Presentation 2008-03-13
Image Recognition System with Parallel-Hierarchical Neural Network Based on Visual Information Processing
Teruo TERAJIMA, Masafumi HAGIWARA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this paper, we propose a new image recognition system based on visual information processing. The proposed system is composed of the network based on Neocognitron that mimics human vision system. The proposed system considers color information employing the framework of Neocognitron, and recognition that considers both color information and shape information is carried out. In addition, the proposed system can automatically decide the number of cell planes that heavily affects the recognition performance. It is confirmed that the proposed system has excellent performance from experiments, for example, the recognition rate for unlearned images is increased by about 15% compared with the conventional Neocognitron.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Image recognition / Neural network / Neocognitron
Paper # NC2007-153
Date of Issue

Conference Information
Committee NC
Conference Date 2008/3/5(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Neurocomputing (NC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Image Recognition System with Parallel-Hierarchical Neural Network Based on Visual Information Processing
Sub Title (in English)
Keyword(1) Image recognition
Keyword(2) Neural network
Keyword(3) Neocognitron
1st Author's Name Teruo TERAJIMA
1st Author's Affiliation Faculty of Science and Technology, Keio University()
2nd Author's Name Masafumi HAGIWARA
2nd Author's Affiliation Faculty of Science and Technology, Keio University
Date 2008-03-13
Paper # NC2007-153
Volume (vol) vol.107
Number (no) 542
Page pp.pp.-
#Pages 6
Date of Issue