Optic disc and optic cup segmentation in retinal images

Automated segmentation of the optic disc (OD) and cup (OC) is important for retinal image analysis diabetic retinopathy systems. For OD segmentation, this paper presents a method done in three steps that combines variance brightness features to localise leading increased accuracy detecting rather than using just one feature. As first step, divided into non-over lapping windows. Then brightest window with maximum intensity selected. Subsequently circular Hough transform (CHT) applied get segmentation. OC two steps: 1) blood vessels inside are eliminated and; 2) restricted region growing performed The efficacy proposed demonstrated standard benchmark DRIVE DIARETDB1 databases by comparing results some methods silico.