Abstract
The use of composite images allows more images to be included in the construction of synthetic discriminant functions without increasing the training-set cardinality. This provides lower computational overhead or allows inclusion of more training information for the same computational load. The procedure is demonstrated with the minimum-average-correlation-energy synthetic discriminant function as an example.
© 1996 Optical Society of America
Full Article | PDF ArticleMore Like This
David Casasent and Wen-Thong Chang
Appl. Opt. 25(14) 2343-2350 (1986)
Mario Montes-Usategui, Juan Campos, and Ignacio Juvells
Appl. Opt. 34(20) 3904-3914 (1995)
Z. Bahri and B. V. K. Vijaya Kumar
J. Opt. Soc. Am. A 5(4) 562-571 (1988)