Automatic seeded region growing for color image segmentation

In this paper, we present an automatic seeded region growing algorithm for color image segmentation. First, the input RGB color image is transformed into YCbCr color space. Second, the initial seeds are automatically selected. Third, the color image is segmented into regions where each region corresponds to a seed. Finally, region-merging is used to merge similar or small regions. Experimental results show that our algorithm can produce good results as favorably compared to some existing algorithms. (c) 2005 Elsevier B.V. All rights reserved.