Diabetes occurs when the pancreas can not secrete enough insulin, slowly affecting the retina of the human eye. As it progresses, a patient's vision begins to deteriorate, leading to diabetic retinopathy. In this sense, the retinal images acquired through the camera help analyze the consequences, the nature and the state of the effect of diabetes on the eye. Diabetic retinopathy is a disease of the retina that affects patients with diabetes mellitus and is the main cause of blindness. It is a disease in which the blood vessels of the retina swell. This damages the retina of the eye and can lead to blindness if the level of diabetes is very high. The most effective treatment is early detection through regular exams. The automatic detection of these images would help doctors to easily detect the patient's condition in a more accurate way. This emphasizes in the determination of retinal images using appropriate techniques of image processing and data mining. With this we can easily classify the normal and abnormal images of the retina, this will reduce the number of revisions for doctors.