Robust estimation of mahalanobis distances in hyperspectral images

Advanced, high-altitude remote sensing platforms are important for surveillance, and their ability to resolve details on the ground is crucial to military operations. With sponsorship from the Air Force Research Laboratory Space Vehicles Directorate, Major Meidunas developed new methods for extracting information from images acquired by these platforms and new ways to accurately model scene statistics essential for properly identifying image content. His new techniques reduce the errors inherent in interpreting captured images and greatly decrease image processing time. His results will lead to faster and more accurate signal processing for airborne and space imaging systems for the Department of Defense.