Image Sparse Representation Research Articles

A Novel Image Denoising Algorithm Using Concepts of Quantum Many-Body Theory

Sparse representation of real-life images is a very effective approach in imaging applications, such as denoising. In recent years, with the growth of computing power, data-driven strategies exploiting the redundancy within patches extracted from one or several images to increase sparsity have become more prominent. This paper presents a novel image denoising algorithm exploiting such an image-dependent basis inspired by the quantum many-body theory. Based on patch analysis, the similarity measures in a local image neighborhood are formalized through a term akin to interaction in quantum mechanics that can efficiently preserve the local structures of real images. The versatile nature of this adaptive basis extends the scope of its application to image-independent or image-dependent noise scenarios without any adjustment. We carry out a rigorous comparison with contemporary methods to demonstrate the denoising capability of the proposed algorithm regardless of the image characteristics, noise statistics and intensity. We illustrate the properties of the hyperparameters and their respective effects on the denoising performance, together with automated rules of selecting their values close to the optimal one in experimental setups with ground truth not available. Finally, we show the ability of our approach to deal with practical images denoising problems such as medical ultrasound image despeckling applications.

Mixed noise reduction via sparse error constraint representation of high frequency image for wildlife image

Mixed noise reduction via sparse error constraint representation of high frequency image for wildlife image

Hyperspectral Image Classification via Deep Structure Dictionary Learning

The construction of diverse dictionaries for sparse representation of hyperspectral image (HSI) classification has been a hot topic over the past few years. However, compared with convolutional neural network (CNN) models, dictionary-based models cannot extract deeper spectral information, which will reduce their performance for HSI classification. Moreover, dictionary-based methods have low discriminative capability, which leads to less accurate classification. To solve the above problems, we propose a deep learning-based structure dictionary for HSI classification in this paper. The core ideas are threefold, as follows: (1) To extract the abundant spectral information, we incorporate deep residual neural networks in dictionary learning and represent input signals in the deep feature domain. (2) To enhance the discriminative ability of the proposed model, we optimize the structure of the dictionary and design sharing constraint in terms of sub-dictionaries. Thus, the general and specific feature of HSI samples can be learned separately. (3) To further enhance classification performance, we design two kinds of loss functions, including coding loss and discriminating loss. The coding loss is used to realize the group sparsity of code coefficients, in which within-class spectral samples can be represented intensively and effectively. The Fisher discriminating loss is used to enforce the sparse representation coefficients with large between-class scatter. Extensive tests performed on hyperspectral dataset with bright prospects prove the developed method to be effective and outperform other existing methods.

FABNet: Fusion Attention Block and Transfer Learning for Laryngeal Cancer Tumor Grading in P63 IHC Histopathology Images.

Laryngeal cancer tumor (LCT) grading is a challenging task in P63 Immunohistochemical (IHC) histopathology images due to small differences between LCT levels in pathology images, the lack of precision in lesion regions of interest (LROIs) and the paucity of LCT pathology image samples. The key to solving the LCT grading problem is to transfer knowledge from other images and to identify more accurate LROIs, but the following problems occur: 1) transferring knowledge without a priori experience often causes negative transfer and creates a heavy workload due to the abundance of image types, and 2) convolutional neural networks (CNNs) constructing deep models by stacking cannot sufficiently identify LROIs, often deviate significantly from the LROIs focused on by experienced pathologists, and are prone to providing misleading second opinions. So we propose a novel fusion attention block network (FABNet) to address these problems. First, we propose a model transfer method based on clinical a priori experience and sample analysis (CPESA) that analyzes the transfer ability by integrating clinical a priori experience using indicators such as the relationship between the cancer onset location and morphology and the texture and staining degree of cell nuclei in histopathology images; our method further validates these indicators by the probability distribution of cancer image samples. Then, we propose a fusion attention block (FAB) structure, which can both provide an advanced non-uniform sparse representation of images and extract spatial relationship information between nuclei; consequently, the LROI can be more accurate and more relevant to pathologists. We conducted extensive experiments, compared with the best Baseline model, the classification accuracy is improved 25%, and It is demonstrated that FABNet performs better on different cancer pathology image datasets and outperforms other state of the art (SOTA) models.

