Automatic glaucoma detection by using funduscopic images

This paper describes an automatic system to identify glaucoma disease from funduscopic images using digital image processing. Glaucoma is caused by the increased pressure of the eye and damages the optic nerve. Since glaucoma does not show early symptoms, it should be diagnosed by the doctor. Through this new detection technique, doctors can quickly and easily identify patient's condition and carryout the treatment necessary. This method also has the added advantage if being affordable. Here glaucoma is identified through cup (optical disc's inner circle) to disc (outer circle) ratio (CDR) calculation and by the orientation of the blood vessels. In this system firstly, cup and disc are extracted using average and maximum grey level pixels respectively with the use of histogram. Then contours are found, which in turn are used to draw the best fitting circle, thus finding the radius of cup and disc. After calculating CDR, the abnormal image can be recognized if CDR exceeds the threshold value. Otherwise it is a normal image. The system extracts the blood vessels and through their orientation glaucoma is identified.