Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group

Optical neural network using pocket-sized liquid-crystal televisions

Not Accessible

Your library or personal account may give you access

Abstract

The improvements in liquid-crystal-television (LCTV) technology have made it possible to build a miniature optical neural network using high-resolution LCTV's to replace the cathode-ray-tube video monitor.1 Two LCTV's built by thin-film-transistor techniques have been used in an optical neural network for displaying the associative-memory matrix and input pattern, respectively. A compact optical neural network of 8 × 8 = 64 neurons has been synthesized, and the number of neurons in the network can be further increased to 15 × 20 = 300. The size of the optical neural network is only ~65 × 12 × 12 cm3, and the system size can be further reduced to ~20 × 7 × 7 cm3 by employing smaller LCTV's. The processing speed of the neural network can be as high as 3 × 106 interconnections.

© 1990 Optical Society of America

PDF Article
More Like This
Self-Learning Optical Neural Network

Francis T. S. Yu, Taiwei Lu, and Don A. Gregory
MB4 Spatial Light Modulators and Applications (SLM) 1990

Real-time pattern recognition using a computer-generated hologram displayed on a liquid-crystal television

Luc LeClerc, Jean-Jacques Drolet, Yunlong Sheng, and Henri H. Arsenault
ThW2 OSA Annual Meeting (FIO) 1990

Optical quadratic perceptron neural network

Alex V. Huynh, John F. Walkup, and Thomas F. Krile
ThY35 OSA Annual Meeting (FIO) 1990

Select as filters


Select Topics Cancel
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.