Design Of Video Detecting System Based On FPGA

ABSTRACT:With the development of science and technology, and ever-increasing demand for video detecting system, there is a higher demands for the miniaturization of video detecting system. Embedded video detection system has the advantages of smaller size, better adaptability to environment, and is cable of processing a huge amount of data with high-speed, which is beneficial to the portable design and miniaturization. And thus Embeded video detection system becomes the future development trend in this field.First, the implementation of embedded video detection system is analyzed in this thesis, and then a video detecting scheme is developed,with the pre-processing based on FPGA.The image pre-processing system is designed to realize the front-end video acquisition, image preprocessing, and a high definition display of image acquisition via HDMI.The image pre-processing system based on FPGA designed in this paper takes Altera Cyclone III device as the core processor. Internal FPGA logic includes video data collection and decoding module,frame storage controller for image module and output buffer module, which can realize the cache and format transformation of digital video datastream,finish Ping-pong access to double SDRAMs, and ensure the integrity of image data.The A/D converter SAA7113 digitizing the raw analog video signal and transmits the digital video data into the FPGA for the pre-processing; After that RGB data was read from FPGA and encoded by TFP410, and then the encoded signal was transmitted to the HDMI to be displayed in real time,which provides a essential foundation to the further and advanced processing.As for the software development, ModelSim/AE6.3 and QuartusII are chosen as the simulating tool and Synplify the synthesis tool. According to the simulation and experimental results,the proposed image pre-processing system based on FPGA, partial of a embeded video detection, which provides high reliability and stability,is proved that it easy to upgrade and maintenance, and can meet the real-time requirements for a video detection system.