Fuzzy Thresholding-Based Brain Image Segmentation Using Multi-Threshold Level Set Model

Abstract

Region of interest with reference to medical image is a challenging task. Clustering or grouping data objects can be used to isolate certain area of interest called image segmentation from human brain MRI scans is considered here. Together with the combination of Multilevel Otsu's thresholding and Level set approach, the most widely used fuzzy-based clustering like fuzzy C means (FCM) thresholding are taken into consideration. Here, the proper thresholding is determined using FCM thresholding. This threshold value can also be used to modify the Multilevel Otsu' method's threshold. The level set technique is then used to this segmented output image, yielding a more precise boundary level estimation. To improve brain MRI image segmentation, the proposed FTMLS system integrates the three aforementioned methods.