This project presents an advanced image processing system designed for the early detection of diabetic retinopathy. The system utilizes pixel-by-pixel assessment of retinal fundus images to identify subtle changes indicative of the disease, such as alterations in the cup-to-disc ratio, and the presence of exudates and hemorrhages. The methodology involves a software-based approach using MATLAB, incorporating pre-processing steps like resizing, color normalization, and edge enhancement to optimize image quality. The core image processing phase employs parallel functions for cup-to-disc ratio analysis, exudate detection, and hemorrhage detection. The system aims to provide a more accurate diagnosis compared to visual assessments, particularly in the early stages of the disease. The project includes requirement analysis, functional specifications, detailed design, coding, and various testing phases, culminating in a user acceptance test to ensure the system's effectiveness. The project's limitations are also acknowledged, specifically the system's focus on diabetic retinopathy symptoms, potentially affecting its ability to detect other ophthalmological issues.
![[object Object]](/_next/static/media/star-bottom.7253800d.svg)
