Sparse Low-rank Representation Research Articles

Cross-dataset discriminant subspace learning algorithm for apple leaf diseases identification

Apple has a large planting area and high yield in China, but it is easily affected by diseases. Artificial intelligence technology has achieved good results in apple leaf disease identification. However, the training and testing data in complex environments often come from different collection scenarios. There are significant differences in the training and testing datasets, which leads to a significant decline in the identification performance. To improve the generalization ability of apple leaf disease identification, we develop a cross-dataset discriminant subspace learning (CDDSL) algorithm by utilizing the idea of transfer learning, low-rank sparse representation, and maximum margin neighborhood preserving embedding (NPE) to cross-dataset scenarios. Firstly, based on the transfer learning and low-rank sparse representation, each target sample can be represented by a linear combination of source domain samples. The sparse constraint on the noise matrix weakens the influence of noise in the sample data. Then, CDDSL progressively projects the subspace features into the semantic space and establishes a discriminant classifier. Meanwhile, CDDSL utilizes the maximal margin NPE to improve the intraclass compactness and interclass separability of projection features, which can enhance the image identification performance. The experiments conducted on real-world apple leaf datasets verify the effectiveness of the CDDSL algorithm in cross-dataset scenario.

Robust transfer subspace learning based on low-rank and sparse representation for bearing fault diagnosis

With the development of industrial intelligence, data-driven fault diagnosis plays an important role in prognostics and health management. However, there is usually a large amount of unlabeled data from different working conditions, making cross-domain fault diagnosis unstable and inflexible. To deal with this issue, we propose two novel transfer subspace learning methods based on the low-rank sparse representation (LRSR), called LRSR-G and LRSR-R. Specifically, LRSR-G integrates an additional matrix with LRSR to characterize the Gaussian noise for robustness, as well as capture global and local structures. Furthermore, LRSR-R adaptively learns the label matrix from samples instead of using the binary labeling matrix in LRSR-G, thus providing the possibility to improve the flexibility. In addition, we develop two efficient algorithms using the alternating direction method of multipliers to solve the proposed LRSR-G and LRSR-R. Extensive experiments are conducted on the Case Western Reserve University dataset and Jiangnan University (JNU) dataset. The results show that the proposed LRSR-G and LRSR-R perform better than the existing methods, while LRSR-R has more potential in cross-domain fault diagnosis tasks.

Extraction method of typical IEQ spatial distributions based on low-rank sparse representation and multi-step clustering

Extraction method of typical IEQ spatial distributions based on low-rank sparse representation and multi-step clustering

CCIM-SLR: Incomplete multiview co-clustering by sparse low-rank representation

CCIM-SLR: Incomplete multiview co-clustering by sparse low-rank representation

Estimating Functional Brain Networks by Low-Rank Representation With Local Constraint.

The functional architecture undergoes alterations during the preclinical phase of Alzheimer's disease. Consequently, the primary research focus has shifted towards identifying Alzheimer's disease and its early stages by constructing a functional connectivity network based on resting-state fMRI data. Recent investigations show that as Alzheimer's Disease (AD) progresses, modular tissue and connections in the core brain areas of AD patients diminish. Sparse learning methods are powerful tools for understanding Functional Brain Networks (FBNs) with Regions of Interest (ROIs) and a connectivity matrix measuring functional coherence between them. However, these tools often focus exclusively on functional connectivity measures, neglecting the brain network's modularity. Modularity orchestrates dynamic activities within the FBN to execute intricate cognitive tasks. To provide a comprehensive delineation of the FBN, we propose a local similarity-constrained low-rank sparse representation (LSLRSR) method that encodes modularity information under a manifold-regularized network learning framework and further formulate it as a low-rank sparse graph learning problem, which can be solved by an efficient optimization algorithm. Specifically, for each modularity structure, the Schatten p-norm regularizer reduces the reconstruction error and provides a better approximation of the low-rank constraint. Furthermore, we adopt a manifold-regularized local similarity prior to infer the intricate relationship between subnetwork similarity and modularity, guiding the modeling of FBN. Additionally, the proximal average method approximates the joint solution's proximal map, and the resulting nonconvex optimization problems are solved using the alternating direction multiplier method (ADMM). Compared to state-of-the-art methods for constructing FBNs, our algorithm generates a more modular FBN. This lays the groundwork for further research into alterations in brain network modularity resulting from diseases.

ARGLRR: A Sparse Low-Rank Representation Single-Cell RNA-Sequencing Data Clustering Method Combined with a New Graph Regularization.

