ORIGINAL PAPER Automatic Tracing of Optic Disc and Exudates from Color Fundus Images Using Fixed and Variable Thresholds Ahmed Wasif Reza & C. Eswaran & Subhas Hati Received: 18 February 2008 / Accepted: 23 April 2008 / Published online: 22 May 2008 # Springer Science + Business Media, LLC 2008 Abstract The detection of bright objects such as optic disc (OD) and exudates in color fundus images is an important step in the diagnosis of eye diseases such as diabetic retinopathy and glaucoma. In this paper, a novel approach to automatically segment the OD and exudates is proposed. The proposed algorithm makes use of the green component of the image and preprocessing steps such as average filtering, contrast adjustment, and thresholding. The other processing techniques used are morphological opening, extended maxima operator, minima imposition, and water- shed transformation. The proposed algorithm is evaluated using the test images of STARE and DRIVE databases with fixed and variable thresholds. The images drawn by human expert are taken as the reference images. The proposed method yields sensitivity values as high as 96.7%, which are better than the results reported in the literature. Keywords Diabetic retinopathy . Exudates . Optic disc . Biomedical applications . Bioimaging Introduction Diabetic retinopathy (DR) and glaucoma are severe and widely spread eye diseases which affect the retina and cause blindness [5]. Manual analysis and diagnosis requires a great deal of time and energy to review retinal images which are obtained by fundus camera. Therefore, automatic screening for eye disease has been shown to be very effective in preventing loss of sight [18]. Detection of optic disc (OD) and exudates forms an important step in the development of such screening systems. The OD is the entrance of the vessels and the optic nerve into the retina. It appears in color fundus images as a bright yellowish or white region. Its shape is more or less circular, interrupted by outgoing vessels. The size of OD varies from patient to patient and its diameter lies between 40 and 60 pixels in 640×480 color photographs [17]. The diameter delivers a calibration of the measurements [19], and it determines approximately the localization of macula [1], the center of vision, which is of great importance as objects in the macular region affect vision immediately. Various methods have been reported for the detection of OD. In [17], OD is detected by means of morphological filtering techniques and watershed transformation. In [12], OD is localized exploiting its high gray level variation. This approach has been shown to work well, if there are no or only a few pathologies like exudates that also appear very bright and are also well contrasted. In [14], an area threshold is used to localize the OD. The contours are detected by means of the Hough transform i.e. the gradient of the image is calculated, and the best fitting circle is determined. This approach is quite time consuming and it relies on conditions about the shape of the optic disc that are not always met. Also, in [11], the Hough transform is used to detect the contours of the OD in infrared and argon- blue image. Despite some improvements, problems have been reported in cases where the optic disc does not meet the shape condition or the contrast of the image is very low. In [1], the OD is localized by backtracing the vessels to their origin. This is certainly one of the safest ways to localize the OD, but it has to rely on the backtracking of vessels. In [6], the morphological filtering techniques and active contours are used to find the boundary of the OD. In J Med Syst (2009) 33:73–80 DOI 10.1007/s10916-008-9166-4 A. W. Reza (*) : C. Eswaran : S. Hati Centre for Multimedia and Distributed Computing, Faculty of Information Technology, Multimedia University, 63100 Cyberjaya, Malaysia e-mail: awreza98@yahoo.com