Research On Real-time Edge Detection System Based On FPGA

Edge detection is one of the key technologies of image processing,and the accurate extraction of edge signal provides the basis and guarantee for subsequent target recognition,target tracking and other technologies.Edge detection algorithm has become more and more mature after many years of research and development,but at the same time,the workload of program execution is also increasing,making the processing time of the program increase,the traditional software serial processing method is increasingly unable to meet the needs of real-time processing.In view of the problem of large amount of edge loss when using the traditional edge detection algorithm to extract edge directly in the haze environment,a combination of image-enhanced edge detection algorithm is designed,and a real-time edge detection system with video image acquisition and display function is designed by using the characteristics of FPGA parallel and high-speed processing data.In order to meet the needs of real-time processing of high-traffic video image data,a real-time edge detection system based on FPGA platform is established,which is composed of image acquisition module,image processing module,data cache module and image display module.The OV7725 image sensor completes the video image acquisition function of the external environment,and then processes the captured video image data through the image processing module,caches the resulting image edge data in SDRAM,and finally outputs it to the display in real time through the VGA interface.In order to improve the edge extraction effect of video images in haze environment,the image processing module uses an improved Canny algorithm combined with grayscale changes to enhance the extraction of edge details through appropriate grayscale variation and expansion of traditional Canny operator gradient templates.In FPGA engineering,the edge detection algorithm is realized through brightness mapping module,Gauss filter module,gradient amplitude and direction calculation module,non-extreme value suppression module and dual threshold processing and conversion module.In the design process,the algorithm effect is verified by MATLAB software,and the function verification of various modules in FPGA project is carried out by Modelsim software.After the completion of the real-time edge detection system,the experimental results show that the real-time system can restore more edge details in the haze environment than the traditional edge detection system.At the same time,the system has real-time,which provides a fast solution for real-time edge detection in haze environment.