The development of single-cell transcriptome sequencing technologies has opened new ways to study biological phenomena at the cellular level. A key application of such technologies involves the employment of single-cell RNA sequencing (scRNA-seq) data to identify distinct cell types through clustering, which in turn provides evidence for revealing heterogeneity. Despite the promise of this approach, the inherent characteristics of scRNA-seq data, such as higher noise levels and lower coverage, pose major challenges to existing clustering methods and compromise their accuracy. In this study, we propose a method called Adjusted Random walk Graph regularization Sparse Low-Rank Representation (ARGLRR), a practical sparse subspace clustering method, to identify cell types. The fundamental low-rank representation (LRR) model is concerned with the global structure of data. To address the limited ability of the LRR method to capture local structure, we introduced adjusted random walk graph regularization in its framework. ARGLRR allows for the capture of both local and global structures in scRNA-seq data. Additionally, the imposition of similarity constraints into the LRR framework further improves the ability of the proposed model to estimate cell-to-cell similarity and capture global structural relationships between cells. ARGLRR surpasses other advanced comparison approaches on nine known scRNA-seq data sets judging by the results. In the normalized mutual information and Adjusted Rand Index metrics on the scRNA-seq data sets clustering experiments, ARGLRR outperforms the best-performing comparative method by 6.99% and 5.85%, respectively. In addition, we visualize the result using Uniform Manifold Approximation and Projection. Visualization results show that the usage of ARGLRR enhances the separation of different cell types within the similarity matrix.

Multi-view subspace clustering for learning joint representation via low-rank sparse representation

Multi-view subspace clustering for learning joint representation via low-rank sparse representation

Adaptive Weighted Low-Rank Sparse Representation for Multi-View Clustering

Ongoing researches on multiple view data are showing competitive behavior in the machine learning field. Multi-view clustering has gained widespread acceptance for managing multi-view data and improves clustering efficiency. Large dimensionality in data from various views has recently drawn a lot of interest from researchers. How to efficiently learns the appropriate lower dimensional subspace which can manage the valuable information from the diverse views is challenging and considerable issue. To concentrate on the mentioned issue, we asserted a novel clustering approach for multiple view data through low-rank representation. We consider the importance of each view by assigning the weight control factor. We combine consensus representation with the degree of disagreement among lower rank matrices. The single objective function unifies all factors. Furthermore, we give the efficient solution to update the variable and to optimized the objective function through the Augmented Lagrange’s Multiplier strategy. Real-world datasets are utilized in this study to exemplify the efficiency of the introduced technique, and it is contemplated to preceding algorithms to demonstrate its superiority.

Adaptive denoising for magnetic resonance image based on nonlocal structural similarity and low-rank sparse representation

Adaptive denoising for magnetic resonance image based on nonlocal structural similarity and low-rank sparse representation

Enhanced Sparse Low-Rank Representation via Nonconvex Regularization for Rotating Machinery Early Fault Feature Extraction

The weak fault feature extraction is key to early fault diagnosis of rotary machinery. However, the existing sparse low-rank fault feature extraction methods have the deficiency of underestimation and low peak signal-to-noise ratio. To solve these problems, this article presents an enhanced sparse low-rank (ESL) representation approach for weak fault feature extraction. Considering the periodic self-similarity and shift invariance of fault feature, a weighted dual approximation regularization is proposed for noise and fault irrelevant harmonic suppression, which provides a cornerstone for rotating machinery weak fault feature extraction. To be specific, the truncated nuclear norm (TNN) and weighted generalized minimax-concave (WGMC) penalty are leveraged to form the weighted dual approximation regularization so that it can inherit their superior properties. The TNN can capture the periodic self-similarity and shift-invariance structure of fault impulses while restraining the noise; the WGMC penalty can enhance the sparsity, restrain the fault irrelevant harmonics, and overcome the deficiency of underestimating the large amplitude components. Therefore, the proposed model can effectively extract the weak fault feature. The proposed approach is applied to bearing and planet gear fault diagnosis to evaluate its effectiveness. Comparison results show the significant improvements of the proposed ESL method, indicating that it has great potentials in fault diagnosis of rotating machinery.

Satellite Video Scene Classification Using Low-Rank Sparse Representation Two-Stream Networks

Satellite video scene classification (SVSC) is a challenging work in remote sensing. The main procedure of SVSC is spatial–temporal feature extraction. Unfortunately, massive numbers of dim small moving targets and the low signal-to-noise ratio (SNR) of satellite video bring great challenges to feature extraction. It is difficult to apply traditional feature extraction methods to SVSC because they are used to classify the actions of high-quality video. According to the theory of low-rank sparse decomposition, a Low-rank Sparse Representation Two-stream Network (LSRTN) is designed to increase the classification accuracy of two-stream networks. First, we propose a Low-rank Sparse Component Analysis Network (LSCAN) to decompose satellite videos into low-rank background images and sparse moving target sequences. The LSCAN possesses the advantage of low-rank sparse decomposition to solve small targets and has the capability to adjust the features using the data. Moreover, the LSCAN can efficiently improve the feature extraction of low SNR video. Second, a two-stream structure that was proven to be effective for multiclass video classification was applied to obtain the spatial features and temporal features in each stream. Finally, a fully connected layer integrates the features to classify the satellite video scenes. To utilize the label information, we refine the loss function to adjust the degree of low-rank sparse characteristics and ensure the classification accuracy of training. The experimental results demonstrate that the proposed method achieves better performance than the baseline methods for the SVSC task.

