中文摘要：

新一代视频编码标准HEVC获得了较高的编码效率，但是同时需要较大的计算量。HEVC并行算法能够提高编码速度，如何开发适用于多核处理器的并行编码算法对于满足高清视频实时传输和大规模共享具有十分重要的意义。提出了一种基于Syntax级分组和多线程处理的HEVC熵编码并行算法。该算法首先将HEVC中一个编码树单元的编码信息按照语法元素进行分组；其次，根据编码块数据间的相关性构建Syntax级并行编码器；然后结合多线程技术实现HEVC帧级编码的并行计算。实验结果表明，在编码图像的主客观质量上没有太大损失的情况下，该并行算法框架与传统的串行算法框架相比具有65%~70%的加速效果。

英文摘要：

As a new generation video coding standard, High Efficiency Video Coding （ HEVC ） achieves higher compression efficiency,but also requires a large amount of calculation. The parallel HEVC encoding algorithm is important to improve the encoding speed,therefore the design of parallel coding algorithm suit-able for multi-core processors is of great significance to meet the requirements of High Density（ HD） video real-time transmission and large-scale sharing. In this paper,a parallel entropy-coding framework based on Syntax-level grouping and multi-thread processing is presented. Firstly,the coding messages of a Cod-ing Tree Unit（ CTU） in HEVC are grouped according to the syntax elements;secondly,the parallel encoder based on Syntax-level is built according to the correlation between the data blocks;finally,frame-level par-allel coding in HEVC is achieved by combining with the multi-threaded calculation. Experimental results show that the encoding speed of the proposed solution is 65% -70% faster than that of the traditional serial architecture,meanwhile the video quality doesn’t lose much.