Advanced Signature-Matched Hyperspectral Change Detection

The proposed program will develop enhanced hyperspectral change detection (HSI-CD) algorithms based on spectral phenomenology, conduct extensive performance evaluations using real HSI data, and deliver a flexible software tool for demonstrating automated HSI-CD processing. The work will build on Space Computer Corporation’s existing HSI-CD system by making novel improvements to key processing elements: (1) precision registration, (2) adaptive spectral prediction and normalization, and (3) target signature matching and detection filtering. The registration effort will improve multi-look image alignment in complex backgrounds using precise geo-projection with fine-scale terrain elevation data combined with illumination insensitive scene-based registration refinement. Physically motivated spectral normalization procedures that enable pixel-level comparisons between different HSI passes over an area will be developed to augment or replace currently-used empirical transformations, providing more effective compensation of localized image variations introduced by the environment rather than by material changes. Similar methods will be used to obtain precise matching of material signatures to in-scene spectra, enabling the use of highly selective nonlinear spectral filters for discrimination of specific changes. The program aims to show significant improvements in change target sensitivity and false alarm rejection and to position HSI-CD technology for transition to one or more Air Force sensor programs.