As part of its ongoing automated target recognition (ATR) program, the US DoD has sponsored an effort to develop and demonstrate methods for evaluating ATR algorithms that utilize multiple data sources, i.e., fusion-based ATR. The paper presents results from this program, focusing on the human-in-the-loop, i.e. assisted image exploitation. Reliance on ATR technology is essential to the future success of intelligence, surveillance, and reconnaissance (ISR) missions. ATR technology is designed to aid the analyst, but the final decision rests with the human. Traditionally, evaluation of ATR systems has focused mainly on the performance of the algorithm. Assessing the benefits of ATR assistance for the user raises interesting methodological challenges. We review the critical issues associated with evaluations of human-in-the-loop ATR systems and present a methodology for conducting these evaluations. Experimental design issues addressed include training, learning effects, and human factors. The evaluation process becomes increasingly complex when data fusion is introduced. Even in the absence of ATR assistance, the simultaneous exploitation of multiple frames of coregistered imagery is not well understood. We explore how the methodology developed for exploitation of a single source of data can be extended to the fusion setting.
Evaluation of assisted image exploitation with extensions to image fusion
2003
9 Seiten, 11 Quellen
Aufsatz (Konferenz)
Englisch
Tactical reconnaissance image exploitation
Tema Archiv | 1979
|Advances in Aerial Image Exploitation
Online Contents | 2005
|Image Exploitation System for Airborne Surveillance
British Library Conference Proceedings | 2007
|