Presentation 1994/5/19
A Pattern Recognition Neural Network Combining Competitive Classification and Multilayer Difference Networks
Kazuhiro Minamimoto, Kenji Nakayama,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this paper,we propose a pattern recognition neural network combining competitive classification and multilayer difference network.Input patterns are classified into several groups.Each group has a standard patten,that is average of all patterns in the same group.The second step is the multilayer NN.The input of NN is the difference between the input pattern and the standard pattern of each group.We applied this network to recognition of two dimensional figures having a variety of shapes.There is necessity of pre-processing to avoide unessential deviations - shift, expansion,contraction,rotation.Computer simulation demonstrate the effectiveness of proposed method.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Pattern recognition / Competitive classification / Multilayer NN / Feature extaction / Difference input
Paper # NC94-11
Date of Issue

Conference Information
Committee NC
Conference Date 1994/5/19(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) A Pattern Recognition Neural Network Combining Competitive Classification and Multilayer Difference Networks
Sub Title (in English)
Keyword(1) Pattern recognition
Keyword(2) Competitive classification
Keyword(3) Multilayer NN
Keyword(4) Feature extaction
Keyword(5) Difference input
1st Author's Name Kazuhiro Minamimoto
1st Author's Affiliation Depertment of Electrical and Computer engineering,Faculty of Technology,Kanazawa University()
2nd Author's Name Kenji Nakayama
2nd Author's Affiliation Depertment of Electrical and Computer engineering,Faculty of Technology,Kanazawa University
Date 1994/5/19
Paper # NC94-11
Volume (vol) vol.94
Number (no) 40
Page pp.pp.-
#Pages 8
Date of Issue