Precision-Based Detection and Centralization of the Optic Disc in Color Retinal Fundus Images: A Novel Approach

Abstract

The detection of Optic Disc (OD) stands as the essential process in color retinal image analysis because it represents the main component of the human retina. The detection of Optic Disc along with its center enables healthcare providers to evaluate the severity of conditions like Diabetic Maculopathy (DM) and glaucoma. This document presents a fresh technique to discover optic disc regions while also determining their central areas. The first process enhances the input image through combination of Gaussian, Median, and Weiner filters to create the enhanced image. After image enhancement the morphological closing operation removes BVs as its next step. Adaptive Histogram Equalization is then applied before extracting Region of Interest using four co-ordinate points that are determined with the help of a drawn grid. The combination of Extended Maxima Transform and Morphological operation produces the OD region as well as its center point. The proposed method delivers evaluation results on MESSIDOR which achieve an accuracy rate of 96.21%. The proposed innovative approach detects OD without using Fovea or Macula while needing grid setup during the first run to obtain Region of Interest (RoI).