Comparative analysis of improved FCM algorithms for the segmentation of retinal blood vessels

The main purpose of identifying and locating the retina vessels is to specify from fundus image various tissues vascular structure, which can be wide or tight. classification in retinal often confronts several challenges, such as low contrast accompanying image, inhomogeneity background lighting, noise. Moreover, fuzzy c-means (FCM) one most frequently used algorithms for medical segmentation due its effectiveness. Hence, many FCM method derivatives have been developed improve their noise robustness time-consuming. This paper aims analyze performance some improved recommend best ones blood vessels. Eight algorithm are detained this study: FCM, EnFCM, SFCM, FGFCM, FRFCM, DSFCM_N, FCM_SICM SSFCA. analysis conducted three viewpoints: robustness, performance, At first, clustering evaluated using a synthetic degraded by types levels Then, ability selected segment assessed based on images DRIVE STARE databases after pre-processing phase. Finally, time consumption each measured. experiments demonstrate that FRFCM DSFCM_N achieve better results terms segmentation. Regarding running time, requires less than other images. study extensively discussed, suggestions proposed at end paper.