Medical Image Fusion Using Deep Learning Mechanism

Abstract: Sparse representation(SR) model named convolutional sparsity based morphological component analysis is introduced for pixel-level medical image fusion. The CS-MCA model can achieve multicomponent and global SRs of source images, by integrating MCA and convolutional sparse representation(CSR) into a unified optimization framework. In the existing method, the CSRs of its gradient and texture components are obtained by the CSMCA model using pre-learned dictionaries. Then for each image component, sparse coefficients of all the source images are merged and then fused component is reconstructed using the corresponding dictionary. In the extension mechanism, we are using deep learning based pyramid decomposition. Now a days deep learning is a very demanding technology. Deep learning is used for image classification, object detection, image segmentation, image restoration. Keywords: CNN, CT, MRI, MCA, CS-MCA.

A Neurodynamic Approach to Nonsmooth Quaternion Distributed Convex Optimization With Inequality and Affine Equality Constraints

Quaternions have appeared in many practical fields, such as image processing and data mining and so on. This paper focuses on designing an efficient quaternion-valued neurodynamic approach (QNA) based on multi-agent systems to solve nonsmooth convex quaternion distributed optimization problems (QDOPs) with inequality and affine equality constraints. Each agent in the system cooperatively solves the minimum of the global objective function through the information of itself and its neighbors. At first, the related convex analysis of the quaternion field are given, which provides a theoretical basis for solving the nonsmooth QDOP. Then the considered QDOP is equivalently transformed by using the connectivity of the communication topology. After that, a distributed QNA is presented, where the adaptive controller is introduced to ensure that the penalty terms with respect to the inequality constraints can self-adjust according to the local states. It is shown that all agents reach a consensus while obtaining the optimal solution to the related QDOP. Finally, a numerical example and an application in dictionary sparse representation of color images based on quaternion are realized to intuitively describe the effectiveness and practical significance of the proposed QNA.

A Novel Image Denoising Algorithm Using Concepts of Quantum Many-Body Theory

• An image denoising algorithm based on an image-dependent basis inspired by the quantum many-body theory. • The local structural similarity of images is formalized through a term akin to interaction in quantum physics. • The versatility of the adaptive basis extends its scope of applications regardless of the image characteristics and noise statistics. Sparse representation of real-life images is a very effective approach in imaging applications, such as denoising. In recent years, with the growth of computing power, data-driven strategies exploiting the redundancy within patches extracted from one or several images to increase sparsity have become more prominent. This paper presents a novel image denoising algorithm exploiting such an image-dependent basis inspired by the quantum many-body theory. Based on patch analysis, the similarity measures in a local image neighborhood are formalized through a term akin to interaction in quantum mechanics that can efficiently preserve the local structures of real images. The versatile nature of this adaptive basis extends the scope of its application to image-independent or image-dependent noise scenarios without any adjustment. We carry out a rigorous comparison with contemporary methods to demonstrate the denoising capability of the proposed algorithm regardless of the image characteristics, noise statistics and intensity. We illustrate the properties of the hyperparameters and their respective effects on the denoising performance, together with automated rules of selecting their values close to the optimal one in experimental setups with ground truth not available. Finally, we show the ability of our approach to deal with practical images denoising problems such as medical ultrasound image despeckling applications.

Quality Assessment of DIBR-Synthesized Views Based on Sparsity of Difference of Closings and Difference of Gaussians.

