Retinal vessel tracking using bilateral filter based on Canny method

This paper presents a new method to measure retinal blood vessel with sub-pixel accuracy. The method is based on a modified Canny edge detection method with a bilateral filter. There are three stages of the method. First, the bilateral filter is used to remove the vessel background noises, and then the Canny detector is used to detect all vessel edges. Second, vessel sample profiles crossed vessel boundaries are obtain based on Canny edges. Third, the new vessel positions are measured from the Gaussian fits of the sample profiles. The validation of the method is tested by real images with outputs of the vessel walls and center lines in sub-pixel accuracy.