Abstract: The main objective of this proposed system is to detect Glaucoma in the retinal image. Glaucoma is an eye condition that can’t be healed once it happens. In case the corrective therapy does not continue, it causes a permanent visual disability so it cannot be ignored. Treatment will be helpful when the disease is identified at an early stage. Most of the research describes different techniques widely incorporated in the detection of Glaucoma disease.
In this proposed system, the detection of Glaucoma is identified through Image Pre-processing and SVM algorithm. Pre-processing operators like Segmentation, Enhancement, Binarization, and Thresholding are used to extract the optic cup and optic disc from the retinal image to find the CD R. This proposed technique is based on OTSU’s segmentation method to locate the Optic cup and disc. Calculating only the CDR (Cup-to-Disc ratio) does not help to distinguish all the images as Glaucomatous or normal. Thus, RDR (Rim-to-Disc ratio) is considered another feature for Glaucoma assessment. The SVM (Support Vector Machine) algorithm plays an important role.
Keywords: Cup to Disc Ratio (CDR), Rim to Disc Ratio (RDR), Support Vector Machine (SVM), Optic Disc (OD), Optic Cup (OC), and Region of Interest (ROI).
|
DOI:
10.17148/IJARCCE.2023.12589
