Hardware Software Co-Simulation and Real-Time Video Processing For Edge Detection Using Matlab Simulink Model Blockset

Abstract

FPGA can be useful in many different applications, such as developing more complex systems to be compatible. Using the blocks designed by Xilinx for its System Generator software, a simple algorithm can be designed and tested using Simulink and an FPGA development board, for example edge detection algorithm. Edge detection of image significantly reduces the amount of data and filters out unwanted or insignificant information and gives the significant information in an image. Implementing edge detection is not limited to software development but also hardware development. Edge detection implemented in hardware have emerged as the most viable solution for improving the performance of image processing systems.

This application note will make use of the Spartan-3A DSP development board to explain how to co-simulate Xilinx’s highest level form of an Edgefilter on a still image and implementation sobel edge detection on real-time video based on FPGA board. This document begins by explaining how to convert a still image into a format useable by Matlab and Simulink, shows the construction of a basic algorithm using Xilinx’s System Generator blocks, and also illustrates the procedure of implementing the design real-time video processing using the FPGA development board. The light condition and color contrast will affect the edge detection result. The result of this implementation is detecting edges from the real-time video that has been captured by the camera.

Keywords: Sobel Fiter, Edge Detection, Co-simulation, FPGA, Matlab, Simulink