关于HEVC帧内预测快速模式选择算法的研究

发布时间：2018-05-30 12:45

本文选题：HEVC + 帧内预测　；参考：《南京理工大学》2014年硕士论文

【摘要】：高效率视频编码(HEVC, High Efficiency Video Coding)是由ITU-T和ISO/IEC联合制定的新一代视频编码标准,其目的是为了更显著的提高视频压缩的效率。相比较现有的视频标准H.264/AVC, HEVC能够在保证相同视觉质量的前提下,减少50%左右的比特率。HEVC的帧内预测方向由H.264的9种增加到了35种,使得帧内预测的精度有了显著的提高。然而,大量的预测模式同样也带来了极高的计算复杂度。因此,如何降低HEVC视频编码复杂度,已成为视频编码研究的热点。本文主要研究的是HEVC中的帧内预测算法,具体内容如下： 1)针对HEVC粗略模式选择(RMD,Rough Mode Decision)和率失真优化计算(RDO,Rate Distortion Optimization)过程中候选模式个数较多的问题,本文在分析当前预测块参考像素纹理特性的基础上,利用RMD的最优模式与当前编码块最佳预测模式之间的相关性,提出了一种基于参考像素纹理和RMD最优模式特性的快速模式选择算法,并通过仿真对该算法进行验证。仿真结果表明,本文提出的算法在保证编码性能的前提下,相比较HM10.0中的原始算法,编码时间平均减少了8.71%-36.61%。 2)针对HEVC帧内预测中角度模式的分布结构,本文在此基础上设计了一种新的环形分布结构,该结构更利于阐明角度模式之间的相关性；然后利用最有可能预测模式(MPM,Most Probable Modes)与RMD最优模式之间的相关性,提出了一种基于MPM特性的快速模式选择算法,该算法能够自适应的减少参与RMD和RDO过程的预测模式个数。仿真结果表明,本文提出的算法在保证编码性能的前提下,相比较HM10.0中的原始算法,编码时间平均减少了29.96%-33.51%。
[Abstract]:High efficiency video coding (High Efficiency Video Coding) is a new generation video coding standard developed jointly by ITU-T and ISO/IEC. Its purpose is to improve the efficiency of video compression. Compared with the existing video standard H.264 / AVC, HEVC can reduce the bit rate by about 50% and increase the intra prediction direction from 9 to 35 with the same visual quality. However, a large number of prediction models also bring high computational complexity. Therefore, how to reduce the complexity of HEVC video coding has become a hot spot in video coding research. In this paper, we mainly study the intra prediction algorithm in HEVC. The main contents are as follows: 1) in order to solve the problem of the large number of candidate modes in the process of HEVC rough mode selection and rate-distortion optimization calculation, this paper analyzes the texture characteristics of the reference pixels in the current prediction blocks. Based on the correlation between the optimal mode of RMD and the best prediction mode of the current coding block, a fast pattern selection algorithm based on the reference pixel texture and the optimal pattern characteristics of RMD is proposed, and the algorithm is verified by simulation. The simulation results show that the proposed algorithm can reduce the coding time by 8.71-36.61in comparison with the original algorithm in HM10.0 on the premise of ensuring the coding performance. 2) aiming at the distribution structure of angle mode in HEVC intra prediction, this paper designs a new ring distribution structure, which is more helpful to clarify the correlation between angle modes; Based on the correlation between MPMM-Most Probable modesand RMD optimal pattern, a fast pattern selection algorithm based on MPM characteristics is proposed, which can adaptively reduce the number of prediction modes involved in RMD and RDO processes. Simulation results show that the proposed algorithm can reduce the average coding time by 29.96-33.51 compared with the original algorithm in HM10.0 on the premise of ensuring the coding performance.
【学位授予单位】：南京理工大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN919.81

【参考文献】