HYPERSPECTRAL ANOMALY DETECTION WITH AN IMPROVED APPROACH: INTEGRATION OF GO DECOMPOSITION ALGORITHM AND LAPLACIAN MATRIX MODIFIER

Abstract

In this study, a hyperspectral anomaly detection method based on Laplacian matrix (HADLAP) is proposed. This paper addresses the problem of determining covariance matrix inversion in high-dimensional data and proposes a new approach for identifying anomalies in hyperspectral images (HSIs). The study’s goals are to find anomalous locations in HSIs and to deal with the problem of calculating the inversion of the covariance matrix of high dimensional data. The method is centered on two main concepts. The low-rank and the sparse matrices have been extracted first from hyperspectral data. Then, Mahalanobis Distance (MD) is implemented by the image's sparse component. In this study, HSI data is decomposed using go decomposition (GoDec) algorithm that yields low-rank and sparse matrices. The sparse matrix is then subjected to MD, producing an anomaly detection map. A distinctive aspect of the proposed approach is computation of the covariance matrix inversion in MD using the Laplacian matrix, setting it apart from previous studies. The empirical findings present that proposed method performs remarkably well in anomaly identification when compared to state-of-the-art methods on a variety of hyperspectral datasets.