SUMATRA: Supervised Modeling of ATR Algorithms

Automated and assisted target recognition (ATR) is an important aspect of an Intelligence Surveillance and Reconnaissance (ISR) system. The role of ATR systems is to detect, classify, recognize and identify targets based on input sensor data from a variety of sources. ATR system performance varies with the quality of the sensor data, the difficulty of the environment being assessed, and the separability of the target signatures in some feature space, both from each other as well as potentially confounding objects. Performance prediction and modeling of operator in the loop ATR systems can be very valuable to judge the efficacy of such ATR systems. Our work on the SUMATRA system (Supervised Modeling of ATR Algorithms) relates image variability (such as target feature visibility, image resolution, time of day effects) and operator in loop demands (operator dwell time, operator work load and training, event rate etc.) to predict false alarm rates. This paper presents our results on operator in the loop ATR performance. A Graphical User Interface is developed as a simulation aid.