Abstract | ||
---|---|---|
A region-based variational model for color image segmentation is proposed using the chromaticity-brightness decomposition. By this decomposition, we extend the Wasserstein distance based method to color images. The chromaticity term of the proposed functional follows the data term of the color Chan-Vese model with constraint on unit sphere, and the brightness term is formulated by the Wasserstein distance between the computed probability density function in the local windows (e.g. 3 by 3 or 5 by 5 window) and its estimated counterparts in classified regions. Experimental results on synthetic and real color images show that the proposed method performs well for the segmentation of different image regions. |
Year | DOI | Venue |
---|---|---|
2010 | 10.1007/978-3-642-11301-7_31 | MMM |
Keywords | Field | DocType |
chromaticity term,real color image,brightness term,chromaticity-brightness decomposition,color chan-vese model,different image region,color image segmentation,variational color image segmentation,data term,wasserstein distance,color image,probability density function | Computer vision,Earth mover's distance,Scale-space segmentation,Color histogram,Pattern recognition,Computer science,Segmentation,Chromaticity,Color balance,Image segmentation,Artificial intelligence,Brightness | Conference |
Volume | ISSN | ISBN |
5916 | 0302-9743 | 3-642-11300-1 |
Citations | PageRank | References |
5 | 0.44 | 11 |
Authors | ||
4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Zheng Bao | 1 | 5 | 0.44 |
Yajun Liu | 2 | 9 | 6.27 |
Yaxin Peng | 3 | 73 | 16.82 |
Guixu Zhang | 4 | 128 | 25.80 |