A novel initialization scheme for the fuzzy c-means algorithm for color clustering
Abstract—Abstract A novel initialization scheme for the fuzzy c-means (FCM) algorithm is proposed for the color clustering problem. Given a set of color points, the proposed initialization scheme extracts the most vivi dandd is tinguishable colors, referred to here as the dominant colors. The color points closest to these dominant colors are selected as the initial centroids in the FCM calculations. To obtain the dominant colors and their closest color points, we introduce reference colors andd efine a fuzzy membership model between a color point anda reference color. The effectiveness and reliability of the proposed methodis demonstrated through various color clustering examples..