Elliptically Contoured Distributions for Anomalous Change Detection in Hyperspectral Imagery

• Published in: IEEE geoscience and remote sensing letters Vol. 7; no. 2; pp. 271 - 275
• Main Authors: Theiler, James, Scovel, Clint, Wohlberg, Brendt, Foy, Bernard R
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 01-04-2010; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Adaptive signal detection; Algorithms; Change detection algorithms; Computer simulation; Contouring; covariance matrices; Covariance matrix; data-model ellipsoids; Detectors; Ellipsoids; Gaussian; Gaussian distributions; Hyperspectral imaging; Hyperspectral sensors; Image analysis; Layout; Mathematical models; pattern recognition; Remote sensing; Signal detection; Tail
• ISSN: 1545-598X; 1558-0571
• DOI: 10.1109/LGRS.2009.2032565

We derive a class of algorithms for detecting anomalous changes in hyperspectral image pairs by modeling the data with elliptically contoured (EC) distributions. These algorithms are generalizations of well-known detectors that are obtained when the EC function is Gaussian. The performance of these EC-based anomalous change detectors is assessed on real data using both real and simulated changes. In these experiments, the EC-based detectors substantially outperform their Gaussian counterparts.