Title :
Development of a hyper-parallel optical fuzzy-neural network using images
Author_Institution :
Dept. of Electron., Hiroshima Inst. of Technol., Japan
Abstract :
A hyper-parallel fuzzy-neural network system that uses optical images to infer conclusions is proposed. This system is formed on the basis of the idea of the optical fuzzy inferring system proposed by the author. The hyper-parallel neural network can transform input data applied to a great number of input units into outputs generated from a great number of output units. This paper shows the fundamental optical architectures of the fuzzy-neuro system and the hyper-parallel fuzzy-neural network. Pattern recognition of numerical letters by the fuzzy-neuro system and the calculation of correlation between two-dimensional patterns by the hyper-parallel fuzzy-neural network are also demonstrated
Keywords :
fuzzy neural nets; inference mechanisms; neural net architecture; optical images; optical neural nets; parallel architectures; pattern recognition; fundamental optical architectures; fuzzy-neuro system; hyper-parallel optical fuzzy-neural network; optical fuzzy inferring system; optical images; pattern recognition; two-dimensional patterns; Displays; Fuzzy systems; High speed optical techniques; Lenses; Neural networks; Neurons; Optical computing; Optical fiber networks; Optical filters; Optical network units;
Conference_Titel :
Circuits and Systems, 1995. ISCAS '95., 1995 IEEE International Symposium on
Print_ISBN :
0-7803-2570-2
DOI :
10.1109/ISCAS.1995.520350
