2 November 2023 in IEEE Transactions on Geoscience and Remote Sensing
Recently, hyperspectral anomaly detection (HAD) methods based on tensor low-rank representation (TLRR) have received widespread attention. However, most of them tend to emphasize the utilization of multiple types of prior knowledge to characterize background components, while the prior information about anomaly components is limited. Additionally, the constructed background dictionary is also susceptible to noise and outliers. To address these challenges, this paper focuses on both the background and abnormal components, proposing a robust tensor low-rank sparse representation with saliency prior (RTLSR-SP) method for HAD. Specifically, for the background component described by the dictionary tensor and the corresponding coefficient tensor, tensor nuclear norm (TNN) constraint and sparsity constraint are imposed on the coefficient tensor simultaneously to capture the global and local spatial-spectral structure information of the hyperspectral image (HSI), respectively. For the anomalous component, we design a sparse saliency prior weight tensor to enhance the saliency of anomalous targets. Meanwhile, the tensor ℓ F,1 -norm is also integrated into the model to better separate abnormal targets from the background. Furthermore, combining tensor robust principal component analysis (TRPCA) and skinny tensor singular value decomposition (skinny t-SVD), a robust background dictionary is constructed. Finally, an efficient iterative algorithm based on the alternating direction method of multipliers (ADMM) is derived to optimize the RTLSR-SP model. Comprehensive experimental findings on one simulated dataset and six real hyperspectral datasets demonstrate the effectiveness and superiority of the proposed algorithm compared with eight state-of-the-art HAD algorithms.
6 November 2023 in IEEE Geoscience and Remote Sensing Letters
Nowadays, some tensor-based hyperspectral anomaly detection (HAD) approaches are still insufficient in utilizing the spatial-spectral structure information of hyperspectral images (HSIs), resulting in the inability to isolate the background and abnormal targets well. In this letter, an enhanced tensor low-rank representation learning (ETLR) model is proposed for HAD. Specifically, the original 3-D hyperspectral image (HSI) data is firstly decomposed into a structural background component, an anomaly component and a noise component. Among them, with the help of multi-subspace learning technology, the structural background component is reformulated by the t-product of the background dictionary tensor and the corresponding coefficient tensor. Then, tensor nuclear norm (TNN) is adopted to preserve the global low-rank property of the background component in both spatial and spectral dimensions. For the abnormal component, an ℓ 2,1,1 -norm is designed to enhance the group sparsity of abnormal pixels. For the noise component, a tensor F -norm constraint is imposed to suppress the confusion of noise and anomalies. Meanwhile, a robust dictionary tensor that can adequately characterize the background is constructed by using tensor robust principal component analysis (TRPCA). Furthermore, to reduce the interference of redundant information on detection accuracy, the optimal clustering framework (OCF) method is utilized for band selection. Finally, extensive experiments on one simulated and three real HSI datasets confirm that our algorithm is superior than current HAD algorithms.
15 April 2024 in IEEE Transactions on Geoscience and Remote Sensing
In recent years, tensor representation (TR)-based hyperspectral anomaly detection (HAD) approaches have attracted more and more attention. However, two urgent issues still need to be addressed: 1) existing tensor decomposition approaches for HAD cannot make full use of the spectral–spatial correlation of background components in hyperspectral images (HSIs) and 2) most approaches based on TR overlook the piecewise-smooth of background components that exist simultaneously in the spectral and spatial domains. To this end, with the aid of an advanced multiscale entanglement renormalization ansatz (MERA) tensor network, this article proposes an algorithm based on MERA decomposition and enhanced total variation regularization (MERAETV) for HAD. Specifically, MERA decomposes the background tensor by contracting a top-level factor with the remaining semiorthogonal and orthogonal factors. Due to the intricate interplay between semiorthogonal (low-rank) and orthogonal factors, low-rank MERA approximation exhibits a robust representational capacity that effectively captures the spectral–spatial correlation of the background component. Meanwhile, an enhanced total variation (ETV) regularization is devised to capture the inherent piecewise-smooth of the background component in both spectral and spatial domains. Furthermore, our algorithm incorporates group sparsity constraint and Gaussian noise term to enhance the discrimination between anomalies and background. Finally, a highly efficient update scheme based on the alternating direction method of multipliers (ADMM) is designed. A large number of experiments on one synthetic and seven real HSIs demonstrate the superiority of our proposed approach.
