Abstract: Fuzzy Gmeans (FCM) clustering is one of well-known unsupervised clustering techniques, which has been widely used in automated image segmentation. However, when the classical FCM algorithm is used for image segmentalion, there are also some problems, such as setting the initial number of clusters in advance, not considering the effects of various image features. An improved adaptive FCM image segmentation algorithm based on the ReliefF was proposed, which can accomplish image segmentation by considering the effects of various image features, incorporating the cluster validity exponent to ascertain the initial number of clusters automatically, extracting the image feature according to Laws texture measure. Experimental results show that the proposed method is simple and work well for most images, and has better segmentation effect than the existing FCM image segmentation.

Key words: Image segmentation,Fuzzy c-means clustering,ReliefF,Cluster validity