H.264中一种基于R-Q模型的自适应码率控制算法研究

doi:10.3969/j.issn.1000-436x.2012.12.005

摘要/Abstract

摘要：

摘要：提出一种新的基于帧复杂度的二级码率控制算法（TSRCA,two-stage rate control algorithm）。总结码率控制算法的最新研究成果，分析码率控制基本原理，继而提出一种二级码率控制算法，以解决“蛋鸡悖论”。其次，用RDO模式选择中的比特数信息代替平均绝对残差（MAD,mean absolute difference）来预测帧的复杂度，提高了预测的准确度。所提出的码率控制算法能够显著提高码率控制精度和编码性能。

关键词: H.264, 码率控制, 帧复杂度

Abstract:

The foundation theory of rate control theory in H.264 video coding standard was analyzed.The latest research work about rate control algorithm were summarized.On the basis of analysis.A novel frame-level rate control algorithm was proposed.Firstly,in order to reduce the inter-dependency between RDO (rate-distortion optimization) and rate control,a kind of two-stage rate control algorithm was presented.Secondly,to provide more accurate prediction of the frame complexity,instead of the MAD (mean absolute difference),bits information in the RDO-based mode decision process was employed to predict the frame complexity.Thirdly,a self-adaptive exponential R-Q model was proposed.The proposed rate control algorithm overcomes the disadvantages in the rate control precision and encoding performance of former algorithms.

Key words: H.264, rate control, frame complexity

郑新资,骆冰清,孙知信. H.264中一种基于R-Q模型的自适应码率控制算法研究[J]. 通信学报, 2012, 33(12): 35-42.

Xin-zi ZHENG,Bing-qing LUO,Zhi-xin SUN. Research on adaptive rate control algorithm for H.264 based on R-Q model[J]. Journal on Communications, 2012, 33(12): 35-42.