Images synthesized using depth-image-based-rendering (DIBR) techniques may suffer from complex structural distortions. The goal of the primary visual cortex and other parts of brain is to reduce redundancies of input visual signal in order to discover the intrinsic image structure, and thus create sparse image representation. Human visual system (HVS) treats images on several scales and several levels of resolution when perceiving the visual scene. With an attempt to emulate the properties of HVS, we have designed the no-reference model for the quality assessment of DIBR-synthesized views. To extract a higher-order structure of high curvature which corresponds to distortion of shapes to which the HVS is highly sensitive, we define a morphological oriented Difference of Closings (DoC) operator and use it at multiple scales and resolutions. DoC operator nonlinearly removes redundancies and extracts fine grained details, texture of an image local structure and contrast to which HVS is highly sensitive. We introduce a new feature based on sparsity of DoC band. To extract perceptually important low-order structural information (edges), we use the non-oriented Difference of Gaussians (DoG) operator at different scales and resolutions. Measure of sparsity is calculated for DoG bands to get scalar features. To model the relationship between the extracted features and subjective scores, the general regression neural network (GRNN) is used. Quality predictions by the proposed DoC-DoG-GRNN model show higher compatibility with perceptual quality scores in comparison to the tested state-of-the-art metrics when evaluated on four benchmark datasets with synthesized views, IRCCyN/IVC image/video dataset, MCL-3D stereoscopic image dataset and IST image dataset.

Directional Analytic Discrete Cosine Frames

Block frames called <i>directional analytic discrete cosine frames</i> (DADCFs) are proposed for sparse image representation. In contrast to conventional overlapped frames, the proposed DADCFs require a reduced amount of 1) computational complexity, 2) memory usage, and 3) global memory access. These characteristics are strongly required for current high-resolution image processing. Specifically, we propose two DADCFs based on discrete cosine transform (DCT) and discrete sine transform (DST). The first DADCF is constructed from parallel separable transforms of DCT and DST, where the DST is permuted by row. The second DADCF is also designed based on DCT and DST, while the DST is customized to have no DC leakage property which is a desirable property for image processing. Both DADCFs have rich directional selectivity with slightly different characteristics each other and they can be implemented as non-overlapping block-based transforms, especially they form Parseval block frames with low transform redundancy. We perform experiments to evaluate our DADCFs and compare them with conventional directional block transforms in image recovery.

Image Super-Resolution Through Compressive Sensing-based Recovery.

The primary aim of image super-resolution techniques is to produce a high resolution (HR) image from a low resolution (LR) image efficiently. Deep learning algorithms are being extensively used to address the ill-posed problem of single image super-resolution which requires extremely large data-sets and high processing power. When one does not have access to large data-sets or have limited processing power, an alternative technique may be in order. In this study, we have developed a novel positive scale image resizing method inspired by compressive sensing (CS). We have considered the image super-resolution as a CS recovery problem in which a low resolution image is assumed as a compressed measurement and the required interpolated image is treated as output of the CS-based recovery. In the proposed HR recovery method, a deterministic binary block diagonal measurement matrix, (DBBD), is used as measurement matrix since it maintains the visual similarity between the low and high resolution images. Then along with a sparsification matrix, the sparse representation of HR image is first recovered and subsequently the dense HR image is obtained. The proposed method is applied to medical and non-medical images. The HR images obtained using the traditional proximal, bilinear and bi-cubic interpolation techniques are compared with those obtained using the proposed method. The proposed CS inspired method delivers superior HR images than the traditional techniques. The superiority of the proposed method is attributed to the unique usage of the DBBD matrix and the CS recovery algorithm to obtain a high resolution image without any prior training data-set.

