A Novel Method for Fovea Extraction from Retinal Fundus Images Chitra Raju I Electronics and Communication Dept. LBSITW chitraraju.1991@gmil.com Lizy Abraham Electronics and Communication Dept. LBSITW lizytvm@yahoo.com Abstract—Medical image analysis and processing has great significance in the field of medicine, especially in non-invasive treatment and clinical study. Retinal fundus images are normally manually graded by specially trained clinicians in a time consuming and a resource intensive process. Fovea is one of the important features of a fundus retinal image. This paper aims in presenting a simple and fast method using morphological operations and thresholding methods to automatically extract the fovea region. Proposed method is based on the structure of the blood vessels extracted from green channel component images. The thresholding method uses local threshold values that separate the foreground from the background region which helps in identifying the fovea region more clearly. The proposed method was implemented in Matlab and evaluated on a publicly available DRIVE database and the results have been promising. Keywords—fovea; blood vessels; thresholding; morphology; fundus image. I. INTRODUCTION Retinal fundus images are used for diagnosis by trained clinicians to check for any abnormalities or any change in the retina. They are captured using special devices known as opthalmoscopes. A typical fundus image with its feature marked is shown in Fig.1. Each pixel in the fundus image consists of three values namely red, green and blue, each value being quantized to 8 bits. Fovea is the most fundamental aspect of retina. It is the key region which is responsible for high acuity colour vision and is responsible for human vision. It consists of many delicate cones which if destroyed may lead to blindness. Conventional manual detection of fovea region by ophthalmologists is time consuming and resource intensive. Early detection and diagnosis of retinopathies decrease the risk of blindness and effectively control the illness. However, due to problems of retinal image quality, such as poor contrast and physicians’ subjective observation, the diagnosis can be unstable and uncertain. In countries like India, automation field is highly developing. Computer analysis systems for retinal images can offer an efficient and stable assistance for diagnosis. The performance of the computer analysis system is generally influenced by retinal anatomy detection. The fovea is difficult to observe. The fovea region is always darker than the surrounding retinal tissue. The fovea is located in the center of the macula region of the retina. It is responsible for sharp central vision (also known as foveal vision), which is necessary for humans to read, drive, and any activity where visual detail is of primary importance. Fovea size is relatively small when compared to the rest of retina, but the fovea is very important for seeing fine details and color. Usually this fovea zone is approximated to a circle of diameter 400 micron. The central region called the macula is a circular area measuring about 4 to 5mm in diameter. A small depression in the centre of the macula is called fovea. Fig.1. Illustration of fundus image Manual detection of fovea region by ophthalmologists is time consuming and resource intensive. Due to unavailability of trained ophthalmologists especially in developing countries like India, automation is highly needed. The paper is organized into five sections. Section II discusses related works in detail and the proposed methodology is explained in Section III. In this section a detailed explanation about the method is given followed by a discussion on the results obtained in Section IV. Finally, this paper is concluded in Section V. II. RELATED WORK Retinal image analysis is a complicated task particularly because of the variability of the images in terms of the color, the morphology of the retinal anatomical pathological structure and the existence of particular features in different