Automatic Localization of Macula and Identification of Macular Degeneration in Retinal Fundus Images

In aged people, the central vision is affected by Aged-Related Macular Degeneration (AMD). The purpose of this research is to make safe, non-contact and cost-effective platform that can be used for the localization of the macula and monitoring system for dry AMD. Firstly, the fundus image is pre-processed using Contrast-Limited Adaptive Histogram Equalization (CHALE) and bottom-hat transformation function. Secondly, the enhanced image is used to extract the vascular structure that is blood vessels by using Kirsch's Template Method and the OSTU thresholding is used to segment out dark regions because macula are dark in color. The extraction of the blob like structures, dark region and background, estimated image is subtracted from extraction blood vessels image. The shape features are removed in order to differentiate between blobs like structure from vessel segments and the eccentricity of the candidate region further differentiate between vessels like structures from blob-like structures. This system results were evaluated and compared with state-of-the-art techniques. The result is demonstrated that the proposed system is appropriate for localization of macula and early detection of dry AMD.