Title :
An application of “neocognitron” neural network for integral chip image processing
Author :
Sadykhov, R. Kh ; Vatkin, M.E.
Author_Institution :
Byelorussian State Univ., Minsk, Byelorussia
Abstract :
The architecture of the “neocognitron” neural network in the task of searching for structural units in a gray scale image of an integrated circuit is considered. The updated rule for activation of the network neurons invariant to distortions of brightness is represented. The comparative outcomes of recognition have shown the advantages of a neural network approach
Keywords :
circuit analysis computing; feature extraction; feedforward neural nets; image recognition; integrated circuits; multilayer perceptrons; neural net architecture; brightness; gray scale image; integral chip image processing; integrated circuit; neocognitron neural network architecture; network neurons; recognition; structural unit searching; updated activation rule; Brightness; Computer vision; Cybernetics; Image analysis; Image processing; Image recognition; Informatics; Multi-layer neural network; Neural networks; Neurons;
Conference_Titel :
Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications, International Workshop on, 2001.
Conference_Location :
Crimea
Print_ISBN :
0-7803-7164-X
DOI :
10.1109/IDAACS.2001.942014
