Ensemble fuzzy c-means clustering algorithms based on KL-Divergence for medical image segmentation

Image segmentation plays an important role in medical imaging for clinical purposes. In this paper, an image segmentation method using the ensemble of fuzzy clustering is proposed, in which we classify the pixels in an image according to heterogeneous clustering methods, and then combine the clustering results by a KL-Divergence based fuzzy clustering algorithm to provide the final image segmentation results. Experimental results show that the proposed method performs better than some existing clustering-based methods in medical image segmentation problems.