An Improved FCM Algorithm Incorporating Spatial Information for Image Segmentation

Fuzzy c-means (FCM) clustering algorithm is a popular model widely used in segmentation of magnetic imaging (MRI) data. The conventional FCM does not take into account the spatial information of image and get the unexpected results of segmentation when dealing with some MRI contaminated by noise. Considering the intensities of ideal MRI are piecewise constant, we present an improved model to fuzzy c-means algorithm using membership smoothing constraint. The proposed algorithm can reasonably use the spatial information of image and improve the accuracy of segmentation. Simulation MR brain image with different noise levels and real MR brain image are presented in the experiments. The results of experiments show better robustness of our algorithms to noise than other segmentation algorithms.