Pai, Vidya and Grantha, K N and Siddalingaswamy, PC (2014) Automatic Detection of Retinal Blood Vessels using Spatial Gabor Filters and Fuzzy C means Clustering. In: Proceedings of International Conference on Communication and Computing (ICC ’2014) , June 12th to 14th 2014 , Bengaluru, India .

PDF elesvierdoc.pdf - Published Version Restricted to Registered users only Download (381kB) | Request a copy

Abstract

In ophthalmic images, retinal blood vessels are one of the significant anatomical structures. Automatic segmentation of retinal vasculature is important for the early detection of Diabetic Retinopathy (DR). In this paper, a two stage method for efficient segmentation of multi-oriented blood vessels in colour retinal images is proposed. Initially, spatial Gabor filters are used to enhance blood vessels followed by Fuzzy C means clustering for the fine segmentation of vessels. Performance of this method is evaluated on retinal images from DRIVE database. An overall sensitivity of 82 % and specificity of 89% is achieved for the segmentation of blood vessels.

Actions (login required)

View Item