Diabetes, a chronic metabolic disorder, arises from either the insufficient production of insulin or the body’s diminished responsiveness to insulin. This condition gives rise to severe complications, including cardiovascular disorders, vascular diseases, strokes, kidney dysfunction, neuropathy, and diabetic retinopathy—a condition affecting the eyes. Diabetic retinopathy, characterized by progressive damage to retinal blood vessels, poses a substantial threat to vision. This paper focuses on the automated detection of diabetic retinopathy (DR) in fundus images through the application of advanced image processing and machine learning techniques. The quantification of disease progression within the retina is achieved by extracting relevant features, such as blood vessels, haemorrhages (associated with non-proliferative diabetic retinopathy, NPDR), and exudates (associated with proliferative diabetic retinopathy, PDR), from raw fundus images using image processing techniques. From the utilized dataset, 70% of the images are designated for training and the remaining 30% for testing. These findings highlight the potential of machine learning techniques in providing a precise and efficient automated diagnosis of diabetic retinopathy, contributing to the advancement of early intervention strategies in the management of diabetes-related ocular complications.