The Design Of Image Acquisition And Real-time Processing System Based On FPGA

With the improvement of automation, the real-time processing system based on vision detection appears frequently in people’s life, and has been widely used in the field of public security traffic, safety monitoring, mechanical test, industrial detection etc.In the currently existing visual system, visual acquisition and monitoring system is the dominant, because the systems which can complete the system real-time vision processing algorithms is rare,wasting a lot of resources of the hardware.But in the embedded system,the processing ability of the FPGA and DSP is more and more strong.Especially in FPGA,the use of its advantages of parallel processing, some pixel level algorithm that can complete the real-time in FPGA,such as video coding, image filtering, edge extraction.At the same time, when the resources are permitted, FPGA can copy multiple in one chip of this algorithm structure. In other words,FPGA has a great advantage in improving the real-time vision processing system.Based on the analysis of field programmable gate array(FPGA) structure and working mode, this paper points out that FPGA devices is used for image processing, in order to solve the problem of real-time system. This paper designs a real-time image acquisition and processing system based on OV7725 camera, which adopts the ALTERA company’s Cyclone IV as the core processing chip to complete the configuration and the driver of OV7725 camera, image processing and image data to the data processing and real-time display.Compared with the image acquisition and processing system of the system with the traditional,this system can be reconfigured, good real-time performance, high processing speed and flexibility etc.The whole system includes camera driver module, image acquisition module,image pre-processing module, video decoding module, image storage module and image display module. The entire system completes the test in the Quartus II platform. Each functional module of the logic implemented by using Verilog HDL language, especially on the output of OV7725 video stream RGB888 to YCb Cr444 algorithm and YCb Cr422 to RGB888 algorithm HDL implementation, using Modelsim, Signal Tap II software simulation. The results show that image processing can effectively improve the processing speed of hardware platform, satisfy the requirement of real-time image processing, and has higher application value.