Share Email Print

Proceedings Paper

Optical implementation of a shift-invariant neocognitron
Author(s): Tien-Hsin Chao; William W. Stoner; William J. Miceli
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

An optical neural network based upon the Neocognitron paradigm is introduced. A novel aspect of the architectural design is shift-invariant multichannel Fourier optical correlation within each processing layer. An innovative bipolar neural weights holographic synthesis technique is introduced to implement both the excitatory and inhibitory neural functions. Multilayer processing is achieved by iteratively feeding back the output of the feature correlator to the input spatial light modulator and updating the Fourier filters. By designing the neural net with characteristic features extracted from the target images, successful pattern recognition with intra-class fault tolerance and inter-class discrimination is achieved. A detailed system description is provided. Experimental demonstrations of a two-layer neural network for space objects discrimination is also presented.

Paper Details

Date Published: 2 February 1993
Proc. SPIE 1773, Photonics for Computers, Neural Networks, and Memories, (2 February 1993); doi: 10.1117/12.983186
Show Author Affiliations
Tien-Hsin Chao, Jet Propulsion Lab. (United States)
William W. Stoner, Science Applications International Corp. (United States)
William J. Miceli, Office of Naval Research (United States)

Published in SPIE Proceedings Vol. 1773:
Photonics for Computers, Neural Networks, and Memories
Stephen T. Kowel; John A. Neff; William J. Miceli, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?