<title>Optical-correlator-based ATR system implementation and evaluation</title>

Abstract

The military has invested significant resources for automatically locating and identifying enemy threats and high value targets. Automatic Target Recognition (ATh) systems using the output of imaging sensors, such as video, and Forward Looking InfraRed (FLIR) are continually being developed. The achievement of adequate processing rates required to analyze the tremendous amounts of data from these sensors remains a critical task in bringing these systems to the field. The two Optical Processing Identification Demonstration (OPID I and OPID II) program efforts ,in their use of an optical processor, address the needs for both advances in system algorithms and data throughput enhancement. OPID I PrQgrnrnLitton developed an ATh system with an optically-based recognition subsystem whose design and development were guided by a number of goals and objectives. The primary objective of this DARPA-funded OPID program was to demonstrate the capability of an ATR system that incorporated both digital and optical processors and explored the benefits obtainable using this architecture. The focus of the effort was to evaluate the optical correlator's ability to discriminate between military vehicles and clutter, and its ability to discriminate between different vehicles, even in the presence of significant target signature distortions.The C2NVEO Terrain Board facility provided the environment for the testing of the system. The facility supports simulations of a target acquisition mission for an airborne or ground platform with a FUR sensor. Specially painted 1/400 scale models of military vehicles in a simulated terrain background provided the targets for the OPID I test. The key variables of the test were terrain clutter level, range to targets and target thermal signature. The OPID I system was evaluated at target ranges between 1500 and 4500 meters. Five different grey shades varying from white to black were painted on the targets. Up to two of these shades could be applied to a single target. Three clutter levels were used to represent mountainous, desert and high vegetation environments. Targets could appear at any aspect perspective but were not allowed variation in look-down angle or in-plane rotation. Each scene contained several of three types of targets, M60 tanks, M35 trucks, and Ml 13 APCs.The OPID I ATR system consisted of two main subsystems, the image preprocessor (a Sun with a slave Datacube image processing subsystem) that found targets and generated boundaries for those targets, and a recognition subsystem (a second Sun with correlator support electronics and the 50