Tensor Subspace Learning andFolded-concave Function Regularizationfor Hyperspectral Anomaly Detection

Hyperspectral anomaly detection focuses on identifying and localizing the anomalous targets in remote sensing. The complex scenarios in hyperspectral imagesmake it more difficult to effectively distinguish anomalous objects from background data, especially in noisy environments. Furthermore, matrix factorizationoften unfolds hyperspectral cubic data into two-dimension form, but this causesthe structural knowledge to be lost. To surmount the above disadvantages, weproposes a tensor subspace-based learning strategy with folded-concave regularization for hyperspectral anomaly detection. First, hyperspectral data undergoesinitial preprocessing through band selection and robust tensor principal component analysis to generate a dictionary representing the background. Then,a tensor subspace learning approach aims to factorize hyperspectral data intothe background and anomaly tensors, in which the folded-concave function isleveraged to minimize minor components for denoising. Next, lF,1 norm on tensor is used to extract abnormal information from hyperspectral data. Finally,comprehensive experiments on several real datasets show that the proposed algorithm performs better than the comparative benchmark methods in detection performance.