Share Email Print

Proceedings Paper

New cost function for backpropagation neural networks with application to SAR imagery classification
Author(s): Hossam M. Osman; Steven D. Blostein
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

This paper proposes the minimization of a new cost function while training backpropagation (BP) neural networks to solve pattern classification problems. The new cost function is referred to as the gain-weighted normalized-target mean-square error (GWNTMSE). The paper proves that the minimization of the GWNTMSE is optimal in the sense of yielding network classifier with minimum variance from the optimal Bayes classifier in the limit of an asymptotically large number of statistically independent training patterns. Experimental results are presented. The application selected is the classification of ship targets in airborne synthetic aperture radar (SAR) imagery. The number of ship classes is 8. They represent 2 destroyers, 2 cruisers, 2 aircraft carries, a frigate, and a support ship. The obtained results indicate that BP classifiers trained by minimizing the GWNTMSE consistently outperform those trained by minimizing the standard MSE.

Paper Details

Date Published: 24 August 1999
PDF: 8 pages
Proc. SPIE 3718, Automatic Target Recognition IX, (24 August 1999); doi: 10.1117/12.359941
Show Author Affiliations
Hossam M. Osman, Queen's Univ. (United States)
Steven D. Blostein, Queen's Univ. (Canada)

Published in SPIE Proceedings Vol. 3718:
Automatic Target Recognition IX
Firooz A. Sadjadi, Editor(s)

© SPIE. Terms of Use
Back to Top