Segmentation of Retinal Blood Vessel by Hybrid Approach
Abstract
The complications caused by diabetes in the human eye is referred to as Diabetic Retinopathy. Several methods are used for the segmentation of blood vessels of the retina. On analyzing the segmented retinal vessels, the detection of Diabetic Retinopathy can be easily done. The research paper elaborated on a novel method using enhanced fuzzy c-means using boundary extraction with Haar feature extraction for the retinal blood vessel segmentation. The proposed method has been detailed via flowchart and algorithm. The images from DRIVE databases have been used for conducting segmentation. The results have been obtained on different parameters relevant to the segmentation of retinal images (Sensitivity=0.780, Specificity= 0.9887).
Index Terms-- Fuzzy C-Means, Haar Feature Extraction, Vessel Segmentation