New approaches for detecting groups of targets

In military operations, targets are often deployed in well defined groups, such as tank divisions. It is possible to exploit this information and develop algorithms to detect such groups of targets for automated target cueing (ATC). These candidate areas would then be revisited by a higher resolution sensor. The paper describes one such ATC algorithm, which significantly reduces the amount of area revisited by an automatic target recognition (ATR) system. In addition, the authors demonstrate that performance can be greatly improved by reconfirming the existence of a valid target group during the higher resolution revisits.<>