Design And Implementation Of A Video Image Processing System Based On FPGA

At present,the technology of the computer software for video image processing technology provides a good development platform,computer processor powerful processing ability in all kinds of video image processing algorithm are also obtained the good application.High integration,high performance,low power consumption of Field Programmable Gate Array(FPGA)(Field Programmable Gate Array,FPGA)device for video image system designer has provided a broad prospects for development,FPGA unique performance advantages in embedded video image processing system occupies a very important position.Now many colleges and universities have opened the FPGA theory teaching course,in order to enhance the teaching practice,this thesis through the design of FPGA implementation part of the video and image processing technology,hope to play a good role in the two aspects,one is the large amount of calculation based on the embedded video and image processing algorithm by hardware implementation in order to accelerate the operation speed and reduce the hardware cost;The second is for teaching provides a good example of the FPGA,FPGA teaching lively and interesting,to make the students easy to understand,and through the example can learn system programming,improve the ability of hands-on practice and application.This paper designs the system in order to Alter the Cyclone IV series as the core device,was introduced in detail on the chip of the logical system design,including the camera configuration module,data acquisition module,SDRAM module,VGA controller module and a variety of image algorithm module,etc.On the basis of this research is mainly focused on the image filtering algorithm of median filter and gaussian filter and four kinds of commonly used edge detection algorithm.To sort too many disadvantages in traditional median filter,this paper proposes a fast median filtering algorithm to reduce the number of sort,In order to improve the precision of the Laplace edge detection algorithm,the proposed edge detection to extract more direction to improve accuracy;In order to save system hardware resources of gaussian filter and Sobel edge detection algorithm processing was optimized by using the displacement,and through the system resource usage report to prove the advantage of this design,at the same time use the Modelsim simulation experiment to prove that the FPGA parallel processing and pipelining design speed advantage.After the system design is completed,the experimental platform is built and the system is tested and analyzed.The experimental results show that the median filter and gaussian filter inthe design of this paper have a good inhibitory effect on the noise and gaussian noise respectively,and can meet the requirements of fast real-time filtering.The experimental results show that the improvement of Laplace operator can improve the detection accuracy.Finally,the detection effect of Sobel operator is best in the comparison of four edge detection.The experimental results show that the system can meet the requirements of the fast real-time processing of video images and can be applied to practical engineering practice,while providing a good example for the teaching of FPGA.