Fundus image features extraction

Diabetic retinopathy is the leading cause of the blindness in the working age population. If the disease is detected early and treated promptly, many of the visual loss can be prevented. This paper describes the development of an automatic fundus image processing and analytic system to facilitate diagnosis of the ophthalmologists. The algorithms to detect the optic disk, blood vessels and exudates are investigated. The optic disk is identified by the method of Sobel edge detector and LSR in the candidate area. The blood vessels and the exudates are extracted by Kirsch's method in the different color components of the color fundus image. The processing results of the proposed methods are also presented. More than thirty fundus images have been tested and the results of the system seem promising and useful to clinical work.