Abstract

Traditionally, correlation-based target-recognition systems are evaluated by observation of the number of images successfully recognized and the errors made in the classification process. However, the rigorous use of hypothesis tests and confidence intervals in these evaluations does not appear with any regularity in the optics literature. The optical target-recognition community should adopt these standard methods to evaluate objectively the statistical performance of automatic target recognition systems. We review the necessary steps for making statistically significant inferences about the performance of automatic target recognition systems and show their application by reevaluating some previous results as case studies.

© 1994 Optical Society of America

Sparsity-motivated automatic target recognition

Vishal M. Patel, Nasser M. Nasrabadi, and Rama Chellappa
Appl. Opt. 50(10) 1425-1433 (2011)

Integrated approach for automatic target recognition using a network of collaborative sensors

Abhijit Mahalanobis and Alan Van Nevel
Appl. Opt. 45(28) 7365-7374 (2006)

Comparison of automatic target recognition system performance with full- and reduced-resolution correlators

Paul C. Miller
Appl. Opt. 38(23) 5014-5018 (1999)

Automatic target recognition employing signal compression

Pradeep Ragothaman, Wasfy B. Mikhael, Robert R. Muise, and Abhijit Mahalanobis
Appl. Opt. 46(21) 4702-4711 (2007)

Demonstration of improved automatic target-recognition performance by moment analysis of correlation peaks

Paul C. Miller and Robert S. Caprari
Appl. Opt. 38(8) 1325-1331 (1999)