Target Image Transformation and Transfer

Frontier Technology, Inc. (FTI) and University of Florida (UF) propose to develop and analyze designs and prototype software for a system that will resize, transform, compress, and transfer target imagery and associated reference points from a Fire Control System to an autonomous small caliber projectile for storage. The proposed POSIT system would assist the LFLSP (future SSMSA) onboard target classifier in the recognition/identification of features or behaviors of targets during movement among defilade positions. We propose to analyze and enhance POSIT performance to support: (a) Target image selection by command from a fire control system at the launch point, followed by transformation and resizing to appear as to the projectile's imaging system at turn-on, (b) Image transmission and storage within the projectile's recognition and tracking system to enable target identification using conventional points of reference in the field of view. Phase-I will evaluate, extend and exploit FTI/UF's successful, NVESD-sponsored R&D for dynamic automated target recognition from spinning projectiles. Analysis will emphasize space/time complexity, cost, space/power consumption, and effect on the accuracy of the current LFLSP pattern recognition system. Phase-II will develop and test prototype hardware and software.