Hyperspectral Anomaly Detection Based on Regularized Background Abundance Tensor Decomposition

Abstract

The low spatial resolution of hyperspectral images means that existing mixed pixels rely heavily on spectral information, making it difficult to differentiate between the target of interest and the background. The endmember extraction method is powerful in enhancing spatial structure information for hyperspectral anomaly detection, whereas most approaches are based on matrix representation, which inevitably destroys the spatial or spectral information. In this paper, we treated the hyperspectral image as a third-order tensor and proposed a novel anomaly detection method based on a low-rank linear mixing model of the scene background. The obtained abundance maps possessed more distinctive features than the raw data, which was beneficial for identifying anomalies in the background. Specifically, the low-rank tensor background was approximated as the mode-3 product of an abundance tensor and endmember matrix. Due to the distinctive features of the background’s abundance maps, we characterized them by tensor regularization and imposed low rankness through CP decomposition, smoothness, and sparsity. In addition, we utilized the ℓ1,1,2-norm to characterize the tube-wise sparsity of the anomaly, since it accounted for a small portion of the scene. The experimental results obtained using five real datasets demonstrated the outstanding performance of our proposed method.

FCAE-AD: Full Convolutional Autoencoder Based on Attention Gate for Hyperspectral Anomaly Detection

“…In order to analyze the detection results effectively, this paper adopts the corresponding three-dimensional receiver operating characteristic (3-D ROC) curve, the corresponding two-dimensional ROC (2-D ROC) curve (P D , P F ), 2-D ROC curve (P D , τ), 2-D ROC curve (P F , τ) [34], and some evaluation indicators AUC TD , AUC BS , AUC SNPR , AUC TDBS , AUC OTD for quantitative analysis. Where AUC TD , AUC BS , AUC SNPR , AUC TDBS , AUC OTD are generated by calculating (P D , P F ), (P D , τ), (P F , τ), respectively.…”

CRNN: Collaborative Representation Neural Networks for Hyperspectral Anomaly Detection

“…To utilize the spatial features in the hyperspectral data, an attribute and edge-preserving filtering detector (AED) [25] and other filtering-based methods [26,27] were proposed. Tensor-based methods treat the hyperspectral cube as a third-order tensor and utilize the spectral and spatial information [28]. A prior-based tensor approximation (PTA) method for hyperspectral anomaly detection is proposed in [29], which decomposed the hyperspectral data into a background tensor and an anomaly tensor under a low-rank and a piecewise-smooth prior.…”