Localization of Optic Disc in Color Fundus Images

The objective of Medical image analysis field is to develop computational tools which will assist quantification and visualization of interesting pathology and anatomical structures. Diabetic retinopathy is a medical condition where the ration is damaged because fluid leaks from blood vessels into the retina.. The detection of the optic disk and the quantitative analysis of the evolution of its shape and its size can bring clinical information of big importance. This paper describes a novel method for localization of the optic disk boundary of retinal images. The proposed method consists of two steps: in the first step, a circular region of interest is found by first isolating the brightest area in the image by means of contrast enhancement and in the second step, the Hough transform is used to detect the main normal circular feature, the optical disk within the positive horizontal gradient image within this region of interest. The technique is tested on publicly available DRIVE, diaretdb0, diaretdb1 databases and Live images form hospital. Results obtained by applying these are presented on GUI (Graphic User Interface). Initial results on a database of fundus images show that the proposed method is effective and favorable in relation to comparable techniques. The intention of using GUI as an automated detection program is to allow simple and easy access for doctors or nurses to perform quick analysis and observation without the need of programming skills. The proposed method achieves an average accuracy of 94.7 percentage for localization of optic disk.