Presentation 1999/12/17
Automatic Design of Image Emphasis Filter with Multi-layer Neural Networks
Shinya Miyazaki, Junichi Hasegawa,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper proposes a method of designing an optimum linear filter that emphasize the region to be extracted in a grayscale image by using multi-layer neural networks. By using a binary image that indicates the emphasized region, a set of gray levels of local fields in the original image and the corresponding pixel values in the binary image are given as input patterns and teacher patterns, respectively, in the learning of the network. It makes the values of the link weights of the network converge to the optimum weight function of the filter. Hints on how to give the learning data and comparisons with the ordinal way of a combination of typical linear filters are also discussed.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) image emphasis / multi-layer neural network / linear filter / automatic design / sanlple figure.
Paper # PRMU99-178
Date of Issue

Conference Information
Committee PRMU
Conference Date 1999/12/17(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 Pattern Recognition and Media Understanding (PRMU)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Automatic Design of Image Emphasis Filter with Multi-layer Neural Networks
Sub Title (in English)
Keyword(1) image emphasis
Keyword(2) multi-layer neural network
Keyword(3) linear filter
Keyword(4) automatic design
Keyword(5) sanlple figure.
1st Author's Name Shinya Miyazaki
1st Author's Affiliation School of Computer and Cognitive Sciences, Chukyo University()
2nd Author's Name Junichi Hasegawa
2nd Author's Affiliation School of Computer and Cognitive Sciences, Chukyo University
Date 1999/12/17
Paper # PRMU99-178
Volume (vol) vol.99
Number (no) 515
Page pp.pp.-
#Pages 6
Date of Issue