Video Image Acquisition And Processing System Design And Implementation Base On FPGA

In the information age,with the rapid development of IC(Integrated Circuit)technology and image acquisition,processing and storage technology,a real-time processing system that integrates video image acquisition,processing,and storage is widely used in security,industry,medical,aviation,and security systems,The importance of its status cannot be overemphasized.The edge detection algorithm can extract the edge information of the corresponding image,which is an important cornerstone for the realization of algorithms such as target recognition.However,due to the complexity of the algorithm and the higher and higher resolution of the image,the traditional use of software platforms to process images has become more and more limited,and the performance in real-time performance is poor.FPGA(Field Programmable Gate Array)can process images in real-time due to its parallel data processing method,pipeline technology,and rich register resources.This paper takes the FPGA platform as the core and designs a video image acquisition and processing system by improving the traditional Canny operator.In this thesis,the related algorithms in the image processing module are improved and their feasibility is verified.Use MATLAB to simulate the relevant edge detection algorithms.After the comparison and screening,the Canny algorithm is selected to implement the edge detection system,and the Canny algorithm is improved.First,according to the shortcomings of the Gaussian filter algorithm,the median filter algorithm is used to replace the Gaussian filter;The 3 × 3 template in the Sobel algorithm replaces the 2 × 2 template in the traditional Canny algorithm,and the direction of the convolution template is extended,adding convolution templates in 45° and 135° directions,which can make full use of the pixels in the image.The direction information of the points is used to enhance the image edge detection effect.Finally,Finally,it is improved on the basis of Bernsen’s algorithm,the size of the matrix window is changed to 3 × 3 and a threshold value K is given,the difference between Gmax and Gmin is compared with it,and then the local threshold is determined,and this method is used to replace The threshold selection method in the traditional Canny operator is used to realize the dynamic processing of the threshold.At the same time,the sorting method in the median filtering algorithm is also improved,thereby greatly reducing the number of comparisons in the algorithm implementation process and improving the system.Overall operation rate.In this system,the Artix7 series chip of Xilinx company is used as the main control chip,and the overall system construction is completed on the FPGA development platform with this chip as the core,and the image data is cached by the off-chip DDR3(Double Data Rate 3)to complete the image acquisition and processing.system design.The system includes the following modules: OV5640 camera acquisition module,image processing module,DDR3 controller module,and HDMI(High Definition Multimedia Interface)display module.The design scheme of each module in FPGA is introduced in detail,and use Verilog language to complete the design of related functional modules.Modelsim software simulates and verifies the functions of each module after adding excitation,and the simulation results show that the design of each module meets the overall circuit requirements;After completing the design verification of the entire video image acquisition and processing system,complete the relevant process of the FPGA design in the Vivado software,and download the final generated bitstream file to the FPGA development board to complete the overall verification of the system.The experimental results show that the system can collect,process and display video images in real time.Compared with the traditional edge detection algorithm,the improved algorithm proposed in this design has greatly improved the detection effect and can effectively reduce the occurrence of false edges.The probability of detection makes the detected edge more accurate,and it can also remove the influence of salt and pepper noise,filtering noise and non-uniform lighting conditions on edge detection,which achieves the desired effect of the system.