of SAR images, the product model is utilized to describe SAR images. The advantages of sparse representation and the product model are combined to realize a more accurate sparse representation of the SAR image. Moreover, in order to weaken the influences of the speckle noise on recognition, the speckle noise of SAR images is modeled by the Gamma distribution, and the sparse vector of the SAR image is obtained from q statistical standpoint. Experiments are conducted on the moving and stationary target acquisition and recognition (MSTAR) database. The experimental results validate the effectiveness and robustness of the proposed algorithm, which can achieve higher recognition rates than some of the state-of-the-art algorithms under different circumstances.

Compressive sensing MRI reconstruction using empirical wavelet transform and grey wolf optimizer

Magnetic resonance imaging (MRI) has exhibited an outstanding performance in the track of medical imaging compared to several imaging modalities, such as X-ray, positron emission tomography and computed tomography. MRI modality suffers from protracted scanning time, which affects the psychological status of patients. This scanning time also increases the blurring levels in MR image due to local motion actions, such as breathing as in the case of cardiac imaging. An acquisition technique called compressed sensing has contributed to solve the drawbacks of MRI and decreased the acquisition time by reducing the quantity of the measured data that is needed to reconstruct an image without significant degradation in image quality. All recent works have used different types of conventional wavelets for sparsifying the image, which employ constant filter banks that are independent of the characteristics of the input image. This paper proposes to use the empirical wavelet transform (EWT) which tunes its filter banks to the characteristics of the analyzed images. In other words, we use EWT to produce a sparse representation of the MRI images which yields a more accurate sparsification transform. In addition, the grey wolf optimizer is used to optimize the parameters of the proposed method. To validate the proposed method, we use three MRI datasets of different organs: brain, cardiac and shoulder. The experimental results show that the proposed method outperforms the state-of-the-art methods in terms of signal-to-noise ratio and structure similarity metrics.

Video despeckling using Shearlet tensor-based anisotropic diffusion

This paper provides an effective method for video speckle noise reduction based on 3D Tensor-based Anisotropic Diffusion technique in the Shearlet domain (V-STAD). The proposed model exploits the multi-scale geometric representation and the sparsity property of the shearlet transform to apply a robust tensorial diffusion at each shearlet coefficients. In fact, the robustness of diffusion tensor image smoothing was not well investigated. Therefore, we adopted the robust Tukey's biweight function in the proposed tensor-based anisotropic diffusion. By this way, the filter benefits from robust statistics and sparse directional image representation property of the shearlet transform in addition to the intrinsic temporal correlations between frames to be adopted to the anisotropic nature of diffusion tensor. The experimental results demonstrate promising despeckling solution as compared to well-known state-of-the-art video denoising methods, like VBM3D and VIDOLSAT. The proposed method has clearly shown superior despeckling capability and it simultaneously demonstrated better local image structures preservation without introducing artifacts.

Bearing fault diagnosis based on spectrum image sparse representation of vibration signal

Bearings are crucial for industrial production and susceptible to malfunction in rotating machines. Image analysis can give a comprehensive description of vibration signal, thus, it has achieved much more attention recently in fault diagnosis field. However, it brings lots of redundant information from a single spectrum image matrix behind rich fault information, and massive spectrum image samples lead to exacerbation of this situation, which readily results in the accuracy-dropping problem of multiple local defective bearings diagnosis. To solve this issue, a novel feature extraction method based on image sparse representation is proposed. Original spectrum images are acquired through fast Fourier transformation. Sparse coefficient that reveals the underlying structure of spectrum image based on raw signals is extracted as the feature by implementing the orthogonal matching pursuit and K-singular value decomposition algorithm strategically, and then two-dimensional principal component analysis is applied for further processing of these features. Finally, fault types are identified based on a minimum distance strategy. The experimental results are given to demonstrate the effectiveness of the proposed method.

A competitive scheme for storing sparse representation of X-Ray medical images.

