Automatic Target Recognition for Joint STARS Using Flexible Template Matching and Genetic Algorithms
Department of Defense
Agency Tracking Number:
Solicitation Topic Code:
Small Business Information
Scientific Systems Company,
500 West Cummings Park, Suite, 3950, Woburn, MA, 01801
Socially and Economically Disadvantaged:
Dr R. K. Mehra/dr S. Mahm
AbstractThe success of C3I and battle management systems depends critically on their ability to perform automatic target recognition (ATR) and hostile target identification (HTI). The technologies of sensor data fusion and adaptive signal processing including intelligent imaging are key to the success of ATR and HTI. An innovative approach to intelligent imaging and target recognition is proposed using methods of Bayesian classification, Flexible Template Matching (FTM), and Genetic Algorithms (GA). The overall ATR problem under low SNR conditions is formulated as a model-based stochastic maximaization problem. The global maxima are found using a Genetic Algorithm. A Nueral Network (NN) approach is used for detection and preclassification to reduce the number of false alarms. The FTM-GA-NN approach will be applied to real radar data representative of that generated by the current Joint STARS system and/or near-term technologically feasible enhancements to the Joint STARS system to classify targets into tanks, trucks, and helicopters. Norden Systems, Inc., the designer and manufacturer of the Joint STARS radar system as well as several other advanced technology radar sensors, will support Scientific Systems during Phases I and II for R&D and during Phase III for commercialization of the innovative signal processing algorithms.
* information listed above is at the time of submission.