图/表 13

图1

图2

图3

图4

表1

图5

图6

图7

图8

图9

图10

表2

表3

参考文献 29

[1]	Draft ITU-T Recommendation and Final Draft International Standard of Joint Video Specification,Document JVT-G050.doc,ITU-T Rec.H.264 \|ISO/IEC 14496-10 AVC[S]. Joint Video Team (JVT) of ISO/IEC MPEG and ITU-T VCEG, 2003.
[2]	TSAI AC , PAUL A , WANG J C ,et al. Intensity gradient technique for efficient intra-prediction in H.264/AVC[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2008,18(5): 694-698.
[3]	Test Model Editing Committee. ISO/IEC JTCI/SC29 WG11/93225b-1993 MPEG-2,Test Model 5[S]. 1993.
[4]	ITU Study Group. ITU-T/SG15-1997 Video Codec Test Model,TMN8[S]. 1997.
[5]	Vedio Group. ISO/IEC JTC1/SC29/WG11-1997 Coding of Moving Pictures and Associated Audio MPEG 97/W1796[S]. 1997.
[6]	ZHANG F , STEINBACH E . Improved ?-domain rate control with accurate header size estimation[A]. Acoustics,Speech and Signal Processing (ICASSP),2011 IEEE International Conference[C]. Prague,Czech, 2011. 813-816.
[7]	LI Z G , PAN F , PANG K ,et al. Adap tive basic unit laye control for JVT:JVT- G012.doc[EB/OL]. , 2006.
[8]	MA S , GAO W , LU Y . Rate-distortion analysis for H.264 /AVC video coding and its app lication to rate control[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2005,15(12): 1533-1544.
[9]	WU Y , LIN S X , ZHANG Y D ,et al. Optimum bit allocation and rate control for H.264/AVC[A]. 15th meeting Busan[C]. Korea, 2005. 16-22.
[10]	CUI Z G , GAN Z L , ZHU X C . Structural similarity optimal MB layer rate control for H.264[A]. Wireless 2011 International Conference,Communications and Signal Processing (WCSP)[C]. Nanjing,China, 2011. 1-5.
[11]	SHIMANO M , CHEUNG G , SATO I . Adaptive frame and QP selection for temporally super-resolved full-exposure-time video[A]. 2011 18th IEEE International Conference Image Processing (ICIP)[C]. Belgium, 2011. 2253-2256.
[12]	WU S J , LIU P L , HUANG Y Q ,et al. On bit allocation and Lagrange multiplier adjustment for rate-distortion optimized H.264 rate control[A]. IEEE International Workshop on Multimedia Signal Processing[C]. Ride Janeiro,Brazil, 2009. 1-6.
[13]	WANG J , YIN H B , ZHOU B Q ,et al. A low-cost MAD prediction algorithm for H.264 rate control facilitating hardware implementation[A]. 2011 4th International Congress on Image and Signal Processing (CISP)[C]. Shanghai,China, 2011. 18-21.
[14]	KIM M J , HONG M C . Adaptive rate control based on statistical modeling in frame-layer for H.264/AVC[A]. 13th International Conference on Advanced Communication Technology (ICACT)[C]. Phoenix Park,Korea, 2011. 680-683.
[15]	HRARTI M , SAADANE H , LARABI M ,et al. A macroblock-based perceptually adaptive bit allocation for H264 rate control[A]. 2010 5th International Symposium I/V Communications and Mobile Network (ISVC)[C]. Las Vegas,Nevada,USA, 2010. 1-4.
[16]	XIE Z , BAO Z , XU C ,et al. Optimal bit allocation and efficient rate control for H64/AVCsed on general rate-istortion model and enhanced coding complexity measure[J]. Image Processing,IET, 2010,4(3): 172-183.
[17]	LEE P J , KUO K T , HUANG S H ,et al. Coding mode determination using fuzzy reasoning in H.264 motion estimation[J]. Signal Processing,IET, 2011,5(2): 242-250.
[18]	CHANG C Y , CHOU C F , CHAN D Y ,et al. A q-domain characteristic-based bit-rate model for video transmission[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2008,18(2): 1307-1311.
[19]	MALEWAR A , SINGH R K , GADRE V . Novel target bit rate estimation for improved frame level H.264 rate control[A]. 2010 International Conference on Signal Processing and Communications (SPCOM)[C]. Bangalore,India, 2010. 1-5.
[20]	DONG J , LING N . A context-adaptive prediction scheme for parameter estimation in H.264/avc macro block layer rate control[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2009,19(5): 1108-1117.
[21]	GRAJEK T , DOMA?SKI M . New model of MPEG-4 AVC/H.264 video encoders[A]. 17th IEEE International Conference on Image Processing (ICIP)[C]. Hongkong,China, 2010. 961-964.
[22]	MALEWAR A , SINGH R K , GADRE V . Novel target bit rate estimation for improved frame level H.264 rate control[A]. 2010 International Conference on Signal Processing and Communications (SP OM)[C]. 2010. 1-5.
[23]	史久根, 许林, 张超 . 一种场景自适应的 H.264 帧层码率控制算法[J]. 计算机应用研究, 2010. SHI J G , XU L , ZHANG C . Scene adaptive frame-lager rate control algorithm for H.264[J]. Application Research of Computers, 2010.
[24]	LI Z G , PAN F , LIM K P ,et al. Adaptive rate control for H.264[A]. Proc IEEE Int Conf Image Process[C]. Singapore, 2004. 449-452.
[25]	KWON D , SHEN M , KUO C C J . Rate control for H.264 video with enhanced rate and distortion models[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2007,17(5): 517-529.
[26]	KAMACI N , ALTINBASAK Y , MERSEREAU R M . Frame bit allocation for the H.264/AVC video coder via cauchy density-based rate and distortion models[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2005,15(8): 994-1006.
[27]	Q042 of JVT,Revised H.264/MPEG-4 AVC Reference Software Manual[S].
[28]	O079 of JVT,Text Description of Joint Model Reference Encoding Methods and Decoding Concealment Methods[S].
[29]	Joint Video Team,JVT reference software encoder,version14.2[EB/OL]. , 2009.

参数	参数设置
测试平台	JM14.2
GOP结构	IPPP
Hadamard变换	on
率失真优化	on
编码帧数	100
搜索范围	16
熵编码模式	CAVLC
参考帧数目	4
运动估计搜索范围	-16~+16
运动估计精度	1/4 pixel
帧率	30frame/s

测试序列	编码算法	编码时间/s	误差平方和
Foreman	JM14.2算法	435.124	26.27
	本文算法	424.326	25.21
Carphone	JM14.2算法	428.391	22.43
	本文算法	309.592	22.29
News	JM14.2算法	435.791	61.53
	本文算法	421.347	59.68

测试序列	编码算法	PSNR/dB	码率/(kbit·s^{- 1})	目标码率/(kbit·s^{- 1})
Foreman	JM14.2算法	28.89	128.27	128
	本文算法	29.12	128.10
Silent	JM14.2算法	40.69	128.56	128
	本文算法	40.89	128.32
Bus	JM14.2算法	28.67	256.97	256
	本文算法	28.69	256.27
Footall	JM14.2算法	33.21	256.79	256
	本文算法	33.54	256.26
bridge	JM14.2算法	36.42	33.41	32
	本文算法	36.79	32.21
Suzie	JM14.2算法	35.25	33.61	32
	本文算法	35.36	32.34