The window to retinal vascular system is the eye. Retinal blood vessels play a vital role in the detection of eye diseases like diabetic retinopathy, hyper tension retinopathy and glaucoma. In this work, Gabor filter is used to enhance the blood vessel. The optimal threshold value is obtained for the gray scale images. Otsu based threshold technique is used to segment the retinal blood vessel. The vessel segmentation approach has been tested on the publically available DRIVE dataset. The extracted features are classified by K-Nearest Neighbor classifier. The performance of the system is proved by the experimental results.