Invariant Spectral/Spatial Techniques for Target Detection and Material Identification in Cluttered Backgrounds

"In this project we will develop innovative algorithms for anomaly detection,target detection, and atmospheric quantification. The algorithms will combinespectral and spatial information in hyperspectral images and will be applicableto the visible through short-wave infrared and thermal infrared spectral ranges.Targets will be represented using invariant physics-based spectral subspaces thatdescribe target signatures over spatial and temporal variability in the illumination,atmospheric, and thermal environments. Background clutter will be represented usingadaptive spectral subspaces that can represent arbitrary mixtures of several materials.Optimized algorithms will be derived that use the representations to detect andidentify low-contrast targets in unknown changing environments with substantial clutter.Since the algorithms are based on physical models, they can also be used to estimateparameters that describe the atmospheric conditions. The proposed algorithms are basedon compact subspace representations and can be implemented efficiently. The algorithmswill be evaluated using imagery from hyperspectral data collections. The newalgorithms will significantly improve the capability of systems that process airborneand space-based electro-optical imagery."