Open AccessDetection and Classification of Early Stage Lesions in Diabetic Retinopathy using Color Fundus Images
Author(s) -
S. Sudha*,
A. Srinivasan,
T. Gayathri Devi
Publication year - 2019
Publication title -
international journal of recent technology and engineering
Language(s) - English
Resource type - Journals
ISSN - 2277-3878
DOI - 10.35940/ijrte.c6806.098319
Subject(s) - artificial intelligence , diabetic retinopathy , fundus (uterus) , pattern recognition (psychology) , computer science , classifier (uml) , cluster analysis , stage (stratigraphy) , computer vision , medicine , ophthalmology , diabetes mellitus , biology , paleontology , endocrinology
Detection of lesions and classification of Diabetic Retinopathy (DR) play an important role in day-to-day life. In this proposed system, colour fundus image is pre-processed using morphological operations to recover from noises and it is converted into HSV colorspace. Fuzzy C-Means Clustering algorithm (FCMC) is used for segmenting the early stage lesions such as Microaneurysms (Ma), Haemorrhages (HE) and Exudates. Hybrid features such as colour correlogram and speeded up robust features (surf) are extracted to train the classifier. Cascaded Rotation Forest (CRF) classifier is used for classification of diabetic retinopathy. The proposed system increases the accuracy of detection and it has got high sensitivity.