Researches On H.264 Encoder Optimization For Real Time Compression

ITU-T H.264/MPEG-4 Part 10 Advanced Video Coding (commonly referred to as H.264/AVC) is the newest international video coding standard. It was developed by a Joint Video Team (JVT) consisting of experts from ITU-T's Video Coding Experts Group (VCEG) and ISO/IEC's Moving Picture Experts Group (MPEG). H.264 is composed of many advanced video coding techniques. Compared to existing video coding standards, H.264 has made great progress in coding efficiency. For example, it improves the compression efficiency by a factor of about two (on average) over MPEG-4. In addition, H.264 provides a very network friendly interface. However, the improvement in compression efficiency of H.264 comes from the dramatic increase in computational complexity, this makes it a must to reduce the computational complexity of algorithms in H.264 in real time applications.Firstly the basic theories for video coding standard and the development of video coding are introduced in this thesis.Secondly, a brief introduction to the main parts of H.264 standard, with focus on the main technology in video coding layer through which the high compression rate of H.264 is achieved.Thirdly, the performance and computational complexity of H.264 encoder is analyzed in detail. The computionally complex part-prediction mode selection is researched deeply.Finally, the fast intra-prediction method is analyzed in detail. Based on the analysis, we proposed a fast mode selection method based on threshold. The threshold can be adjusted adaptively based on both the image texture and the QP value to determine whether the most probable mode can be selected as the optimal mode to end the selection process. The simulation results show that without considerable performance degradation, the proposed method decreases complexity of the intra-prediction mode decision part in H.264, and saves the computing costs greatly.