Novel Method to Detect Exudates Lesions in Color Retinal Images by Using Morphology Mean Shift Algorithm

Exudates are a serious complication inflicting blindness in diabetic retinopathy patients. The most important goal of this paper is to enhance a novel technique to observe exudates lesions in shade retinal pictures by means of the use of a morphology imply shift algorithm. The proposed approach begin with a normalization of the retinal image, distinction enhancement, noise removal, and the localization of the OD. Then, a coarse segmentation technique by way of the usage of imply shift gives a set of exudates and non-exudates candidates. Finally, a classification the use of the mathematical morphology algorithm (MMA) manner is utilized in order to hold solely exudates pixels. The most effective price parameters of the MMA will facilitate an expand of the accuracy effects from the totally MSA technique by means of 13.10%. Based on a contrast between the outcomes and floor reality images, the proposed technique got an common sensitivity, specificity, and accuracy of detecting exudates as98.40%, 98.13%, and 98.35%, respectively.