Presentation 2002/1/17
Edge Detection and Texture Segmentation based on Independent Component Analysis
Yen-Wei CHEN, Xian-Yan ZENG,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this paper, we propose a new edge detection and texture segmentation technique based on independent component analysis (ICA). We apply ICA to natural scene patches to learn the basis functions and then these basis functions are used as pattern templates for feature detections. The gray-scale image is transformed into a pattern map (feature map) in which edge, and background pixels are classified into different class by pattern matching. Compared with conventional mathmatics-based templates such as Haar transform and Gabor functions, the proposed method is based on the statistics of the image. High accuracies of edge detection and texture segmentation have been achieved. In addition to high accuracies, the proposed method is free of the influence of illumination and easy to calculation.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) independent component analysis / template matching / basis function / edge detection / texture segmentation
Paper # 2001-HIP-89
Date of Issue

Conference Information
Committee HIP
Conference Date 2002/1/17(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 Human Information Processing (HIP)
Language ENG
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Edge Detection and Texture Segmentation based on Independent Component Analysis
Sub Title (in English)
Keyword(1) independent component analysis
Keyword(2) template matching
Keyword(3) basis function
Keyword(4) edge detection
Keyword(5) texture segmentation
1st Author's Name Yen-Wei CHEN
1st Author's Affiliation Faculty of Engineering, University of the Ryukyus()
2nd Author's Name Xian-Yan ZENG
2nd Author's Affiliation Faculty of Engineering, University of the Ryukyus
Date 2002/1/17
Paper # 2001-HIP-89
Volume (vol) vol.101
Number (no) 594
Page pp.pp.-
#Pages 5
Date of Issue