A competitive scheme for economic storage of the informational content of an X-Ray image, as it can be used for further processing, is presented. It is demonstrated that sparse representation of that type of data can be encapsulated in a small file without affecting the quality of the recovered image. The proposed representation, which is inscribed within the context of data reduction, provides a format for saving the image information in a way that could assist methodologies for analysis and classification. The competitiveness of the resulting file is compared against the compression standards JPEG and JPEG2000.

Sparse Range-Constrained Learning and Its Application for Medical Image Grading

Sparse learning has been shown to be effective in solving many real-world problems. Finding sparse representations is a fundamentally important topic in many fields of science including signal processing, computer vision, genome study, and medical imaging. One important issue in applying sparse representation is to find the basis to represent the data, especially in computer vision and medical imaging where the data are not necessary incoherent. In medical imaging, clinicians often grade the severity or measure the risk score of a disease based on images. This process is referred to as medical image grading. Manual grading of the disease severity or risk score is often used. However, it is tedious, subjective, and expensive. Sparse learning has been used for automatic grading of medical images for different diseases. In the grading, we usually begin with one step to find a sparse representation of the testing image using a set of reference images or atoms from the dictionary. Then in the second step, the selected atoms are used as references to compute the grades of the testing images. Since the two steps are conducted sequentially, the objective function in the first step is not necessarily optimized for the second step. In this paper, we propose a novel sparse range-constrained learning (SRCL) algorithm for medical image grading. Different from most of existing sparse learning algorithms, SRCL integrates the objective of finding a sparse representation and that of grading the image into one function. It aims to find a sparse representation of the testing image based on atoms that are most similar in both the data or feature representation and the medical grading scores. We apply the new proposed SRCL to two different applications, namely, cup-to-disc ratio (CDR) computation from retinal fundus images and cataract grading from slit-lamp lens images. Experimental results show that the proposed method is able to improve the accuracy in CDR computation and cataract grading.

A new cluster tendency assessment method for fuzzy co-clustering in hyperspectral image analysis

Abstract The assessment of cluster tendency is a method determining whether a considering data-set contains meaningful clusters. The raised questions often are: How many clusters is the data-set reasonably partitioned into and How is the data-set disposed? In this paper, we proposed a new assessment method of cluster tendency which is called Silhouette-Based Assessment of Cluster Tendency (SACT). The SACT algorithm appraises the cluster tendency of the data-set in terms of the number of clusters and the initial prototypes which can be used to simultaneously determine the suitable number of clusters and the prototypes. The information of the suitable number of clusters and the prototypes helps the clustering algorithms to improve the performance. The hyperspectral image analysis is one of the complex problems which need to improve the speed of the SACT algorithm by using the Image Patch Distance technique for sparse hyperspectral image representation, i.e., reducing the size of the input data of the SACT algorithm. Experiments were conducted on some labeled synthetic data sets, color images and hyperspectral images. The proposed algorithm exhibited high performance, reliability and accuracy compared to previously algorithms in the assessment of cluster tendency.

An improved coupled dictionary and multi-norm constraint fusion method for CT/MR medical images

To solve the problems that a single dictionary is difficult to obtain accurate sparse representation of images, and a single norm as activity level measurement of the source image block does not preserve more details of the image, leading to poor image fusion results, this paper proposes an improved coupled dictionary and multi-norm constraint image fusion method for CT/MR images. In the paper, CT and MR image pairs are used as training set, and the coupled CT dictionary and the MR dictionary are obtained by using the improved K-SVD algorithm respectively. The fusion dictionary is obtained by combining coupled CT dictionary and MR dictionary with the spatial domain method. First, the registered source images are compiled into the column vectors and the means are removed. The exact sparse representation coefficients are calculated by the CoefROMP algorithm under the fusion dictionary. Then the multi-norm constraint of the sparse representation coefficients is taken as activity level measurement of the source image blocks, and the sparse representation coefficients are fused by the rule of “choosing the maximum”. Finally, the fused images are obtained by reconstruction. The experimental results show that the proposed method in this paper can effectively retain more image details, improve fusion image contrast and clarity, focal prominent, accelerate the running speed of the algorithm and be applied to clinical diagnosis and auxiliary treatment.

