| With the development of wireless network and audio and video technology, the application of mobile multimedia plays an important role in social life. In order to utilize limited bandwidth resource efficiently, multimedia should be compressed before being transmitted. At the same time, data compression must satisfy a certain criterion for sharing multimedia information in network. As a digital audio and video coding and decoding standard established and owned by China, AVS, Audio and Video Standard, includes a video standard, AVS-M, specified for mobile multimedia. Products according with AVS-M could be used for broadband network streaming, mobile videoconference, mobile video surveillance, etc. The dissertation work is developed around AVS-M mobile video standard.The AVS-M standard framework is described in detail at first and then comparied with H.264/AVC, the main video standard internationally, for every key component. After that, a new algorithm is provided: rapid intra prediction mode decision algorithm. It is implemented by the sort ascending of the result of SAD in luma block for cost computation and determining a candidate mode set in advance.For optimizing and implementing of AVS-M codec software, detailed analysis is done and then a synthetical encoding optimization scheme combined with rapid algorithms and single instruction multiple data technology is carried out. Simulation results show the scheme is effective. The complexity of decoder is fixed relatively and the optimization is focused on reducing time and space complexity. At the same time, the feasibility of transplanting decoder to embedded operating system is also considered.For spreading AVS-M application, there must have efficient and robust hardware encoder products and the decoder supports mobile terminals as more as possible. So at the end of this dissertation, AVS-M real-time encoding on DSP and decoding software on main operating systems and platform for handsets such as BREW is discussed in detail. |
PDF Full Text Request