International Journal of Scientific & Engineering Research Volume 3, Issue 7, July-2012 1 ISSN 2229-5518 IJSER © 2012 http://www.ijser.org Detection of Diabetic Retinopathy Using Sobel edge detection method in DIP Jyoti Patil Assistant Professor I2IT, Hinjewadi, Pune-411057 India Email: jyot.physics@gmail.com Dr. A. L. Chaudhari H.O.D. Department of Electronics MGSM’s Arts, Science & Commerce College, Chopda Dist. Jalgaon 425 107 India Email: chaudharial@yahoo.co.uk Abstract— Diabetic retinopathy, a complication of diabetes that occurs as a result of vascular changes in the retina, It is a major cause of loss of vision. Automated image processing has the potential to assist in the early detection of diabetes, by detecting changes in blood vessel patterns in the retina. Image processing techniques can reduce the work of ophthalmologists and the tools used automatically locate the exudates. 0In this paper the process and knowledge of Digital Image Processing (DIP) is used. Automated analysis techniques for retinal images have been an important area of research for developing screening programmers. By using MATLAB for programming to develop the DIP tool for diagnosis of eye infection . Sobel edge detection algorithm is a method to find the edge pixels in an image. Edges are pixels which carry important information in an image. Thus sobel method is best technique for features are extended & used to classify the pixels in the patch into vessel and non vessel. Keywords— Diabetic Retinopathy, DIP, MATLAB, 1. INTRODUCTION D iabetic retinopathy (DR) is a severe eye disease that affects many diabetic patients. Diabetic retinopathy is the most common cause of blindness which a complication of diabetes mellitus, so it is necessary to diagnosed early. The eye, a vital organ of the human body, gives us the sense of color, shape and state of physical objects. But if Abnormalities occurs in the eye because of diseases such as Conjunctivitis, Fungal Keratitis, glaucoma, diabetic retinopathy, fungal infection, diabetes then eye may be damaged [1]. The complicated images obtained from infected eye will be processed using digital Image Processing (DIP) technique, which manipulates the image for the purpose of either extracting information from the image or produces an alternative representation of the image. Thus screening is the most effective method to detect early signs of diabetic retinopathy [2-3]. Using screening method big blood clots called hemorrhages, Hard exudates, The bright circular region from the blood vessels called optic disk, The fovea defines the center of the retina, and is the region of highest visual acuity, exudates and microaneurysms, irregular shaped, and found in the posterior pole of the fundus can be detected. Ma et al. [3] defined a quality descriptor according to three classes, namely, out-of-focus images, motion blurred images and severely occluded images of eyelids and eyelashes. Zhu et al. [4] proposed a quantitative quality measure using discrete wavelet decomposition. Analyzing and interpreting retinal images have become a necessary and important diagnostic procedure in ophthalmology. We are interested in vessel segmentation in color images for screening of diabetic retinopathy. Thus to remove noise, enhance objects of interest - blood vessels, damaged areas, Changes in the blood Vessel Structure we can use Sobel algorithm, and Laplacian of Gaussian operator, which detects the edges of blood vessels [5]. Micro aneurysms [tiny dilations of the blood vessels] are the first apparent sign of diabetic retinopathy so that their detection in fundus images through photography might be detect the disease in an early stage. 2. METHODS OF DETECTION SOBEL EDGE DETECTION METHOD Detection of vessels, exudates, and hemorrhages, blood clots, Hard exudates, optic disk is possible using Sobel method. Edge detection is the process of localizing pixel intensity transitions. The Sobel operator is an algorithm for edge detection in images discovers the boundaries between regions also it determine and separate objects from background in an image. It’s an important part of detecting features and objects in an image [6]. The Sobel method finds edges using the Sobel approximation to the derivative. It returns edges at those points where the gradient of I is maximum. Where the gradient of the considered image is maximum. The horizontal and vertical gradient matrices whose dimensions are 3 × 3 for the Sobel method has been generally used in the edge detection operations [7]. If we define A as the source image, and Gx and Gy are two images which at each point contain the horizontal and vertical derivative approximations, the computations are as follows.