Speckle Suppression Based on Sparse Representation with Non-Local Priors

As speckle seriously restricts the applications of remote sensing images in many fields, the ability to efficiently and effectively suppress speckle in a coherent imaging system is indispensable. In order to overcome the over-smoothing problem caused by the speckle suppression algorithm based on classical sparse representation, we propose a non-local speckle suppression algorithm that combines the non-local prior knowledge of the image into the sparse representation. The proposed algorithm first applies shearlet to sparsely represent the input image. We then incorporate the non-local priors as constraints into the image sparse representation de-noising problem. The denoised image is obtained by utilizing an alternating minimization algorithm to solve the corresponding constrained de-noising problem. The experimental results show that the proposed algorithm can not only significantly remove speckle noise, but also improve the visual effect and retain the texture information of the image better.

Fractional-order sparse representation for image denoising

Sparse representation models have been shown promising results for image denoising. However, conventional sparse representation-based models cannot obtain satisfactory estimations for sparse coefficients and the dictionary. To address this weakness, in this paper, we propose a novel fractional-order sparse representation U+0028 FSR U+0029 model. Specifically, we cluster the image patches into K groups, and calculate the singular values for each clean U+002F noisy patch pair in the wavelet domain. Then the uniform fractional-order parameters are learned for each cluster. Then a novel fractional-order sample space is constructed using adaptive fractional-order parameters in the wavelet domain to obtain more accurate sparse coefficients and dictionary for image denoising. Extensive experimental results show that the proposed model outperforms state-of-the-art sparse representation-based models and the block-matching and 3D filtering algorithm in terms of denoising performance and the computational efficiency.

Learning a collaborative multiscale dictionary based on robust empirical mode decomposition

Abstract Dictionary learning is a challenging topic in many image processing areas. The basic goal is to learn a sparse representation from an overcomplete basis set. Due to combining the advantages of generic multiscale representations with the learning-based adaptivity, multiscale dictionary representation approaches have the power in capturing structural characteristics of natural images. However, the existing multiscale learning approaches still suffer from three main weaknesses: inadaptability to diverse scales of image data, sensitivity to noise and outliers, difficulty to determine the optimal dictionary structure. In this paper, we present a novel multiscale dictionary learning paradigm for sparse image representations based on an improved empirical mode decomposition. This powerful data-driven analysis tool for multi-dimensional signals can fully adaptively decompose the image into multiscale oscillating components according to intrinsic modes of data self. This treatment can obtain a robust and effective sparse representation, and meanwhile generates a raw dictionary at multiple geometric scales and spatial frequency bands. This dictionary is refined by selecting the optimal oscillating atom based on frequency clustering. In order to further enhance sparsity and generalization, a tolerant dictionary is learned using a coherence regularized model. A fast proximal scheme is developed to optimize this model. The multiscale dictionary is considered as the product of an oscillating dictionary and a tolerant dictionary. Experimental results demonstrate that the proposed method has superior performance compared with several competing methods for sparse image representations. We also have shown the promising results in image denoising application.

Improving Medical CT Image Blind Restoration Algorithm Based on Dictionary Learning by Alternating Direction Method of Multipliers

