11 November 1991 Optical interconnect and neural network applications based on a simplified holographic N4 recording technique
Author Affiliations +
Abstract
This paper presents a large-scale static holographic optical neural network (HONN) based on volume holographic materials. A 1024 neuron HONN has been experimentally demonstrated for pattern recognition operations. It is possible to build a compact size and cost-effective HONN of 64,000 neurons which is capable of storing up to 20,000 training patterns and performing parallel processing in high speed (10 exp 13 interconnections per second).
© (1991) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Freddie Shing-Hong Lin, Freddie Shing-Hong Lin, Thomas Taiwei Lu, Thomas Taiwei Lu, Andrew A. Kostrzewski, Andrew A. Kostrzewski, Hung Chou, Hung Chou, } "Optical interconnect and neural network applications based on a simplified holographic N4 recording technique", Proc. SPIE 1558, Wave Propagation and Scattering in Varied Media II, (11 November 1991); doi: 10.1117/12.49646; https://doi.org/10.1117/12.49646
PROCEEDINGS
8 PAGES


SHARE
RELATED CONTENT

Holographic generation of non-diffractive beams
Proceedings of SPIE (November 10 2014)
Multilayer optical neural networks
Proceedings of SPIE (February 01 1993)
Adaptive holographic implementation of a neural network
Proceedings of SPIE (June 30 1990)
Holographic associative memory with bipolar features
Proceedings of SPIE (November 10 1991)

Back to Top