Automated Optic Disc and Optic Cup Segmentation for Glaucoma Screening Using Fundus Images
Contributors
Dr. Arun Vikas Singh
Shashikant Gupta
Keywords
Proceeding
Track
General Track
License
Copyright (c) 2026 Sustainable Global Societies Initiative

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Abstract
Glaucoma is a degenerative optic-nerve disease in which structural changes can precede severe visual loss. For computer-assisted screening, reliable extraction of the optic disc (OD) and optic cup (OC) from color fundus images is important since the cup-to-disc ratio and related geometric measures directly depend on the quality of the boundaries. Two approaches are considered. The first is Reaction Diffusion Level Set Segmentation (RDLSS), an extension of level-set evolution in which a diffusion term is added to stabilize the evolving boundary. The second combines harmonic-mean filtering, connected-component analysis (CCA), morphological removal of retinal vessels, Three-Sigma control limits, and active contour refinement. The RDLSS experiment used 175 retinal images and achieved 98.95% average accuracy. The second approach was tested on 100 RIGA images of size 526×720 pixels; it was able to locate fully visible optic discs and failed when the disc was partial or absent. The results show that preprocessing and region-focused segmentation are essential to dependable optic-disc/optic-cup extraction and provide a practical front end for an automated glaucoma analysis [1], [2].