Presentation 2005-02-03
Segmentation of Color Images Using Anisotropic Diffusion and Texture Features with Unsharp Masking
Fumihito OSAKI, Junji MAEDA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) The purpose of the paper is to provide rough segmentation only at the texture regions while maintaining precise segmentation at an overall region in natural color images. We propose to use two features that have different characteristics. One is the smoothed color feature that is obtained by applying anisotropic diffusion in order to remove small textures and edges from an original image. The other is the enhanced texture feature that extracts the statistical geometrical features from an original image after applying unsharp masking. The image segmentation method that combines clustering and region expanding is performed based on the two features and we experimentally show the effectiveness of the proposed method.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Image Segmentation / Anisotropic Diffusion / Unsharp Masking / Statistical Geometrical Features / Clustering
Paper # ITS2004-51,IE2004-185
Date of Issue

Conference Information
Committee IE
Conference Date 2005/1/27(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 Image Engineering (IE)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Segmentation of Color Images Using Anisotropic Diffusion and Texture Features with Unsharp Masking
Sub Title (in English)
Keyword(1) Image Segmentation
Keyword(2) Anisotropic Diffusion
Keyword(3) Unsharp Masking
Keyword(4) Statistical Geometrical Features
Keyword(5) Clustering
1st Author's Name Fumihito OSAKI
1st Author's Affiliation Division of Computer Science and Systems Engineering, Graduate School of Engineering, Muroran Institute of Technology()
2nd Author's Name Junji MAEDA
2nd Author's Affiliation Division of Computer Science and Systems Engineering, Graduate School of Engineering, Muroran Institute of Technology
Date 2005-02-03
Paper # ITS2004-51,IE2004-185
Volume (vol) vol.104
Number (no) 648
Page pp.pp.-
#Pages 6
Date of Issue