Share Email Print

Optical Engineering

Multilayer neural network with a fuzzy controlled learning method for optical pattern training
Author(s): Xin Lin; Masahiko Mori
Format Member Price Non-Member Price
PDF $20.00 $25.00

Paper Abstract

A fast learning method for neural networks learning system using fuzzy controlling is proposed. By the fuzzy controlled theory, the learning rate and the nonlinear gain of the output function in the back- propagation algorithm are adjusted based on training error and training time. The effectiveness of this learning method is demonstrated in optical pattern training.

Paper Details

Date Published: 1 October 2000
PDF: 6 pages
Opt. Eng. 39(10) doi: 10.1117/1.1290583
Published in: Optical Engineering Volume 39, Issue 10
Show Author Affiliations
Xin Lin, National Agriculture Research Ctr. (Japan)
Masahiko Mori, Electrotechnical Lab. (Japan)

© SPIE. Terms of Use
Back to Top