Computer — Aided localization of the optic disc based on textural features

In this paper, we developed a simple and efficient method for identifying and automatic detection of the position of the optic disc and also the optic nerve in retinal images. The detection is performed by applying a gliding-box type algorithm and a decision criterion based on the extreme of the selected features. The size of the windows considered as sub-images was taken so that they include the optic disc of all the images from STARE database. The results indicate a good rate of detection of the optic disc.