Sparse flight spotlight mode 3-D imaging of spaceborne SAR based on sparse spectrum and principal component analysis

The spaceborne synthetic aperture radar (SAR) sparse flight 3-D imaging technology through multiple observations of the cross-track direction is designed to form the cross-track equivalent aperture, and achieve the third dimensionality recognition. In this paper, combined with the actual triple star orbits, a sparse flight spaceborne SAR 3-D imaging method based on the sparse spectrum of interferometry and the principal component analysis (PCA) is presented. Firstly, interferometric processing is utilized to reach an effective sparse representation of radar images in the frequency domain. Secondly, as a method with simple principle and fast calculation, the PCA is introduced to extract the main features of the image spectrum according to its principal characteristics. Finally, the 3-D image can be obtained by inverse transformation of the reconstructed spectrum by the PCA. The simulation results of 4.84 km equivalent cross-track aperture and corresponding 1.78 m cross-track resolution verify the effective suppression of this method on high-frequency sidelobe noise introduced by sparse flight with a sparsity of 49% and random noise introduced by the receiver. Meanwhile, due to the influence of orbit distribution of the actual triple star orbits, the simulation results of the sparse flight with the 7-bit Barker code orbits are given as a comparison and reference to illuminate the significance of orbit distribution for this reconstruction results. This method has prospects for sparse flight 3-D imaging in high latitude areas for its short revisit period.

MRI reconstruction based on Bayesian piecewise sparsity constraint and adaptive 3D transform

Sparse representation of image is crucial to a promising reconstruction in compressed sensing magnetic resonance imaging (CS-MRI), which significantly accelerates imaging process by highly undersampling k-space data. However, realizing high sparsity and estimation accuracy of image coefficients directly affects the visual quality of the reconstructed image. In this work, an adaptive 3D transform learning method is developed to efficiently enhance the sparsity of coefficients produced by 3D transform on similar patches. In addition, we also consider the statistical characteristics of 3D coefficients, and build a probabilistic model that employs Gaussian scale mixture (GSM) prior for the grouped 3D coefficients with close magnitude levels. Based on Bayesian inference, a piecewise sparsity constraint is derived from maximum a posterior (MAP) estimation of 3D coefficients. Furthermore, the combination of piecewise sparsity constraint and adaptive 3D transform allows one to establish a novel CS-MRI reconstruction model and the corresponding numerical algorithms are deduced under the framework of alternating direction method of multipliers (ADMM). Compared to several advanced CS-MRI methods, the proposed approach better suppresses artifacts and preserves more image features with superior performance metrics.

Adversarial network integrating dual attention and sparse representation for semi-supervised semantic segmentation

Semantic segmentation is the important task of assigning a semantic label to each pixel. However, semantic segmentation based on the deep neural network usually requires massive annotations consumption to acquire better performance. To avoid the problem, some algorithms based on weakly-supervised and semi-supervised conditions have been proposed and achieved gradually improving performance in recent years. In this paper, we propose a novel semi-supervised adversarial network to alleviate the shortage of labeled data, which only requires a few labeled images to get competitive performance. The model is composed of two parts: the segmentation network and the discriminator network. The first part aims to semantically generate a segmented result that has the same size as the input color image. The discriminator network is designed in a fully convolutional manner to distinguish the predicted probability maps depending on the ground truth distribution. In particular, the probability maps are regarded as focal attention maps, which are fed back to the segmentation network to make the model converge faster, and the process can induce the model to focus on pixels that are hard to segment. To enhance the representation ability of image features, sparse representation and dual attention are adopted in the segmentation network. The sparse representation module aims to emphasize the object edges and locations by learning the convolutional sparse representation of the input color images, and the dual attention module can exploit the semantic interdependencies in two different dimensions. Moreover, the semi-supervised mechanism is introduced to the network, in which the adaptive parameter T that controls the sensitivity of the self-taught phase is proposed, and the training dataset is split into two parts for fully-supervised learning and semi-supervised learning. Specifically, the first part is unlabeled data, which is applied to provide supervised signals for semi-supervised training. The labeled data drawn from the other part is utilized for fully-supervised learning. Our semi-supervised adversarial framework can improve the learning ability and achieve higher performance, also, provide a novel approach to tackle the semantic segmentation task. Finally, comprehensive experiments on the PASCAL VOC 2012 and Cityscapes datasets are conducted to verify the effectiveness of the proposed model, which achieves the expected performance.

