DocumentCode
2571806
Title
An effective approach towards color image segmentation for micro-vessel detection
Author
Juan Chen ; Quan Wen ; Zhifei Pang ; Mete, Mutlu
Author_Institution
Sch. of Comput. Sci. & Eng., Univ. of Electron. Sci. & Technol. of China, Chengdu, China
fYear
2012
fDate
19-21 Oct. 2012
Firstpage
59
Lastpage
63
Abstract
In this paper, we propose an effective approach towards color image segmentation for micro-vessel detection in the virtual slide of double stained liver tissue. The contribution of the proposed method lies in three aspects. First, the dominant colors of white/gray, blue, red and brown are modeled in the RGB color space. Second, the CART (classification and regress tree) method is implemented to segment the modeled colors. Third, the micro-vessel regions are identified in the color images based on the association of both red and brown pixels. Extensive experiments are carried out to validate the performance of the proposed approach. The experimental results demonstrate that our method is quite promising.
Keywords
image colour analysis; image segmentation; regression analysis; RGB color space; color image segmentation; double stained liver tissue; microvessel detection; regress tree method; Color; Decision trees; Humans; Image color analysis; Image segmentation; Liver; Tumors; classification and regress tree; color segmentation; double stain; micro-vessel; virtual slide;
fLanguage
English
Publisher
ieee
Conference_Titel
Computational Problem-Solving (ICCP), 2012 International Conference on
Conference_Location
Leshan
Print_ISBN
978-1-4673-1696-5
Electronic_ISBN
978-1-4673-1695-8
Type
conf
DOI
10.1109/ICCPS.2012.6384283
Filename
6384283
Link To Document