International Journal of Advanced Innovative Technology in Engineering (IJAITE)

---

Automatic Identification Of Diabetic Retinopathy Using Support Vector Machine

Prof. Chetan A. Mugal

Abstract :

Diabetic Retinopathy (DR) is a main cause for blindness, detecting it as early by exudates that form in the retina. The old method followed by opthalmogists is the regular supervision of the retina. Diabetic retinopathy (DR) is a disease that cause by diabetic and it is described by the development of retinal soft exudate like microaneurysms and hard exudates. The exudates are extracted from the fundus image by applying threshold and removal of optic disk and region of interest using morphological operation like closing, dilation, erosion. The features are extracted from processed image and used for classifying into exudates and non-exudates In early stage of disease, the damage occurred can be prevented. In this paper, were comparing Support Vector Machine (SVM), K N N c l a s s i f i e r s f o r automatic microaneurysms a n d e x u d a t e s detection in images acquired through non-dilated pupils. Micro-aneurysms and exudates are first detected and features are get extracted from the image. These features are fed to SVM and kNN classifier.

Keywords :

Full Text :

Download PDF

DOI :

Cite this paper :

References :

---