An Image Reconstruction Algorithm based on Sparse Representation for Image Compressed Sensing

It is difficult to control the balance between artifact suppression and detail preservation. In addition, the information contained in the reconstructed image is limited. For achieving the purpose of less lost information and lower computational complexity in the sampling process, this paper proposed a novel algorithm to realize the image reconstruction using sparse representation. Firstly, the principle of algorithm for sparse representation is introduced, and then the current commonly used reconstruction algorithms are described in detail. Finally, the algorithm can still process the image when the sparsity is unknown by introducing the sparsity theory and dynamically changing the step size to approximate the sparsity. The results explain that the improved algorithm can not only reconstruct the image with unknown sparsity, but also has advantages over other algorithms in reconstruction time. In addition, compared with other algorithms, the reconstruction time of the improved algorithm is the shortest under the same sampling rate.

C-CNN: Contourlet Convolutional Neural Networks.

Extracting effective features is always a challenging problem for texture classification because of the uncertainty of scales and the clutter of textural patterns. For texture classification, spectral analysis is traditionally employed in the frequency domain. Recent studies have shown the potential of convolutional neural networks (CNNs) when dealing with the texture classification task in the spatial domain. In this article, we try combining both approaches in different domains for more abundant information and proposed a novel network architecture named contourlet CNN (C-CNN). The network aims to learn sparse and effective feature representations for images. First, the contourlet transform is applied to get the spectral features from an image. Second, the spatial-spectral feature fusion strategy is designed to incorporate the spectral features into CNN architecture. Third, the statistical features are integrated into the network by the statistical feature fusion. Finally, the results are obtained by classifying the fusion features. We also investigated the behavior of the parameters in contourlet decomposition. Experiments on the widely used three texture data sets (kth-tips2-b, DTD, and CUReT) and five remote sensing data sets (UCM, WHU-RS, AID, RSSCN7, and NWPU-RESISC45) demonstrate that the proposed approach outperforms several well-known classification methods in terms of classification accuracy with fewer trainable parameters.

MRI reconstruction based on Bayesian group sparse representation

Compressed sensing (CS) theory speeds up the magnetic resonance imaging (MRI) by undersampling the k-space data. Utilizing a proper sparse representation for image and incorporating prior information are vital to yield a high-quality reconstruction in CS-MRI. In this work, the multivariate Gaussian scale mixture (GSM) model is developed to precisely characterize to the statistical properties of sparse coefficients of group formed by similar patches, and a Bayesian group sparse representation (BGSR) is derived from maximum a posterior (MAP) estimation. The efficient multiclass orthogonal dictionaries learning is further integrated in BGSR driven CS-MRI reconstruction to enable high sparsity as well as performance enhancement. The solution is obtained by a use of alternating direction method of multipliers (ADMM) iteration, and a closed-form solution for each subproblem is separately deduced. The experimental results show that the proposed method provides a superior visual quality and performance indexes over the regularization based CS-MRI methods. Hence, the proposed method can be utilized to further accelerate MRI and produce the highly accurate reconstruction for subsequent image processing as well as clinical diagnosis. Furthermore, this work can be extended to other imaging applications and provide some references for Bayesian probabilistic model based CS-MRI reconstruction.

Identification of wool and cashmere fibers based on multiscale geometric analysis

The identification of wool and cashmere fibers has always been an essential topic in the field of textile. In this work, a fiber recognition algorithm based on multiscale geometric analysis and deep convolutional neural network is introduced. As a new tool for signal analysis with multiscale and multi-direction, curvelet transform not only has multiscale and multi-resolution characteristics of the wavelet transform, but also has very strong directionality in the sense that it can provide an optimally sparse representation of fiber image with a large number of fiber contours information. The proposed method was based on multiscale geometric analysis to reduce the dimension of wool/cashmere fiber images and reduce the calculation of redundant data. The deep convolutional neural network was used to classify and recognize wool/cashmere fiber images. Then 400 fiber images of two kinds of fiber samples including wool and cashmere fibers, were collected respectively and processed by the methods including random interception and rotation to obtain a total of 600 fiber images respectively for the experiment analysis. The results show that the recognition accuracy is up to 96.67%. Compared with the traditional feature extraction and classification algorithm, this method dramatically improves model performance and identification accuracy.

