Color Image Segmentation Based on Multi-Valued Eigenfunction

Image segmentation is a key step in image process, and also is a core technology of a vision system in soccer robot. In this paper, a method of color image segmentation based on multi-valued characteristic is presented. A superset of the objective sample is constructed in color space, multi-valued characteristic function is employed to describe objective clustering, ultimately implement the division of color image. Experiments show this algorithm is preferable in both functionality and speed of image processing.