Determination of compatibility coefficients for color edge detection by relaxation

In this paper, we propose a new algorithm which detects boundaries of objects from a color image. The result of this method is a binary image where only the boundary pixels are represented. This approach uses a co-operating relaxation process in order to enhance edge probabilities. The relaxation algorithm processes probabilities resulting from a gradient output of the different features of the color images. So, this algorithm is able to detect edges from variations of red, green and blue images. We present in the first part of this paper the relaxation scheme. In the second part, we propose two kinds of compatibility coefficients evaluations. These coefficients are compared in the third part of this paper using a rather complex image