Recently, Diabetic Retinopathy (DR) is found as a serious chronic disease and conquered the third portion of diabetic patients over entire world. DR is the main reason of vision loss, and the prevalence of diabetes and the day-by-day count of diabetic patients are increasing rapidly in the world. The detection of DR at its initial stages is slightly tough task but if it leaves as, it is, it may consequences to severe DR followed by permanent vision loss. So, developing an automatic DR detection system based on retinal fundus images is seriously required. In retinal images, the major signs and characteristics of DR can are observed through different components like Retinal Vessel structure, Optic Disk, Hard Exudates, Microaneurysms, Hemorrhages, and Cotton Wool Spots. The presence of one or all of these features in retinal images determines the status of DR. Hence, the automatic DR diagnosis system requires the segmentation and analysis of these components.