Global Unsupervised Anomaly Extraction And Discrimination In Hyperspectral Images Via Maximum-Orthogonal Complement Analysis
Year: 2008
Type: article
Abstract: In this paper we address the problem of global unsuper-vised detection, discrimination, and population estimation of anomalies in hyperspectral images. The proposed ap-proach, denoted as Anomaly Extraction and Discrimina-tion Algorithm (AXDA), detects anomalies via analysis of a signal-subspace obtained by the recently developed Maxi-mum Orthogonal-Complement Algorithm (MOCA). MOCA is unique in providing an unsupervised combined estimation of signal-subspace that includes anomalies, and its rank. The main idea of AXDA is to iteratively reduce the anomaly vec-tor subspace-rank, making the related anomalies to be poorly represented. This helps to detect them by a statistical anal-ysis of the ℓ2,∞-norm of data residuals. As a by-product, AXDA provides also an anomaly-free robust background sub-space and rank estimation. 1.
DOI: 10.5281/zenodo.40865
Authors: Barzohar, Meir, Kuybeda, Oleg, Malah, D.
Institution: Technion – Israel Institute of Technology
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W1526060269