A DCT-based local and non-local fuzzy C-means algorithm for segmentation of brain magnetic resonance images.

•A transform-based local and nonlocal fuzzy C-means (DCT-LNLFCM) for brain MRI segmentation is developed.•Both the local and the nonlocal information are used in the transform domain for image segmentation.•Detailed experimental analysis is performed both on simulated and real MRI.•The DCT-LNLFCM achieves high segmentation accuracy as compared to the state-of-the-art methods.•DCT-LNLFCM provides a good tradeoff between noise insensitivity and preservation of image details.