基于场景检测的自适应Intra帧差错掩盖算法

为了减小视频传输中的错误对解码端重建视频质量的影响，提出了一种基于场景检测的自适应Intra帧差错掩盖算法。该算法利用H.264/AVC的帧内预测模式分布差异检测Intra帧的场景切换，对非场景切换的Intra帧进行空时域掩盖的失真度估算，在此基础上进行自适应的空时域掩盖。实验结果表明，该算法充分利用了Intra帧的时域相关性，与已有的Intra帧差错掩盖算法比较，该算法恢复的视频序列平均峰值信噪比提高了1~2 dB，具有更好的差错掩盖效果。

Adaptive error concealment algorithm based on scene change detection

In order to reduce the influence of transmission errors on the reconstructed image quality in decoder, a novel adaptive error concealment algorithm based on scene change detection for intra frames of H.264/AVC is proposed. The distribution difference of intra prediction mode for H.264/AVC is used to detect scene change. Then distortion estimation is carried out for these intra frames not happened at the moment of scene change. Finally, combined with distortion estima-tion, the new method adaptively selects the spatial-temporal error concealment schemes. Simulation results show that the presented algorithm can use the high correlation of adjacent frames sequence and effectively suppresses error propagation. Compared with other conventional methods, the proposed algorithm can achieve 1~2 dB gain in terms of average PSNR and provides much better error concealment results.