Blind motion deblurring via [formula omitted] sparse representation

The method of blind deblurring based on machine learning can effectively deal with the blurred images in the real world. However, existing multi-level architectures can lead to problems such as the inability to reserve edges, the expected introduction of artifacts and ghosts when deblurring. Multi methods have found that using L0 norm to realize image sparse representation can help keeping main structure of images. In this paper, an edge extraction module based on L0 sparse representation is proposed to preserve the edge of images, which is embedded in a multi-scale recurrent network(SRN). When the current scale transmits information to the next scale, edge enhancement is performed using the edge extraction module. Furthermore, considering the correlation among pixels and the correlation among channels, we introduce dual-attention mechanism into the encoder-decoder structure. The deblurring experiment was carried out on the GOPRO dataset. Comparing with 5 state-of-art methods qualitatively and quantitatively, the experimental results show that the proposed method can better preserve the image edges and effectively avoid the artifact of the image. And the peak signal-to-noise ratio of the proposed method are improved compared with other methods.

A novel denoising algorithm for medical images based on the non‐convex non‐local similar adaptive regularization

Sparse representation is a powerful statistical image modelling technique and has been successfully applied to image denoising. For a given patch, a non-convex non-local similarity adaptive method is adopted for sparse representation of images. First, it uses the autoregressive model to perform dictionary learning from sample patch datasets. Second, the sparse representation of an image introduces non-convex non-local self-similarity as the regularization term. In order to make better use of the sparse regularization method for image denoising, the parameters used in this study are estimated using adaptive methods. This model is more efficient and accurate, Compared with K-means singular value decomposition (KSVD) algorithm, a generalized K-means clustering method, total variation of population sparsity (GSTV) algorithm, adaptive sparse domain selection (ASDS) algorithm, forward denoising convolutional neural network (DnCNNs), a fast and flexible Convolutional Neural Network image denoising method (FFNNet) and operator-splitting algorithm to minimize the Euler elastica functional (OSEEF). Image noise-reduction experiments confirmed that using the adaptive regularization method, the results in peak signal to noise ratio (PSNR) and visual opinion are better than other algorithms.

Compressed Sensing-Based Simultaneous Recovery of Magnitude and Phase MR Images via Dual Trigonometric Sparsity

Complex-valued Magnetic Resonance Imaging (MRI) is widely used in clinical diagnosis. The magnitude images are mainly used for structure visualization, and the phase images reveal tissue properties, such as magnetic susceptibility and fluid flow information. While MRI is slow, compressed sensing (CS) can be used to reconstruct the accelerated acquisitions. Current CS-based MRI reconstruction algorithms mostly focus on magnitude image recovery, with less attention to the recovery of phase images. In this paper, we propose a novel CS algorithm to simultaneously recover the magnitude and phase MR images based on the sparsity of the trigonometric function. The CS method requires a sparse representation of the original images, and it is observed the trigonometric functions of phase images in the Wavelet domain promote sparsity. Therefore, rather than transforming the phase images directly into the Wavelet domain, we calculate the sine and cosine of the phase images whose Wavelet transforms are set as the L1-norm regularization term. The combination of the dual trigonometric functions captures a unique, faithful four-quadrant phase information, which also improves the reconstructed magnitude images through an alternating optimization procedure. Reconstructions of simulated images and in vivo images are studied, and both show the superiority of the proposed method over compared phase recovery algorithms.