A REVIEW OF AUTOMATIC DETECTION OF MICRO ANEURYSM AND DIABETIC RETINOPATHY GRADING IN FUNDUS RETINAL IMAGES

Abstract

 From an image processing standpoint, the automatic detection of micro aneurysms is a challenging task, since their color and size are same as the vessels, they have a variable size and often they are so small that can be easily mystified with the image noise. It is also difficult to discriminate whether a red lesion is a micro aneurysm or small dot hemorrhage. This problem increases the number of false candidates that naturally deteriorates the overall accuracy of the detectors. Automatic early detection could limit the severity of the disease and assist ophthalmologists in investigating and treating the disease more effectively and efficiently. Therefore, time required for examination and effect of the disease on the patient could be reduced if the detection system could succeed on images taken from patients with non-dilated pupils.