Localization of Fovea Region in Retinal Fundus Images

Medical image analysis is important for early detection of several diseases related to the eye. The retinal fundus photographs are widely used in the diagnosis and treatment of various eye diseases such as diabetic retinopathy, glaucoma etc. Retina can be analyzed by detecting the features such as blood vessels, optic disk, macula, fovea, etc. The fovea is a small, slightly concave area without retinal capillaries and its detection is very important for ocular diagnosis systems. It appears as a dark area, in contrast to the rest of the retina where retinal vessels are enhanced by the fluorescent liquid. The fovea is so important because some specialized cells that provide central vision lie in it. This paper describes an automated detection of fovea. Detection of fovea is dependent upon approximate location of Optic Disk.Initially the retinal blood vessels is detected using mathematical morphology. Since the optic disk is usually the brightest component on the fundus, the high intensity value of pixels is used to locate it. From these values the highest pixel value is determined which is considered as the centre of the optic disc. Using this, the position of the optic disk is located approximately using the simple circular fitting method. Geometrical features are then used to localize the macula region successfully. The centre of the macula is the localized fovea region.