Presentation 2007-12-13
Graph Cuts by using Local Texture Features of Wavelet Coefficient for Image Segmentation
Keita FUKUDA, Tetsuya TAKIGUCHI, Yasuo ARIKI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper proposes a approach to image segmentation using Graph Cuts based on local texture features of wavelet coefficient. Using Multiresolution Analysis based on Haar Wavelet, low-frequency range (scaling image) is used for n-link and high-frequency range (local texture features) is used for t-link with color histogram. Iterated Graph Cuts, the propose method can segment the region of object for images with a lot of edge, with colors similar to foreground and background, and with heavy texture change. Experimental results showed our method (Err : 1.57%) reduced segmentation error rates by approximately 35.1% compared to previous methods (Err : 2.42%).
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Image Segmentation / Graph Cuts / Multiresolution Analysis / Local Texture Feature
Paper # PRMU2007-138
Date of Issue

Conference Information
Committee PRMU
Conference Date 2007/12/6(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) Graph Cuts by using Local Texture Features of Wavelet Coefficient for Image Segmentation
Sub Title (in English)
Keyword(1) Image Segmentation
Keyword(2) Graph Cuts
Keyword(3) Multiresolution Analysis
Keyword(4) Local Texture Feature
1st Author's Name Keita FUKUDA
1st Author's Affiliation Graduate School of Engineering, Kobe University()
2nd Author's Name Tetsuya TAKIGUCHI
2nd Author's Affiliation Faculty of Engineering, Kobe University
3rd Author's Name Yasuo ARIKI
3rd Author's Affiliation Faculty of Engineering, Kobe University
Date 2007-12-13
Paper # PRMU2007-138
Volume (vol) vol.107
Number (no) 384
Page pp.pp.-
#Pages 6
Date of Issue