In this paper, a new pixon-based method is presented for image segmentation. In the proposed algorithm, bilateral filtering is used as a kernel function to form a pixonal image. Using this filter reduces the noise and smoothes the image slightly. By using this pixon-based method, the image over segmentation could be avoided. Indeed, the bilateral filtering, as a preprocessing step, eliminates the unnecessary details of the image and results in a few numbers of pixons, faster performance and more robustness against unwanted environmental noises. Then, the obtained pixonal image is segmented using the hierarchical clustering method (Fuzzy C-means algorithm). The experimental results show that the proposed pixon-based approach has a reduced computational load and a better accuracy compared to the other existing pixon-based image segmentation techniques.
Signal, Image and Video Processing, 2013, Vol 7, Issue 5, p. 855-863