Abstract

An optical neural network based on the neocognitron paradigm [ IEEE Trans. Syst. Man Cybern. SMC-13, 826– 834 ( 1983)] is introduced. A novel aspect of the architectural design is shift-invariant multichannel Fourier optical correlation within each processing layer. Multilayer processing is achieved by feeding back the output of the feature correlator iteratively to the input spatial light modulator and by updating the Fourier filters. By training the neural net with characteristic features extracted from the target images, successful pattern recognition with intraclass fault tolerance and interclass discrimination is achieved. A detailed system description is provided. Experimental demonstrations of a two-layer neural network for space-object discrimination is also presented.

© 1993 Optical Society of America

Optical neural network using vector-feature extraction

Yasunori Kuratomi, Akio Takimoto, Koji Akiyama, and Hisahito Ogawa
Appl. Opt. 32(29) 5750-5758 (1993)

Optical implementation and applications of closest-vector selection in neural networks

Stuart A. Collins, Stanley C. Ahalt, Ashok K. Krishnamurthy, and Daniel F. Stewart
Appl. Opt. 32(8) 1297-1303 (1993)

Neural-network classifiers for automatic real-world aerial image recognition

Shlomo Greenberg and Hugo Guterman
Appl. Opt. 35(23) 4598-4609 (1996)

Optical implementation of neural networks for face recognition by the use of nonlinear joint transform correlators

Bahram Javidi, Jian Li, and Qing Tang
Appl. Opt. 34(20) 3950-3962 (1995)

Optical neural networks: an introduction by the feature editors

Kelvin Wagner and Demetri Psaltis
Appl. Opt. 32(8) 1261-1263 (1993)