Design Of Image Processing System Based On FPGA And GigE Vision Camera

With the rapid development of industrial technology in China,machine vision system is becoming more and more popular in the fields of aerospace,national defense,industry and medical treatment.Especially in the field of intelligent manufacturing and automation industry,digital image processing systems are an indispensable part,and the requirements for image quality and transmission speed are also getting higher and higher.At the same time,more stringent requirements are placed on the efficiency of image processing.Among them,the image data acquisition part is the basis of the machine vision system,and the efficiency of image acquisition directly affects the performance of the entire image processing system.This article is mainly to use machine vision technology to solve the problem of detecting defects in the date coding of packaging boxes on the milk production line.In order to adapt to the fast running requirements of industrial assembly line,this paper designs a date coding defect detection system based on FPGA and GigE Vision camera.The Code-spurting defect detection system designed in this paper is composed of three parts: image acquisition and transmission module,image processing(code defect detection)module and system software operation interface.The image acquisition and transmission module is based on Xilinx's ZYNQ series FPGA chip,and the GigE Vision camera is configured into an image acquisition system with ZYNQ as the core and Gigabit Ethernet as the data transmission method.At the same time,the Linux system is connected to the Visual Studio 2017 platform to cross-platform design the GigE Vision camera image acquisition interface program and image transmission program,and the image processing(printing defect detection)module is designed based on the Open CV computer vision library,and finally the system is designed based on the Py Qt5 platform Operation interface of the software.The test data shows that the image acquisition and processing system based on FPGA and GigEVision camera designed in this paper not only has fast image acquisition speed and low logic resource occupancy rate,about 10%,but also can complete image processing function in real time.Under the premise that the image size is 1920x2560,the image acquisition speed reaches 20 frames per second,reaching the peak of the camera.Therefore,the image acquisition and processing system designed in this paper is suitable for various industries,such as food and beverage packaging industry,consumer electronics processing industry,and mechanical parts processing industry.The data Code-spurting defects detection system for milk carton designed in this paper has been tested on site,and the accuracy rate of date-coded defect detection has reached 97%.After comparison,its detection speed and accuracy are much greater than manual detection.Therefore,this system not only improves production efficiency,but also reduces production costs,and has high industrial practical value.