Images segmentation based on interval type-2 Fuzzy C-Means

Segmentation process helps to find region of interest in a particular image. The main goal is to make image more simple and meaningful. This work is an improvement of an existing method which is Fuzzy C-Means (FCM) to partitioning an image into several constituent components - type 2 Fuzzy C-Means-. First, membership function defined by Hamid R Tizhoosh is used to measure the image fuzziness. Second, new membership functions are proposed. The evaluation of adopted approaches was compared using the validity functions: Partition Coefficient Vpc, Partition Entropy Vpe and Peak Signal and Noise Ratio PSNR. The experimental results on real images prove that the proposed approaches are more accurate and robust than the standard FCM approach.