Implementation of segmentation of blood vessels in retinal images on FPGA

Retinal image analysis is a nonintrusive diagnosis method in modern ophthalmology. In this paper, a method to segment blood vessels in the retinal images is discussed. The morphology of the blood vessels is an important indicator for diseases like diabetic retinopathy, glaucoma etc. The segmentation of the retinal images allows ophthalmologist to perform mass vision screening examinations for early detection of Diabetic Retinopathy and its treatment planning. The images from the public dataset DRIVE are used. For better segmentation performance Spartan 3 FPGA hardware is used. The preprocessing is done in MATLAB and further processing is carried out using FPGA. For the software implementation Xilinx Platform Studio 10.1 is used.