An edge detection by using self-organization

This paper proposes a self-organized edge detection. In this method, several clusters are yielded and self-organized according to a gray scale level and the location of pixels. In addition, the comparison among these clusters results in estimated edge. However, after self-organization, clusters are classified into some group according to their properties. In this report, the method which represents the detail distribution of each duster is introduced. In addition, by using this method, it is shown that the proposed detection of edges can improve the accuracy in some experiments.