Presentation 1998/5/25
Recognition of particular patterns and each of its position in an image, by an integrated neural network consisting of many unit-networks of identity transformation
Masahiro Ishido, Shuhei Takagi,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) The integrated neural network consists of many unit-networks of five layered neural network of identity transformation, connected two-dimensionally in parallel in order to deal with an image. The unit-network learns some particular patterns and is copied, and then they are connected to construct the integrated network. When an image containing particular patterns and others is given to the input units, only the particular patterns appear on the output image at the same position. A new pattern and its position is recognized by connecting another unit-network in parallel which has learned the new pattern. It is not needed that the previous unit-networks learn it.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Identity transformation / Inner state / Generalization / BP learning
Paper #
Date of Issue

Conference Information
Committee NC
Conference Date 1998/5/25(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) Recognition of particular patterns and each of its position in an image, by an integrated neural network consisting of many unit-networks of identity transformation
Sub Title (in English)
Keyword(1) Identity transformation
Keyword(2) Inner state
Keyword(3) Generalization
Keyword(4) BP learning
1st Author's Name Masahiro Ishido
1st Author's Affiliation Department of Computer and Systems Engineering, Kobe University()
2nd Author's Name Shuhei Takagi
2nd Author's Affiliation Department of Computer and Systems Engineering, Kobe University
Date 1998/5/25
Paper #
Volume (vol) vol.98
Number (no) 77
Page pp.pp.-
#Pages 6
Date of Issue