Skip to Main Content
In this paper, based on the analysis of the characteristics of magnetic resonance imaging (MRI), a novel fuzzy clustering algorithm for segmentation of brain MR images is presented. This new algorithm is developed by extending the conventional fuzzy clustering algorithm, which can compensate for not only the noise effects but also the intensity inhomogeneities of the MR images. The proposed technique has been compared and analyzed with the classic fuzzy clustering method and an existing adaptive fuzzy clustering method. Experimental results on segmentation of brain MR images can demonstrate that the proposed method is effective.