Fusion of Hyperspectral-Multispectral images joining Spatial-Spectral Dual-Dictionary and structured sparse Low-rank representation

High spatial resolution hyperspectral images (HR-HSIs) have shown considerable potential in urban green infrastructure monitoring. A prevalent scheme to overcome spatial resolution limitations in HSIs is by fusing low-resolution hyperspectral images (LR-HSIs) and high-resolution multispectral images (HR-MSIs). Existing methods considering the spectral dictionary or spatial dictionary can only reflect the unilateral characteristics of the HSI and cannot completely restore full information in the latent HSI. To overcome this issue, we propose a novel HSI-MSI fusion method, named DDSSLR, which joins spatial-spectral dual-dictionary and structured sparse low-rank representation. The spectral dictionary characterizing generalized spectra and the corresponding spectral sparse coefficients are extracted from LR-HSI and HR-MSI, while sparse low-rank priors of the local structure are imposed on the spectral pixels within the same superpixel in HR-MSI. Additionally, in the spatial domain, we exploit the remaining high-frequency components to learn the spatial dictionary and use the unitary transformation to factorize the spatial sparse coefficient into the sparse low-rank matrix in subspace, establishing the relationship between low-rank and sparse. We formulate the two fusion models as variational optimization problems, which are effectively solved by the alternating direction methods of multipliers (ADMM). Experiments on three HSI datasets show that DDSSLR achieves state-of-the-art performance.

Multi-view low rank sparse representation method for three-way clustering

Multi-view low rank sparse representation method for three-way clustering

Dynamic time warping similarity measurement based on low-rank sparse representation

Similarity measurement of time series is one of the focus issues in time series analysis and mining. Morphology maintenance of time series is a better way for performing a similarity measurement, and phase space reconstruction has the advantage of analyzing the morphology of time series, whereas it is prone to generate high-dimensional data. Linear dimensionality reduction methods have difficulty preserving complete information of time series data. Manifold-based learning methods can better preserve the local characteristics of data. Low-rank representation (LRR) finds the lowest rank representation of all data and is capable of capturing the global structure of data. Therefore, in this paper, we propose a dynamic time warping similarity measurement method based on low-rank sparse representation (LRSE_DTW) to reduce the dimensionality of time series data. We learn the low-rank sparse representation of the phase space and then embed it into low-dimensional space to maintain the morphology of the phase space. DTW is used to measure the distance between discriminant information obtained from the l2,1-norm constraint on the projection matrix. To confirm the effectiveness of LRSE_DTW, time series classification experiments are carried out on public UCR time series classification archive. The results show that LRSE_DTW is superior to several other state-of-the-art time series similarity measurement methods.

Tensor low-rank sparse representation for tensor subspace learning

Many subspace learning methods are implemented on a matrix of sample data. For multi-dimensional data, these methods have to convert data samples into vectors in advance, which often destroys the inherent spatial structure of the sample data. In this paper, we propose a robust tensor low-rank sparse representation (TLRSR) method that can directly perform subspace learning on three-dimensional tensors. Firstly, the dual constraints of low-rankness and sparseness make the representation tensor effectively capture the global structure and local structure of sample data, respectively. Secondly, in order to deal with outliers and noise, we adopt the tensor ℓ2,1-norm to characterize the noise of tensor composed of multiple samples. Thirdly, the denoised tensor instead of the original tensor is used as the dictionary to find the low-rank sparse representation tensor. Finally, an iterative update algorithm is proposed for the optimization of TLRSR, compared with the state-of-the-art methods, clustering on face images and denoising on real images verify the good performance of our proposed TLRSR in tensor subspace learning.

Nonlocal latent low rank sparse representation for single image super resolution via self-similarity learning

<p style='text-indent:20px;'>In this paper, we propose a novel scheme for single image super resolution (SR) reconstruction. Firstly, we construct a new self-similarity framework by regarding the low resolution (LR) images as the low rank version of corresponding high resolution (HR) images. Subsequently, nuclear norm minimization (NNM) is employed to generate LR image pyramids from HR ones. The structure of our framework is beneficial to extract LR features, where we regard the quotient image, calculated between HR image and LR image at the same layer, as LR feature. This LR feature has the same dimension as LR image; however the dimension of commonly used gradient feature is 4 times than LR image. On the other hand, we employ nonlocal similar patch, within the same scale and across different scales, to generate HR and LR dictionaries. In the course of encoding, codes are calculated from both row and column of LR dictionary for each LR patch; at the same time, both low rank and sparse constraints on codes matrix give us a hand to remove coding noises. Finally, both quantitative and perceptual results demonstrate that our proposed method has a good SR performance.</p>

Spatial-Spectral Structured Sparse Low-Rank Representation for Hyperspectral Image Super-Resolution.

