Summary
International Symposium on Nonlinear Theory and Its Applications
2016
Session Number:A2L-D
Session:
Number:A2L-D-1
Letter Reproduction Using a Cellular Neural Network Consisting of Simplified Neurons and Synapses Fabricated by Thin-Film Transistors
Mutsumi Kimura, Ryohei Morita, Sumio Sugisaki, Tokiyoshi Matsuda, Tomoya Kameda, Yasuhiko Nakashima,
pp.-
Publication Date:2016/11/27
Online ISSN:2188-5079
DOI:10.34385/proc.48.A2L-D-1
PDF download (120.1KB)
Summary:
Simplified neurons and synapses fabricated by thin-film transistors (TFTs) are proposed for cellular neural networks. A two-inverter two-switch circuit is used as a neuron, whereas only a transistor is used as a synapse. The neurons and synapses are fabricated by TFTs, which are promising for giant microelectronics. A cellular neural network consists of such neurons and synapses. Particularly in this presentation, it is observed that an alphabet letter is reproduced from a similar pattern by the cellular neural network. This function is available for letter recognition of hand-written letters.