首页 | 本学科首页   官方微博 | 高级检索  
     

面向H.264/AVC的宏块级码率控制算法
引用本文:陈梅丽,刘鹏,姚庆栋.面向H.264/AVC的宏块级码率控制算法[J].浙江大学学报(自然科学版 ),2007,41(6):925-929.
作者姓名:陈梅丽  刘鹏  姚庆栋
作者单位:浙江大学 信息与电子工程学系,浙江 杭州 310027
基金项目:霍英东青年教师奖优选课题资助项目(94031).
摘    要:为了降低计算复杂度和提高H.264/AVC编码器的效率,基于一种相对简单的率失真(R-D)模型提出了新的宏块级码率控制算法,并在新算法中引入了一种自适应方法对目标码率分配和编码复杂度进行调整.相比以往算法在码率分配和参数更新中只利用了宏块绝对平均误差预测值的情况,新算法充分利用了已编码宏块的实际绝对平均误差值和相邻帧之间编码特性的相似性,保证了输出码率的稳定性和解码图像的质量.新算法还采用了一种跳块预判断策略来进一步降低编码复杂度.与JVT-G012码率控制算法相比,新算法在信噪比和输出码率基本不变的条件下,整个编码速度平均提高了约10%.

关 键 词:视频压缩  H.264标准  码率控制  率失真
文章编号:1008-973X(2007)06-0925-05
修稿时间:2006-03-04

Macroblock layer rate control algorithm of H.264/AVC
CHEN Mei-li,LIU Peng,YAO Qing-dong.Macroblock layer rate control algorithm of H.264/AVC[J].Journal of Zhejiang University(Engineering Science),2007,41(6):925-929.
Authors:CHEN Mei-li  LIU Peng  YAO Qing-dong
Affiliation:Department of Information Science and Electronic Engineering, Zhejiang University, Hangzhou 310027, China
Abstract:A new macroblock layer rate control algorithm was proposed based on a relatively simpler rate-distortion(R-D) model in order to reduce the computational complexity and improve the efficiency of H.264/AVC encoder.The algorithm adopted an adaptive method of bit allocation and encoding complexity regulation.Comparing to the previous algorithms used the predicted value of mean of absolute difference(MAD) in bit allocation and parameters update,the new algorithm made full use of the actual MAD and the similarity of encoding mode between two successive frames to ensured the smoothness of the output bit rate and the quality of the decoded picture.A skipped-mode pre-decision method was also proposed to further decrease the computational complexity.Experimental results showed that the proposed algorithm increased the encoding speed by about 10% in average while maintaining the similar peak signal-to-noise ratio(PSNR) and bit rate performance compared with the algorithm adopted in JVT-G012.
Keywords:video compression  H  264 standard  rate control  rate distortion
本文献已被 CNKI 维普 等数据库收录!
点击此处可从《浙江大学学报(自然科学版 )》浏览原始摘要信息
点击此处可从《浙江大学学报(自然科学版 )》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号