Skip to Main Content
This paper proposes two efficient approaches for automatic detection and extraction of Exudates and Optic disk in ocular fundus images. The localization of optic disk is composed of three steps. First the centre of optic disk is estimated by finding a point that has maximum local variance. The color morphology in Lab space is used to have homogeneous optic disk region. The boundary of the optic disk is located using geometric active contour with variational formulation. The Exudates identification involves Preprocessing, Optic disk elimination, and Segmentation of Exudates. In Exudates detection the enhanced segments are extracted based on Spatially Weighted Fuzzy c-Means clustering algorithm. The Spatially Weighted Fuzzy c-Means clustering algorithm is formulated by incorporating the spatial neighborhood information into the standard FCM clustering algorithm. The Experimental results of both approaches are validated with ground truth images. The proposed algorithm for optic disk detection produces 92.53% accuracy. The sensitivity and the specificity of the proposed algorithm for exudates detection are 86% and 98% respectively.