In this paper, the medical CT image blind restoration is translated into two sub problems, namely, image estimation based on dictionary learning and point spread function estimation. A blind restoration algorithm optimized by the alternating direction method of multipliers for medical CT images was proposed. At present, the existing methods of blind image restoration based on dictionary learning have the problem of low efficiency and precision. This paper aims to improve the effectiveness and accuracy of the algorithm and to improve the robustness of the algorithm. The local CT images are selected as training samples, and the K-SVD algorithm is used to construct the dictionary by iterative optimization, which is beneficial to improve the efficiency of the algorithm. Then, the orthogonal matching pursuit algorithm is employed to implement the dictionary update. Dictionary learning is accomplished by sparse representation of medical CT images. The alternating direction method of multipliers (ADMM) is used to solve the objective function and realize the local image restoration, so as to eliminate the influence of point spread function. Secondly, the local restoration image is used to estimate the point spread function, and the convex quadratic optimization method is used to solve the point spread function sub problems. Finally, the optimal estimation of point spread function is obtained by iterative method, and the global sharp image is obtained by the alternating direction method of multipliers. Experimental results show that, compared with the traditional adaptive dictionary restoration algorithm, the new algorithm improves the objective image quality metrics, such as peak signal to noise ratio, structural similarity, and universal image quality index. The new algorithm optimizes the restoration effect, improves the robustness of noise immunity and improves the computing efficiency.

Image compression-based multiple description transform coding using NSCT and OMP approximation

In this paper, we present a novel multiple description transform image coding architecture, which uses an attractive transform called non-subsampled contourlet transform (NSCT). It combines NSCT and orthogonal matching pursuit algorithm (OMP) to give a sparse representation of images, aiming at solving the compression problem due to the redundancy property of NSCT. In this way, OMP turns to give a solution to remove the redundancies. We evaluate the performance of our image coder in the case of four descriptions that are dispatched over different channels. The experimentations show that the proposed method is efficient and the potential using NSCT than DWT in multiple description image coding, is evaluated by PSNR in each case of packet loss, where every description can reconstruct the image with acceptable fidelity, the later is much better if all descriptions are available.

Image compression-based multiple description transform coding using NSCT and OMP approximation

In this paper, we present a novel multiple description transform image coding architecture, which uses an attractive transform called non-subsampled contourlet transform (NSCT). It combines NSCT and orthogonal matching pursuit algorithm (OMP) to give a sparse representation of images, aiming at solving the compression problem due to the redundancy property of NSCT. In this way, OMP turns to give a solution to remove the redundancies. We evaluate the performance of our image coder in the case of four descriptions that are dispatched over different channels. The experimentations show that the proposed method is efficient and the potential using NSCT than DWT in multiple description image coding, is evaluated by PSNR in each case of packet loss, where every description can reconstruct the image with acceptable fidelity, the later is much better if all descriptions are available.

Discriminative Sparse Features for Alzheimer's Disease Diagnosis Using Multimodal Image Data.

Feature extraction in medical image processing still remains a challenge, especially in high-dimensionality datasets, where the expected number of available samples is considerably lower than the dimension of the feature space. This is a common problem in real-world data, and, specifically, in medical image pro- cessing as, while images are composed of hundreds of thousands voxels, only a reduced number of patients are available. Extracting descriptive and discriminative features to represent each sample (image) by a small number of features, which is particularly important in classification task, due to the curse of dimensionality problem. In this paper we solve this recognition problem by means of sparse representations of the data, which also provides an arena to multimodal image (PET and MRI) data classification by combining specialized classifiers. Thus, a novel method to effectively combine SVC classifiers is presented here, which uses the distance to the hyperplane computed for each class in each classifier allowing to select the most discriminative image modality in each case. The discriminative power of each modality also provides information about the illness evolution; while functional changes are clearly found in Alzheimer's diagnosed patients (AD) when compared to control subjects (CN), structural changes seem to be more relevant at the early stages of the illness, affecting Mild Cognitive Impairment (MCI) patients. Classification experiments using 68 CN, 70 AD and 111 MCI images from the Alzheimer's Disease Neuroimaging Initiative database have been performed and assessed by cross-validation to show the effectiveness of the proposed method. Accuracy values of up to 92% and 84% for CN/AD and CN/MCI classification are achieved. The method presented in this work shows that sparse representations of brain images are of importance for codifying and transferring relevant image features, as they may capture the salient features while maintaining lightweight data transactions. In fact, the method proposed in this work outperforms the classification results obtained using projection methods such as Principal Component Analysis for extracting representative features of the images.

