EGMM: An evidential version of the Gaussian mixture model for clustering

Lianmeng Jiao, Thierry Denoeux, Zhun-ga Liu, Quan Pan

Abstract: Highlights•An evidential Gaussian mixture model is proposed to characterize cluster uncertainty.•Model parameters are estimated by Expectation-Maximization algorithm.•A validity index allowing determination of the number of clusters is provided.•Experiments demonstrate the better performance of the proposed method.•An application to multi-modal brain image segmentation is exploited.