Glaucoma Detection Using Support Vector Machine Algorithm

Glaucoma is an eye disorder that is not curable, so early diagnosis is needed to prevent it from developing further. This condition contributes to complete blindness. Various automated systems have been developed to address the disease. In this paper, we explain the calculation of the cup to disc ratio (CDR) using a fuzzy algorithm. One of the major factors in the diagnosis of glaucoma is the cup to disc ratio (CDR). In the proposed process, using the support vector machine and thresholding, the optic disc and optic cup are segmented. The primary identifier to confirm glaucoma for a patient is the Cup to Disc Ratio (CDR) of the image. This classifies the given input image as normal or diseased and if recognised as diseased, classifies the stage of the patient affected by glaucoma based on the CDR ratio, whether mild, extreme or normal. In this paper, we include a range of screenshots that display the effects of various techniques of image processing applied to input images.