Face recognition with Patch-based Local Walsh Transform

In this paper, we present a novel dense local image representation method called Local Walsh Transform (LWT) by applying the well-known Walsh Transform (WT) to each pixel of an image. The LWT decomposes an image into multiple components, and produces LWT complex images by using the symmetrical relationship between them. Cascaded LWT (CLWT) is also a dense local image representation obtained by applying the LWT again to real and imaginary parts of LWT complex images. Applying the LWT once more to real and imaginary parts of LWT complex images increases the success rate especially on low resolution images. In order to combine the advantages of sparse and dense local image representations, we present Patch-based LWT (PLWT) and Patch-based CLWT (PCLWT) by applying the LWT and CLWT, respectively, to patches extracted around landmarks of multi-scaled face images. The extracted high dimensional features of the patches are reduced through the application of the Whitened Principal Component Analysis (WPCA). Experimental results show that both the PLWT and PCLWT are robust to illumination and expression changes, occlusion and low resolution. The state-of-the-art performance is achieved on the FERET and SCface databases, and the second best unsupervised category result is achieved on the LFW database.

Joint Estimation of Dictionary and Image from Compressive Samples

Effective compressed sensing of images relies on the prior knowledge of a well-suited dictionary for sparse representation of the target image. In the absence of such knowledge, which is a typical scenario in real-world applications, the following question arises: Would it be advantageous to take an off-the-shelf overcomplete dictionary and fine-tune the dictionary with respect to (the observed part of) the image? A primary obstacle in this approach is overfitting, i.e., the loss in model generalization to the whole image. In this paper, we establish that local dictionary optimization using the compressive samples reduces the image recovery error—relative to the off-the-shelf recovery—with an overwhelming probability that depends on the sampling matrix. We present joint estimation of dictionary and image (JEDI), an iterative algorithm for dictionary fine-tuning from compressive samples and analyze its performance for image recovery. Our algorithmic analysis is supplemented with numerical simulations under different random sampling patterns and off-the-shelf dictionary initializations.

RETRACTED ARTICLE: Image denoising based on parallel K-singular value decomposition in cloud computing

In order to effectively apply the image sparse representation algorithm to the cloud computing platform in order to improve the efficiency of image processing, a K-singular value decomposition parallel optimization scheme based on Spark platform is proposed and applied to image denoising. Firstly, randomly selected partial image blocks are used to generate the training signal set. Then, the trained dictionary is generated by alternately iterating between the sparse code stage and the dictionary update stage, and then the sparse representation of the whole image is performed by the orthogonal matching tracking algorithm, So as to achieve the effect of denoising. In the sparse code stage, the sparse matrix is approximated by orthogonal matching tracking algorithm. In order to reduce the transmission and calculation of data, each sparse vector is recorded with a 3-tuple structure. In the dictionary update phase, the method is updated in a separate way for each atom. The experimental results show that the proposed method not only can effectively remove the image noise, retain the image texture details, but also has a better acceleration effect.

Exploration of ISAR Imaging Based on Compressed Sensing

The traditional compressed sensing ISAR imaging algorithm is based on row and column stacking, the image matrix is arranged as a long vector, and then a one-dimensional compressed sensing method is used to obtain the sparse representation of the target image. The biggest problem with traditional methods is that the stacked vectors are so large that the measurement matrices are large, making the algorithms very time-consuming. In addition, the traditional algorithm in the real domain, so in the processing of ISAR imaging data, need to complex measurement matrix diagonalization of real processing results such that the expected circular constraint space into square constraint space, increases the range of spatial constraints will inevitably increase the algorithm error. In this paper, a complex domain compressed sensing algorithm for two-dimensional space is proposed. The processing result proves the feasibility of the algorithm.