Supported by the National High-Tech Research and Development Plan of China under Grant Nos.2001AA142140, 2003AA142140 (国家高技术研究发展计划(863)
An algorithm for color image segmentation, based on color and spatial information is proposed in this paper. First, color quantization is performed on an image based on the proposed color coarseness metric, and then an incremental region growing method is exploited to find the spatial connectivity of pixels with similar colors to form the initial segmented regions. Second, the initial regions are hierarchically merged based on the region distance defined by the color and spatial information. A criteria is proposed to decide the termination of the merging process. Finally, the erosion and dilation operators are used to smooth the edges of the segmented regions. The experimental results demonstrate that the color image segmentation results of the proposed approach hold favorable consistency in terms of human perception.