Presentation 2013-02-18
Consideration of Perceptual Segmentation Algorithm for Natural Color Images
Naoya ISHIKURA, Takehito TANI, Junji MAEDA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Image segmentation is a useful preprocessing in image recognition. The purpose of this paper is to develop the perceptual segmentation algorithm of natural color images. We propose the fuzzy-based split-and-merge algorithm as segmentation method that integrates color features and texture features by using fuzzy inference. In the proposed algorithm, however, there are still the problems remaining in the handling of features and parameters. In this research, we propose an improved method to produce the favorable segmented results by incorporating SLIC and automatic determination of the parameters into the algorithm. To demonstrate the effectiveness of the proposed method, we perform the comparative experiments in comparison with the other methods using the quantitative evaluation with ground truth.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Perceptual segmentation / fuzzy-based split-and-merge algorithm / statistical geometrical features / simple linear iterative clustering / ground truth / PR Index / Dongen Index
Paper # ITS2012-31,IE2012-111
Date of Issue

Conference Information
Committee IE
Conference Date 2013/2/11(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) Consideration of Perceptual Segmentation Algorithm for Natural Color Images
Sub Title (in English)
Keyword(1) Perceptual segmentation
Keyword(2) fuzzy-based split-and-merge algorithm
Keyword(3) statistical geometrical features
Keyword(4) simple linear iterative clustering
Keyword(5) ground truth
Keyword(6) PR Index
Keyword(7) Dongen Index
1st Author's Name Naoya ISHIKURA
1st Author's Affiliation Division of Information and Electronic Engineering, Graduate School of Engineering, Muroran Institute of Technology()
2nd Author's Name Takehito TANI
2nd Author's Affiliation Division of Information and Electronic Engineering, Graduate School of Engineering, Muroran Institute of Technology
3rd Author's Name Junji MAEDA
3rd Author's Affiliation Division of Information and Electronic Engineering, Graduate School of Engineering, Muroran Institute of Technology
Date 2013-02-18
Paper # ITS2012-31,IE2012-111
Volume (vol) vol.112
Number (no) 434
Page pp.pp.-
#Pages 6
Date of Issue