Share Email Print
cover

Proceedings Paper

Improved back-propagation algorithm applied to target recognition
Author(s): Yi Ge; Zhenhua Li; Li Zhang; Anzhi He; Ai-ming Lu
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

Despite of its slow learning time, back-propagation (BP) is one of the most widely used neural network training algorithms. In this paper, a nonlinear stretch method is presented that modifies the nonlinear activation function of BP algorithm to speed up the convergence. A invariant target recognition system based on BP neural network using this method and moment invariants is studied. Simulative recognitions on aircrafts and vehicles show that the speed of convergence is increased effectively.

Paper Details

Date Published: 31 December 1996
PDF: 4 pages
Proc. SPIE 2866, International Conference on Holography and Optical Information Processing (ICHOIP '96), (31 December 1996); doi: 10.1117/12.263055
Show Author Affiliations
Yi Ge, Nanjing Univ. of Science and Technology (China)
Zhenhua Li, Nanjing Univ. of Science and Technology (China)
Li Zhang, Nanjing Univ. of Science and Technology (China)
Anzhi He, Nanjing Univ. of Science and Technology (China)
Ai-ming Lu, Nanjing Univ. of Science and Technology (China)


Published in SPIE Proceedings Vol. 2866:
International Conference on Holography and Optical Information Processing (ICHOIP '96)
Guoguang Mu; Guofan Jin; Glenn T. Sincerbox, Editor(s)

© SPIE. Terms of Use
Back to Top