Authors : R. Shanthi and S. Prabakaran
Abstract: Automatic and accurate localization of Optic Disk and Cup (ODC) in retinal fundus image is a necessary process for diagnosis of glaucoma. Glaucoma is the second leading chronic eye disorder which harms the optic nerve head and causes vision loss which leads to permanent blindness. The glaucoma can be detected by segmenting ODC boundary and calculating the Ratio of optic Disc and Cup (CDR). This study presents an effective framework for automatic segmentation of ODC boundary. The framework begins with median filtering to smoothen the original fundus image, followed by region of interest is marked, next superpixel is generated using SLIC method, finally, Hough transform circle and Otsu’s threshold method is used to detect the optic disc and cup boundary. Drive online public dataset is used to assess the performance of the proposed system. The average precision, recall and F-score values obtained for detecting optic disc are 0.98, 0.97 and 0.95, respectively and for optic cup a precision, recall and F-score values are 0.92, 0.95 and 0.93, respectively. The experiment results achieved is showed that the proposed technique gives better result in segmenting the boundary of optic disc and cup in retinal fundus image.
R. Shanthi and S. Prabakaran, 2019. Optic Disc and Cup Boundary Segmentation for Glaucoma Screening in Fundus Image. Journal of Engineering and Applied Sciences, 14: 8804-8811.