A novel enhancement-based rapid kernel-induced intuitionistic fuzzy c-means clustering for brain tumor image

Soft clustering techniques are extensively used for segmenting medical images, and in particular, fuzzy c-means (FCM) clustering is employed to cluster the distinctive regions of the medical image. Specifically, a special attention is needed for the segmentation of brain tumor MR images, since it has more uncertainties. To cope with this impreciseness, intuitionistic fuzzy c-means (IFCM) clustering is utilized which improves the accuracy in segmentation. In this framework, a new approach of clustering brain tumor MR image is proposed to segment brain tumor image. Initially, a novel intuitionistic fuzzy generator (IFG) is derived and the input image is enhanced using it to remove uncertainties. Then, kernel distance-based intuitionistic fuzzy c-means clustering is executed for gray-level histogram of the morphologically reconstructed intuitionistic fuzzy image (IFI). Finally, extensive experiment is conducted for the proposed method and other state-of-the-art methods in clustering to show the efficacy of the proposed method.