Improved Graph-Based Image Segmentation Based on Mean Shift

Jian-wen Mo; Chao-xuan Wang; Tong Zhang; Hua Yuan

8th International Conference on Communications and Networking in China

Research Article

Improved Graph-Based Image Segmentation Based on Mean Shift

Cite: BibTeX Plain Text

@INPROCEEDINGS{10.1109/ChinaCom.2013.6694680,
    author={Jian-wen Mo and Chao-xuan Wang and Tong Zhang and Hua Yuan},
    title={Improved Graph-Based Image Segmentation Based on Mean Shift},
    proceedings={8th International Conference on Communications and Networking in China},
    publisher={IEEE},
    proceedings_a={CHINACOM},
    year={2013},
    month={11},
    keywords={image segmentation; mean shift; l*a*b* color space},
    doi={10.1109/ChinaCom.2013.6694680}
}

Jian-wen Mo
Chao-xuan Wang
Tong Zhang
Hua Yuan
Year: 2013
Improved Graph-Based Image Segmentation Based on Mean Shift
CHINACOM
IEEE
DOI: 10.1109/ChinaCom.2013.6694680

Jian-wen Mo¹^,*, Chao-xuan Wang¹, Tong Zhang¹, Hua Yuan¹

1: Guilin University of Electronic Technology

*Contact email: mo_jianwen@126.com

Abstract

According to the problem of over-segmentation using graph-based image segmentation(GBIS), combined with Mean Shift, the paper proposed an improved graph-based image segmentation algorithm(IGBIS) using Lab* color space. Firstly, the source image was smoothed by mean shift; Second, the mean shift image was converted from RGB color space to Lab* color space; Then, the weight function in GBIS method was improved by a constant S that we introduced to control the color difference degree; Finally, two kinds of Algorithm evaluation method (GCE and VOI) were used to evaluate the improved method. The experiment results show that the improved method depressed over-segmentation, received better segmentation results.

Keywords: image segmentation; mean shift; l*a*b* color space

Published: 2013-11-14
Publisher: IEEE

: http://dx.doi.org/10.1109/ChinaCom.2013.6694680