Translator Disclaimer
25 January 2011 Color image segmentation on region growing and multi-scale clustering
Author Affiliations +
This paper presents a color image segmentation method by combining region growing and color clustering algorithms. This method considers the both color and location information in a transformed color space. After multi-scale clustering (MSC), it does a spatial processing - region growing. MSC can perform better in conquering the over-segment problem than equal distance clustering. Compared with the previous methods only depending on MSC, the region growing can enhance the ability of noise suppression. This method inherits the idea that operates clustering first and then carries out spatial processing. Both clustering algorithm and spatial processing algorithm are improved, so this method (the two algorithms) can obtain more satisfactory results.
© (2011) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Zong-pu Jia, Wei-xing Wang, Jun-ding Sun, and Tai-wen Wei "Color image segmentation on region growing and multi-scale clustering", Proc. SPIE 7866, Color Imaging XVI: Displaying, Processing, Hardcopy, and Applications, 78661N (25 January 2011);


Back to Top