Automatic Target Recognition

We focus on the development and implementation of a hierarchical multi-class target classification system suitable for dynamically changing conditions in the battle field. New support vector representation and discrimination machine (SVRDM) classifiers are developed. Each classifier is a separate classifier that distinguishes one class from all of the rest. This architecture involves a tree structure of C-1 classifiers rather than a single layer set of C-1 classifiers. Our approach uses only a subset of classifiers in any application in traversing a path to a bottom decision node. Thus, the number of required calculations is reduced and the classification results are greatly improved. The multiple classifiers are further fused via statistical decision theory for robust performance of the overall classification.