RED Lesion Recognition in Diabetic Retinopathy using Image Processing
V. Ramya , Saranathan College of Engineering; S.Prakash, Saranathan College of Engineering; R.Raj Mohan, Saranathan College of Engineering; P. Mohan Raj, Saranathan College of Engineering; G. Ruthra, Saranathan College of Engineering
Computer Aided Diagnostic, Diabetic Retinopathy, Fundus Imaging, Dynamic Shape Features, Random Forest
The number of people with diabetes is about 400 million in 2015. The global prevalence of diabetes among adults over 15 years of age has risen from 4.6 % in 1990 to 8.0% in 2015.Diabetes, often referred by doctors as diabetes mellitus that describes as a group of metabolic diseases in which the person has high blood glucose (blood sugar), either because insulin production is insufficient, or because the body's cells do not respond properly to insulin, or both. In this paper, a novel approach for the detection of both microneurysms and hemorrhages in fundus images. It has 4 sections: 1.Image Acquisition 2.Segmentation 3.Feature Extraction 4.Classification. The features represent the evolution of the shape during image flooding and allow to distinguish between lesions and vessel segments. On the Messidor database, when detecting images with diabetic retinopathy, the proposed method takes 131.611as execution time.
|Manuscript Id ||:||IJSTEV3I10101|
|Published in ||:||Volume : 3, Issue : 10|
|Publication Date||:|| 01/05/2017|