Presentation 2010-12-07
The Edge Based Image Segmentation for Extracting Narrow Elongated Area
Takumi UEMURA, Keiichi UCHIMURA, Gou KOUTAKI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) The image segmentation is an important basic process in image analysis. The segmented image will become the input of more advanced processes, and it will affect the accuracy of the results of the advanced processes significantly. The edge detection is the preprocess of some image segmentation methods. However, these methods cannot detect the detail of edge and remains difficult to segment the thin areas because the edge pixels crush the details. We propose the image segmentation method using Boundary code to solve this problem. Thin area will be segmented continuously by using the proposed segmentation method. In the experiment, we will compare the proposed method with the previous methods.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Image segmentation / Edge detection / Road extraction
Paper # ITS2010-28
Date of Issue

Conference Information
Committee ITS
Conference Date 2010/11/29(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 Intelligent Transport Systems Technology (ITS)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) The Edge Based Image Segmentation for Extracting Narrow Elongated Area
Sub Title (in English)
Keyword(1) Image segmentation
Keyword(2) Edge detection
Keyword(3) Road extraction
1st Author's Name Takumi UEMURA
1st Author's Affiliation Graduate School of Science and Technology, Kumamoto University()
2nd Author's Name Keiichi UCHIMURA
2nd Author's Affiliation Graduate School of Science and Technology, Kumamoto University
3rd Author's Name Gou KOUTAKI
3rd Author's Affiliation Graduate School of Science and Technology, Kumamoto University
Date 2010-12-07
Paper # ITS2010-28
Volume (vol) vol.110
Number (no) 327
Page pp.pp.-
#Pages 6
Date of Issue