Research On Video Identification System Based On ARM9

the current development of information, as information carriers ,such as audio and video, is playing an increasingly important role in all areas of industrial production. As the massive amount of information about video is a constraint to its development , one of the key about video research is the how to improve the efficiency of video compression without affecting the video quality, H.264 is the latest in today's international video data compression standard, the compression efficiency is the highest in the similar image quality, the nearly 1-fold increase than H.263 and MPEG4, which greatly increases the efficiency of video compression.H.264 compression performance increase, at the same time in raising the price is the algorithm complexity, the practical application of it demand for the real-time video decoding, there has been many algorithms about general processors based-PC nowdays ,but the frequency of embedded devices ,more limited processing ability, storage capacity is relatively small and the bus speed is relatively low, so we need for code optimization transplantation in order to meet real-time applications.The H.264 standard and its new features are introduced In this paper, the study about the improvement of the time-consuming for modules at the decoder, algorithms ported to ARM platform and optimized for the platform-specific ,it complete decoding video image display and it completes the transmission of H.264 video format in WIN32 environment and gives test results.In the current field of intelligent video surveillance, intelligent object recognition for mobile technology has gradually matured, this paper carried out simulation under WIN32 for the technology using opencv , perfect function of the entire intelligence system was simulated and the results obtained.In this research, Information Industry Co., Ltd. Wuhan East horizontal issues too, "H.264-based video surveillance system " the background of the research project, now completed testing about each module, the test module display movement and stability, the index has reached the desired request.