A fuzzy clustering algorithm with local contextual information and Gaussian function for simultaneous brain MR image segmentation and intensity inhomogeneity estimation

In this paper, we present a fuzzy clustering algorithm by integrating local contextual information and a Gaussian function for simultaneous brain MR image segmentation and intensity inhomogeneity estimation. For each pixel, a local contextual information is integrated due to highly correlation between the image pixels and used to define its fuzzy membership function to belong into a tissue type. Whereas, a Gaussian surface is fitted over each tissue region using the local image gradients to estimate the intensity inhomogeneity (IIH). In doing so, we have introduced global and local membership functions for each pixel. The combined IIH is iteratively removed from the image and the final segmentation result is obtained based on the global membership values. The simulation results on brain MR images show its superiority over other fuzzy-based clustering algorithms.