Hyperspectral image super-resolution by fusing high-resolution multispectral image (HR-MSI) and low-resolution hyperspectral image (LR-HSI) aims at reconstructing high resolution spatial-spectral information of the scene. Existing methods mostly based on spectral unmixing and sparse representation are often developed from a low-level vision task perspective, they cannot sufficiently make use of the spatial and spectral priors available from higher-level analysis. To this issue, this paper proposes a novel HSI super-resolution method that fully considers the spatial/spectral subspace low-rank relationships between available HR-MSI/LR-HSI and latent HSI. Specifically, it relies on a new subspace clustering method named "structured sparse low-rank representation" (SSLRR), to represent the data samples as linear combinations of the bases in a given dictionary, where the sparse structure is induced by low-rank factorization for the affinity matrix. Then we exploit the proposed SSLRR model to learn the SSLRR along spatial/spectral domain from the MSI/HSI inputs. By using the learned spatial and spectral low-rank structures, we formulate the proposed HSI super-resolution model as a variational optimization problem, which can be readily solved by the ADMM algorithm. Compared with state-of-the-art hyperspectral super-resolution methods, the proposed method shows better performance on three benchmark datasets in terms of both visual and quantitative evaluation.

Laplacian regularized low-rank sparse representation transfer learning

In unsupervised transfer learning, it is extremely valuable to effectively extract knowledge from the vast amount of untagged data that exists by utilizing tagged data from other similar databases. In general, the data in the real world often resides in the low-dimensional manifold embedded in the high-dimensional environment space. However, the current subspace transfer learning methods do not consider the nonlinear geometry structure inside the data, so the local similarity information between the data may be lost in the learning process. In order to improve this respect, we propose a new subspace transfer learning algorithm, namely Laplacian Regularized Low-Rank Sparse Representation Transfer Learning (LRLRSR-TL). After introducing the low-rank representation and sparse constraints, the method incorporates Laplacian regularization term to represent the global low-dimensional structure and capture the inherent nonlinear geometry information of the data. Experimental investigation conducted based on five different cross-domain visual image datasets shows that the proposed method has outstanding performance compared with several state-of-the-art transfer learning methods.

Wasserstein autoencoders for collaborative filtering

The recommender systems have long been studied in the literature. The collaborative filtering is one of the most widely adopted recommendation techniques which is usually applied to the explicit data, e.g., rating scores. However, the implicit data, e.g., click data, is believed to be able to discover user’s latent preferences. Consequently, a number of research attempts have been made toward this issue. To the best of our knowledge, this paper is the first attempt to adapt the Wasserstein autoencoders to collaborative filtering problem. Particularly, we propose a new loss function by introducing an $$L_1$$ regularization term to learn a sparse low-rank representation form to represent latent variables. Then, we carefully design (1) a new cost function to minimize the data reconstruction error, and (2) the appropriate distance metrics for the calculation of KL divergence between the learned distribution of latent variables and the underlying true data distribution. Rigorous experiments are performed on three widely adopted datasets. Both the state-of-the-art approaches, e.g., Mult-VAE and Mult-DAE, and the baseline models are evaluated on these datasets. The promising experimental results demonstrate that the proposed approach is superior to the compared approaches with respect to evaluation criteria Recall@R and NDCG@R.

Double feature selection algorithm based on low-rank sparse non-negative matrix factorization

Recently, many feature selection algorithms based on non-negative matrix factorization have been proposed. However, many of these algorithms only consider unilateral information about global or local geometric structure normally. To this end, this paper proposes a new feature selection algorithm called double feature selection algorithm based on low-rank sparse non-negative matrix factorization (NMF-LRSR). Firstly, to reduce the dimensions effectively, NMF-LRSR uses non-negative matrix factorization as the framework to further reduce the dimension of the feature selection which is originally a dimension reduction problem. Secondly, the low-rank sparse representation with the self-representation is used to construct the graph, so both the global and intrinsic geometric structure information of the data could be taken into account in the process of feature selection, which makes full use of the information and makes the feature selection more accurate. In addition, the double feature selection theory is used to this paper, which makes the result of feature selection more accurate. NMF-LRSR is tested on the baseline and the other six algorithms in the literature and evaluated them on 11 publicly available benchmark datasets. Experimental results show that NMF-LRSR is more effective than the other six feature selection algorithms.