11 October 2024 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
In recent years, tensor representation (TR) based approaches have been widely studied in hyperspectral anomaly detection (HAD). However, these methods still suffer from two key issues. Firstly, the various complex regularizations imposed on the background components increase the cost of selecting the best regularized parameters and fail to maximize the effectiveness between these prior regularizations. Secondly, most of them tend to utilize multiple prior knowledge to describe background components, but show obvious deficiencies in mining prior information of abnormal components. To address these two problems simultaneously, we propose an enhanced low-rank and smoothness fusion regularization plus saliency prior (ELRSF-SP) approach. To be specific, for the first problem, we design a weighted tensor-correlated total variation (wt-CTV) to simultaneously characterize the low-rank and smoothness properties of the background tensor. The wt-CTV avoids an additional regularization parameter to balance the two prior regularizations and fully considers the prior distribution information of the singular values of the gradient tensor, thereby improving the ability and flexibility to cope with practical problems. For the second problem, we construct a saliency weight tensor as a constraint of the anomaly tensor to improve the contrast between abnormal pixels and the background. Meanwhile, the tensor $\ell _{1}$ -norm is introduced in ELRSF-SP to characterize the sparsity of the anomaly tensor. Finally, for the optimization of ELRSF-SP, an effective algorithm based on the alternating direction method of multipliers (ADMM) is derived. Extensive experiments demonstrate the effectiveness of the ELRSF-SP approach.
27 October 2025 in IEEE Geoscience and Remote Sensing Letters
As a low-rank regularization technique, the tensor nuclear norm (TNN) has been extensively employed for hyperspectral anomaly detection (HAD). However, traditional TNN and its variants often suffer from rank estimation bias and are difficult to effectively capture the spectral-spatial correlation of complex backgrounds. In addition, the extracted sparse component is often submerged in the background, resulting in abnormal targets being insignificant. To overcome these problems, this letter proposes an enhanced low-rank and joint saliency prior (ELRJSP) method for HAD. Specifically, within the framework of tensor singular value decomposition (t-SVD), we propose an improved weighted tensor nuclear norm (IWTNN), defined as a weighted combination of the weighted tensor nuclear norm (WTNN) and the weighted nuclear norm of the core matrix. This novel norm effectively integrates information from both the original background tensor and its core tensor, thereby leveraging spectral-spatial structural characteristics more comprehensively and leading to enhanced accuracy in background estimation. Furthermore, in order to accurately detect and highlight abnormal targets, under the constraint of the tensor ℓF,1-norm, a visual saliency sparse weight tensor to constrain the abnormal tensor is designed by combining the visual saliency weight graph and the sparse optimized weight graph and extending them to the tensor domain. Comparative experiments on three real hyperspectral datasets demonstrate that the developed ELRJSP algorithm outperforms several advanced algorithms.
1 January 2026 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Recently, models that utilize tensor low-rank representation (TLRR) for hyperspectral anomaly detection (HAD) have demonstrated significant advantages. Nevertheless such models still have deficiencies in accurately characterizing the low-rank and smooth coupling prior of the background and effectively expressing the group sparse structure of abnormal components. In addition, the extracted abnormal components are often easily disturbed by noise. To comprehensively address the above challenges, this paper proposes a tensor multiple subspace learning (TMSL-GTCTV) model with gamma tensor correlated total variation (GTCTV) for HAD. First, a tensor multiple subspace learning scheme is adopted to model the background as a tensor product of a dictionary tensor and a coefficient tensor to characterize the multiple subspace distribution properties of the complex heterogeneous background. Then, by modeling the non-convex gamma quasi-norm in the gradient domain related to the background, a GTCTV regularization is designed and applied to the coefficient tensor to enhance the expression of the low-rank and smooth coupling prior information of the background, thereby improving the accuracy of background modeling. To distinguish sparse non-anomalous targets, under the constraint of the tensor $\ell _{2,1,1}$-norm, a sparse optimization weight tensor for weighting abnormal tensor is designed. Meanwhile, the structured $\ell _{2,1}$-norm is adopted to model the noise component to enhance the ability to identify real abnormal targets. Furthermore, a robust dictionary tensor is constructed by combining GTCTV regularization and sparse optimization weight tensor and integrating them into the tensor robust principal component analysis (TRPCA) framework. Finally, extensive experimental results on a synthetic and five real HSI datasets validate the superiority of our TMSL-GTCTV.