Presentation 2009-02-28
An automatic parameter adaptation for image segmentation using a Pulse Coupled Neural Network
Masato YONEKAWA, Hiroaki KUROKAWA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) The Pulse Coupled Neural Network (PCNN) is a simulation model of neurons' synchronization phenomena in cats visual cortex and an application to the image processing has been studied, recently. The two-dimensional PCNNs are used for the image processing and the parameters in the PCNN are adjusted by trials and errors. However, because of many parameters of PCNN, it is difficult to configure these parameters appropriately. In this study, we propose an parameter adaptation algorithm for PCNN. The proposed algorithm changes parameters according to PCNN's output states and does not require any object image as a teaching data. Simulation results show that the optimal parameters of the PCNN for image segmentation is automatically obtained.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Pulse Coupled Neural Networks / image processing / parameter adaptation
Paper # NLP2008-146
Date of Issue

Conference Information
Committee NLP
Conference Date 2009/2/21(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 Nonlinear Problems (NLP)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) An automatic parameter adaptation for image segmentation using a Pulse Coupled Neural Network
Sub Title (in English)
Keyword(1) Pulse Coupled Neural Networks
Keyword(2) image processing
Keyword(3) parameter adaptation
1st Author's Name Masato YONEKAWA
1st Author's Affiliation Tokyo University of Technology()
2nd Author's Name Hiroaki KUROKAWA
2nd Author's Affiliation Tokyo University of Technology
Date 2009-02-28
Paper # NLP2008-146
Volume (vol) vol.108
Number (no) 442
Page pp.pp.-
#Pages 6
Date of Issue