用于IP网络的差错复原编解码器

随着因特网的飞速发展，基于IP网络的视频通信越来受到重视，但由于IP网络属于易于发生差错的信道，因此，基于IP网络的视频业务必须要到信道差错的影响，为了抗信道差错，差错复原技术已成为易发生差错信道下视频编码的重视组成部分，针对IP网络时常发生拥塞和带宽经常变动而造成的数据分组丢失，时延和抖动问题，提出了用于IP网络的差错复原技术综合应用解决方案，该方案通过综合运用多种差错原复技术，大大提高了压缩码流的抗差错能力和解码器和差错复原能力，同时应用该方案实现了具有差错复原性能的H．263、H．26L视频编码解器。利用Internet网络模型对编解码器的差错复原能力进行了实验，实验结果表明，该视频编解码器具有良好的差错复原性能，重图象的主客观质量得到了明显的改善。

Error Resilience Codec for IP-Based Video Delivery

IP based video communication is becoming increasingly important in recent years. One inherent problem of IP based video communication is channel errors. Current video compression standards achieve efficient compression by using predictive coding and variable length entropy coding. Variable length coding techniques are highly vulnerable to channel errors. The channel errors cause the decoder to lost synchronization with the decoder. Predictive coding techniques make the errors quickly propagate across the entire video sequence. So the effect of these errors not only can cause great degradation of the quality of reconstructed images, even can lead to the failure of the whole video communication system. Error resilience techniques have been employed to reduce the effect of the channel errors. The syntax and the structure of compressed video stream have been modified to get error resilient video stream. So the decoder can much more easily detect errors and to recover the lost data from the errors. Error resilience techniques also use the spatial and the temporal correlations within the video sequence to reduce the degradation of the quality of reconstructed images.This article presents an integrated resolving scheme of error resilience technologies for IP based video delivery. This scheme employs severel error resilience technologies, such as the slice mode, resynchronization and error concealment, etc, it can greatly increase the error resilient capability of the video codec. We also implement error resilient H 263 and H 26L codec based on this scheme. This article also gives some simulation results of error resilience technologies based on Internet models. The result shows error resilience technologies can remarkably improve the subjective and objective quality of the reconstructed image.