A Fast Enhancement/Thresholding Based Blood Vessel Segmentation for Retinal Image Using Contrast Limited Adaptive Histogram Equalization

Automatic detection of blood vessel in retinal fundus image is an important task in the computer aided diagnosis of ophthalmology. This paper presents a fully automatic enhancement/thresholding based vessel extraction method. The input image is enhanced by histogram matching and Contrast Limited Adaptive Histogram Equalization (CLAHE) techniques. Following CLAHE, Wiener filtering is carried in order to remove the background noise. A local entropy based thresholding technique is then used to extract blood vessel from the 2 dimensional Gabor filter response of CLAHE'd image. The performance of the proposed method was evaluated on two publicly available DRIVE and STARE databases and compared with the methods reported recently. The proposed method extracts blood vessels in a DRIVE image within 1.47 seconds(s) with (accuracy, sensitivity, specificity) = (95.74%, 72.59%, 97.99%), and (95.26%, 76.93%, 96.72%) for the DRIVE and STARE databases, respectively. The average predictive value on both the databases were also higher (73.56%) compared to the recent methods.