Presentation 1996/1/19
Extraction of Singular Locations from an Image Using Immune Network
Kazuya KUBO, Toshio ITO, Yukio KANEDA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) We take the distributed diagnostic model as an Immune Network Model and applies it to image processing. The Immune System is composed of units, each of which has several states and can test other units to identify faulty ones among them in the whole system. This model is suitable for parallel processing. At first, we define line patterns extracted by using local Hough Transform as units and introduce several dynamical model into them to extract such blurred locations as blotted spots or noise. Then we compare those locations with contours. Secondly, modeled features boundary, in particular,the boundary of two textures which are modeled as a 2-D Markov Random Fields is extracted.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) immune network / blurred image / contours / boundary extraction / Markov Random Fields
Paper # PRU95-209
Date of Issue

Conference Information
Committee PRU
Conference Date 1996/1/19(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 Pattern Recognition and Understanding (PRU)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Extraction of Singular Locations from an Image Using Immune Network
Sub Title (in English)
Keyword(1) immune network
Keyword(2) blurred image
Keyword(3) contours
Keyword(4) boundary extraction
Keyword(5) Markov Random Fields
1st Author's Name Kazuya KUBO
1st Author's Affiliation The Graduate School of Science and Technology, Kobe University()
2nd Author's Name Toshio ITO
2nd Author's Affiliation Electronics Engineering Div.,Daihatsu Motor Co.,Ltd.
3rd Author's Name Yukio KANEDA
3rd Author's Affiliation Faculty of Engineering, Kobe University
Date 1996/1/19
Paper # PRU95-209
Volume (vol) vol.95
Number (no) 470
Page pp.pp.-
#Pages 6
Date of Issue