Abstract
Automatic segmentation of the normal features such the blood vessels may help to develop the future of medicine. It can give an earlier diagnosis of such eye diseases as diabetic retinopathy and glaucoma; it can support the specialists in their decision as well. In this paper, a method of retinal blood vessel segmentation is proposed. The database used is HRF of a total of 45 fundus images. The fundus images go through a MATLAB code in preprocessing steps of image acquisition, grayscale conversion and contrast enhancement, intensity adjustment, complement and adaptive histogram equalization. Then, the blood vessel segmentation process includes mathematical morphological opening, binarization and noise extraction. The proposed method shows a specificity of 97%, a sensitivity of 69% and an Accuracy of 94%.