This research focuses on the early detection of glaucoma using digital fundus images and MATLAB-based image processing techniques. The process begins by converting the original RGB eye images into grayscale to simplify the data and reduce computational complexity while retaining essential structural information. After grayscale conversion, contrast enhancement is applied to improve the visibility of key features such as the optic disc and cup. Enhancing contrast is critical as it highlights subtle differences in intensity, which may indicate early signs of glaucoma that are not easily noticeable in the original image. Once the image quality is improved, the optic disc is segmented, which is an essential step because the optic disc is a primary region of interest in glaucoma diagnosis. Following segmentation, thresholding is used to differentiate between the optic cup and the surrounding retinal regions. Thresholding is necessary to isolate specific intensity levels, making it easier to distinguish the optic cup from the optic disc and calculate the cup-to-disc ratio (CDR). This ratio is a vital clinical indicator of glaucoma, as a larger CDR often signifies potential damage to the optic nerve. The final stage of the Research involves calculating the CDR and using it as a basis to detect the presence of glaucoma. By automating these steps in MATLAB, the system provides a consistent, objective, and efficient method for glaucoma detection, which can be used by the ophthalmologists in early diagnosis and treatment planning.

In this thesis, we explored the potential of using the Cup-to-Disc Ratio (CDR) as a key metric for the detection and diagnosis of glaucoma. By analysing retinal images and correlating the CDR values with clinical glaucoma indicators, we were able to demonstrate that CDR serves as a reliable and non-invasive measure for early glaucoma detection. The study further confirmed that a higher CDR value often correlates with the progression of glaucoma, offering a promising method for monitoring the disease over time. This approach has significant potential for improving diagnostic accuracy, especially in settings where advanced imaging equipment or specialist expertise may not be available. The results from our analysis provide strong evidence that the CDR value can be an effective tool for identifying glaucoma, but it is also essential to note that it should be used in conjunction with other clinical assessments, such as intraocular pressure (IOP) measurements and visual field tests, to ensure comprehensive patient evaluation.

[1]Abdel-Hamid, ―Glaucoma detection from retinal images using statistical and textural wavelet features‖ Journal of Digital Imaging, Springer 2020, Volume-33, pp. 151-158 [2]P.Perez, M Perez, ME. Perez and O. Arjona, ―Parallel multiscale feature extraction and region growing: Application in retinal blood vessel detection‖, IEEE Transactions on Information Technology in Biomedicine, 2010, vol. 14, no. 2,pp. 500-506. [3]D. Marin, A. Aquino, E. Arias, and M. Bravo, ―A new supervised method for blood vessel segmentation in retinal images by using gray–level and moment invariants–based features‖, IEEE Transactions on Medical Imaging, 2011, vol. 30, pp. 146-158. [4]M. Fraz, P. Remagnino, A.Hoppe, B.Uyyanonvara, A.Rudnicka, C.Owen and S.Barman, ―An ensemble classification–based approach applied to retinal blood vessel segmentation‖, IEEE Transactions on Biomedical Engineering, 2012 vol. 59, pp. 2538-2548. [5]Z.Xiao, M.Adel and S.Bourennane, ―Bayesian method with spatial constraint for retinal vessel segmentation‖, Computational and Mathematical Methods in Medicine, 2013, vol. 13, no. 401413, pp. 1-9. [6]N.Bansal and M. Dutta, ―Retina vessels detection algorithm for biomedical symptoms diagnosis‖, International Journal of Computer Applications, 2013, vol. 71, no. 20, pp. 41-46. [7]A.Budai, R.Bock, A.Maier, J.Hornegger, and G.Michelson, ―Robust vessel segmentation in fundus images‖, International Journal of Biomedical Imaging, 2013, vol. 20, pp. 1-11. [8]D.Wong, J.Liu, N.Tan and F.Yin, ―Automatic detection of the macula in retinal fundus images using seeded mode tracking approach‖, IEEE Annual International Conference of Engineering in Medicine and Biology Society (EMBC) 2012, pp. 4950-4953. [9]A. Khawaja, T. Khan, M. Khan, and S. Nawaz, ―A Multi- Scale Directional Line Detector for Retinal Vessel Segmentation‖, Sensors (Basel), 2019, vol. 19, issue 22, pp. 4949. [10]M. Islam, J. Wang, S. Johnson, M. Thurtell, R. Kardon and M. Garvin, ―A Deep-Learning Approach for Automated OCT En-Face Retinal Vessel Segmentation in Cases of Optic Disc Swelling Using Multiple En-Face Images as Input‖, arvo journal, 2020, vol. 9, issue 2. [11]S. Ceccon, F. David, Heath, P.David Crabb and A. Tucker, ―Exploring Early Glaucoma and the Visual Field Test: Classification and Clustering Using Bayesian Networks”, IEEE Journal of Biomedical and Health Informatics, 2014, vol. 18, pp. 1008-1014. [12]S.Haykin, ―Neural Networks and Learning Machines‖, 3rd Edition, Pearson Publications. [13]M.Hagan, ―Neural Network Design‖, 2nd Edition, Cengage Publication. [14]A. -M. Ştefan, E. -A. Paraschiv, S. Ovreiu and E. Ovreiu, “A Review of Glaucoma Detection from Digital Fundus Images using Machine Learning Techniques,” 2020 International Conference on e-Health and Bioengineering (EHB), 2020, pp. 1-4

© 2025 International Journal of Engineering and Techniques (IJET).

Digital Object Identifier (DOI)

IJET assigns a unique DOI to every accepted article via Zenodo for persistent linking and citation. Your article’s DOI: https://doi.org/{{doi}}

Open DOI About Zenodo DOI

Close