Abstract:SIAM Journal on Imaging Sciences, Volume 17, Issue 2, Page 917-950, June 2024. Anomaly detection in the hyperspectral images, which aims to separate interesting sparse anomalies from backgrounds, is a significant topic in remote sensing. In this paper, we propose a generalized nonconvex background representation learning with dictionary constraint (GNBRL) model for hyperspectral anomaly detection. Unlike existing methods that use a specific nonconvex function for a low rank term, GNBRL uses a class of nonconvex functions for both low rank and sparse terms simultaneously, which can better capture the low rank structure of the background and the sparsity of the anomaly. In addition, GNBRL simultaneously learns the dictionary and anomaly tensor in a unified framework by imposing a three-dimensional correlated total variation constraint on the dictionary tensor to enhance the quality of representation. An extrapolated linearized alternating direction method of multipliers (ELADMM) algorithm is then developed to solve the proposed GNBRL model. Finally, a novel coarse to fine two-stage framework is proposed to enhance the GNBRL model by exploiting the nonlocal similarity of the hyperspectral data. Theoretically, we establish an error bound for the GNBRL model and show that this error bound can be superior to those of similar models based on Tucker rank. We prove that the sequence generated by the proposed ELADMM algorithm converges to a Karush–Kuhn–Tucker point of the GNBRL model. This is a challenging task due to the nonconvexity of the objective function. Experiments on hyperspectral image datasets demonstrate that our proposed method outperforms several state-of-the-art methods in terms of detection accuracy.

Coincidence of the Rao Test, Wald Test and GLRT for Anomaly Detection in Hyperspectral Imagery.

Training Data Assisted Anomaly Detection Of Multi-Pixel Targets In Hyperspectral Imagery

Hyperspectral Anomaly Detection Via Locally Enhanced Low-Rank Prior

Anomaly Detection with Training Data in Hyperspectral Imagery

Coincidence of the Rao Test, Wald Test, and GLRT of MIMO Radar in Gaussian Clutter

Anomaly detection algorithm based on FrFT transform and total variation regularization

Exploring the Intrinsic Probability Distribution for Hyperspectral Anomaly Detection

Multipixel Anomaly Detection With Unknown Patterns for Hyperspectral Imagery

Hyperspectral Anomaly Detection Based on Low-Rank Representation Using Local Outlier Factor

Normalizing Flow-Based Probability Distribution Representation Detector for Hyperspectral Anomaly Detection.

Hyperspectral Anomaly Detection via Optimal Kernel and High-Order Moment Correlation Representation

Anomaly Detection-Oriented Band Selection For Hyperspectral Image

Tensor Approximation With Low-Rank Representation and Kurtosis Correlation Constraint for Hyperspectral Anomaly Detection

Adaptive Robust Subspace Detection Based on GLRT, Rao, Wald, Gradient, and Durbin Tests

A Novel Hyperspectral Remote Sensing Images Classification Using Gaussian Processes with Conditional Random Fields

Modified Rao Test for Adaptive Subspace Detection

Enhanced Tensor Low-Rank Representation Learning for Hyperspectral Anomaly Detection

Generalized Nonconvex Hyperspectral Anomaly Detection via Background Representation Learning with Dictionary Constraint

Low-Rank and Sparse Decomposition With Mixture of Gaussian for Hyperspectral Anomaly Detection

Anomaly Detection Based on High-order Statistics in Hyperspectral Imagery

Unsupervised Anomaly and Change Detection with Multivariate Gaussianization