We propose image processing techniques for the detection of blood vessels in fundus images of the retina. The methods include the design of a bank of directionally sensitive Gabor filters with tunable scale and elongation parameters. Forty images of the retina from the Digital Retinal Images for Vessel Extraction database were used to evaluate the performance of the methods. The results of blood vessel detection using inverted green-channel images were compared with the corresponding manually segmented blood vessels. High efficiency in the detection of blood vessels with the area under the receiver operating characteristics curve of up to 0.96 was achieved with a combination of Gabor filters at three scales.