Using Fuzzy C-means Cluster for Histogram-Based Color Image Segmentation
Zhi-Kai Huang, Yun-Ming Xie, Dehui Liu, Ling-Ying Hou · 2009
In this paper, we proposed a fuzzy c-means (FCM) cluster based adaptive thresholding segmentation algorithm for color image. The main advantage of this method is that, it does not require a priori knowledge about number of objects in the image. It calculates the threshold values automatically with the help of merging process. The first step of the method is that construct the histograms for each color channel. With this aim, information based histogram of the color intensities have been obtained. In the second step of the method, Fuzzy 2-partition is used on each of the three histograms in R(red), G(green) and B(blue) dimensions, color image segmentation is obtained for the performance of the FCM cluster for each color channel. Experiment results show that this method can determine automatically the number of the thresholds levels and achieves good results for color images.