Research And Implementation Of The Error-tolerant Video Coding Algorithm Based On Reference Picture Selection Within X264

Recently, with the rapid development of network technology and the 4th generation mobile communication technology (4G), video communication service has witnessed tremendous popularity so far. But due to the massive information in original video data, direct transmission may place a heavy burden to the communication networks. So it is necessary to remove the redundant data by compressing the original video data, and this method is the so-called video coding technology. The mainstream video coding standard H.264/AVC, which has many advantages such as high compression efficiency, desirable network transmission performance, etc., has been widely employed in various kinds of video communications services. However, as intra/inter frame predictive coding, the variable length entropy coding and other methods are applied during the encoding process, the encoded bit-stream is extremely sensitive to transmission errors. Moreover, once transmission error or packet loss occurred in the transmission process, errors will propagate and spread rapidly in spatial and temporal domain, seriously affecting the quality of the reconstructed video image. In order to improve the anti-error performance of the compressed bit-stream, fault-tolerant video coding technology (error control) has already attracted the attention of many researchers both at home and abroad. This thesis further studies the core reference frame selection algorithm (CRPS) on the basis of preliminary achievements in our laboratory. This thesis firstly implement the CRPS algorithm on the x264 and FFmpeg, and then verify its fault tolerance performances. Finally we further optimizes the CRPS algorithm through the theory of hierarchical coding structure. The main research work is as follows:(1) The codec process of the open source encoder x264 and decoder FFmpeg will be analyzed in details. And then combined with the corresponding stream transmission module, implementation of the CRPS algorithm from the JM test model to x264 and FFmpeg will also be presented.(2) To offset the rate incensement caused by the CRPS algorithm, a further optimization of CRPS algorithm referring to the concept of hierarchical B frame used in HEVC (High Efficiency Video Coding) will be accomplished, and the hierarchical coding structure based CRPS and CKRPS (Core-Key RPS) is obtained as well. Through the hierarchical idea, rate allocation will be more reasonable, the optimized CRPS and CKRPS coding structure can not only reduce the output bit-rate of encoder greatly, but also enhance the anti-error ability of the bit stream significantly.