Aided and automatic target recognition (Ai/ATR) capability is
a critical technology needed by the military services for modern combat.
However, the current level of performance that is available is largely deficient
compared to the requirements. This is largely due to the difficulty
of acquiring targets in realistic environments but has also been due to
the difficulty in getting new concepts from, for example, the academic
community, due to limitations for distribution of classified data. The difficulty
of the performance required has limited the fulfillment of the promise
that is so anticipated by the war fighter. We review the metrics, imagery
data bases, and sensors associated with Ai/ATR performance and suggest
possible technical approaches that could enable new advancements
in military-relevant performance.
|