Özet:

Abstract Diabetic retinopathy affects the retina and is the major cause of blindness among people with diabetes. Screening people on a regular basis to catch diseases early has traditionally required a lot of time and money. Therefore, it would be helpful if these illnesses could be automatically detected using computational approaches. Retina has several characteristics, such as exudates and micro aneurysms. Automatically detecting micro aneurysms (MAs) from colour retinal pictures is a challenging task, but their presence is an early indicator of diabetic retinopathy. To help with this, we are focusing on green Chanel images. The goal of this study is to use a classifier to identify retinal micro-aneurysms and exudates for automated DR screening. The ability to identify dark lesions and brilliant lesions in digital fundus pictures is essential for the creation of an automated DR screening system. Retinal fundus pictures from the Messidor dataset are used to identify micro-aneurysms and exudates. The characteristics are discovered using morphological processes after preprocessing.

Anahtar Kelimeler: