可伸缩视频码流的熵编码组合优化方法

Combination Optimized Method of Entropy Coding for Scalable Video Bitstream

分析可伸缩视频编码系统中由嵌入式零树编码生成的码流特征,提出一种利用最优预测估计思想将算术编码和游程编码相结合,实现熵编码优化的新方法。实验结果表明,该方法比传统自适应算术编码的效率提高了10%左右,达到了进一步压缩数据的目的。

This paper analyzes features of bitstream generated by Embedded Zerotree Wavelet（EZW） in scalable video coding system, and presents a new method that combines arithmetic coding and run length coding though optimal predictor to optimize entropy coding. Experimental results show that the coding efficiency average increases about 10% by using the new method than using traditional arthmetic, and it obtains the purposes of further data compression.