Abstract:
Vascular Imaging is a key component in comprehensive stroke diagnosis. Aiming at the complex background of vascular images, weak contrast between the vessels and the background, a new vascular enhancement method based on classical image processing techniques is proposed. The problem in Canny’s hysteresis thresholding method having two threshold values which yielded low performance in localization and noise rejection has been overcome with a not complicated and accurate algorithm. In this paper an efficient image enhancement algorithm i.e., pre processing, contrast adjustment and edge abstraction is implemented on Spartan -3 FPGA using Simulink model which comprises of Simulink blocks and Xilinx blocks. Accurate and faster results are obtained by interfacing with Matlab and XSG through the workspace of Matlab.

Keywords: Segmentation, Vessel extraction, Canny edge detector, thresholding, Image enhancement, FPGA Implementation, Xilinx System Generator (XSG), Simulink, Co-simulation