Region Growing Using Color and Spatial Information
NING Yu-dan (Shaoyang Vocational and Technical College, Shaoyang 422000, China)
In this paper, the seed region growing algorithm based on the color of the image and spatial information is put forward and it goes for the color image division.First of all, with Euclidean distance the algorithm uses the relative means clustering algorithm to quantify the color image, which results in the initial image segmentation. Then by calculating the divergence of partial color, the grading region is to be combined.Finally, smooth the edge of the region of the segmentation with the algorithm related to morphology.The experiments show that with the algorithm, the meaningful segmentation region corresponded with human's vision judgment can be achieved.
【CateGory Index】： TP391.41