An analysis of hyperspectral imagery data collected during Operation Desert Radiance

Abstract

The utility of hyperspectral imagers for the identification, classification and status of a specific material based on it's spectral characteristics has been demonstrated in the fields of geology, forestry and meteorology. The United States military has an interest in the utility of hyperspectral imagers for a multitude of tactical and strategic purposes. The Hyperspectral MASINT Support to Military Operations Program (HYMSMO) was designed to explore this arena through a series of planned collection operations utilizing the Hyperspectral Digital Imagery Collection Experiment (HYDICE) sensor in addition to other hyperspectral imaging platforms. Operation DESERT RADIANCE explored many of the areas relating to the tactical detection and classification of military targets. Through the use of the Low Probability of Detection and the Principle Components Transformation algorithms contained in the HYDICE Starter Kit and ENVI software package, this thesis shows that the detection of a tactical target by use of it's unique spectral signature is feasible.