基于H.263视频通信的容错技术研究
|
摘要
由于通信系统的信道不可避免的存在噪声，传输数据流必然存在误码。对于H.26X的视频编码流，由于空域和时域的压缩方法以及熵编码方法的应用，即使是一个比特的错误，常会影响一个区域的图像不能正确解码，同时错误的图像还会扩散影响后续数帧图像。因此有必要在标准解码程序的基础上，添加容错功能。对于实际的解码系统，所设计的误码掩盖算法，不但要考虑图像恢复的质量，还要考虑算法的效率，不能影响后续帧的解码速度。本课题对前向错误控制、后处理错误掩盖、编解码器端交互的错误掩盖这三大类容错方法进行了详细的研究，分析了现有算法的一些局限性，基于空域插值和时域补偿这两种基本的掩盖算法，提出了一种混合容错算法。该算法的原理为：当运动向量较大时，图像的空间相关性要大于时间的相关性；此时利用空间相关性进行掩盖就比利用时间相关性掩盖好。这种方法的准则是：如果检测出的错误宏块的运动向量的水平分量或者垂直分量大于某个阈值时，则使用空域插值方法进行错误掩盖，反之则利用时域补偿方法进行掩盖。
课题在开放的H.263测试模型TMN8中实现了该算法，实现过程包括误码的检测，定位和误码部分数据的恢复等工作，并与标准H.263的结果进行了比较，模拟结果表明，经检错和掩盖之后，解码图像质量有了显著的改善，证明了算法的正确性和实用性。
目前在国际上，移动视频通信和网络视频通信已经成为许多公司追求的未来电信市场目标，而容错能力研究又是多媒体通信特别是移动视频通信的研究热点之一，因此，本文所做的工作是具有一定的实用价值的。
RESEARCH OF ERROR RESILIENCE
TECHNOLOGY FOR VIDEO COMMUNICATION BASED ON H.263
CODEC
Major: Signal and Information Processing Data: Author: Meng Yan Signature:
Tutor: Zhang Erhu Liu Shichang Signature:
ABSTRACT
The reconstructed image quality can be degraded substantially by errors during the transmitting over the communication channel, especially when the video is compressed by temporal and spatial coding ways, which leads to error propagation in frame sequence, so the error resilience procedure should be performed to improve the reconstruction quality. This thesis discusses the research status and typical algorithms of error resilience at home and abroad, whose performance are analyzed, too. The error resilience technology means that the decoder can detect errors, skip the inextricable errors, and rectify the errors as soon as possible, one important way of which is error concealment. Recovering the video quality by temporal compensation and spatial interpolation is the basic principle, and the error concealing process includes the detecting, and the binding, and the synchronizing, and the data recovering. Based on the one and a half year' s lucubration, the author gets one new error concealment way with better combination between efficiency and running time in processing the errors in video communication.
According to the experiment results and relative data analysis, we can get a certain conclusion, which is that the algorithms are correct and useful. At present, mobile video communication and IP video communication are future market goals for most companies in the communication field. At the same time, the error resilience technology is one of the keys in realizing a video communication system. So, you can see, the work of this thesis is meaningful and valuable.
课题在开放的H.263测试模型TMN8中实现了该算法，实现过程包括误码的检测，定位和误码部分数据的恢复等工作，并与标准H.263的结果进行了比较，模拟结果表明，经检错和掩盖之后，解码图像质量有了显著的改善，证明了算法的正确性和实用性。
目前在国际上，移动视频通信和网络视频通信已经成为许多公司追求的未来电信市场目标，而容错能力研究又是多媒体通信特别是移动视频通信的研究热点之一，因此，本文所做的工作是具有一定的实用价值的。
RESEARCH OF ERROR RESILIENCE
TECHNOLOGY FOR VIDEO COMMUNICATION BASED ON H.263
CODEC
Major: Signal and Information Processing Data: Author: Meng Yan Signature:
Tutor: Zhang Erhu Liu Shichang Signature:
ABSTRACT
The reconstructed image quality can be degraded substantially by errors during the transmitting over the communication channel, especially when the video is compressed by temporal and spatial coding ways, which leads to error propagation in frame sequence, so the error resilience procedure should be performed to improve the reconstruction quality. This thesis discusses the research status and typical algorithms of error resilience at home and abroad, whose performance are analyzed, too. The error resilience technology means that the decoder can detect errors, skip the inextricable errors, and rectify the errors as soon as possible, one important way of which is error concealment. Recovering the video quality by temporal compensation and spatial interpolation is the basic principle, and the error concealing process includes the detecting, and the binding, and the synchronizing, and the data recovering. Based on the one and a half year' s lucubration, the author gets one new error concealment way with better combination between efficiency and running time in processing the errors in video communication.
According to the experiment results and relative data analysis, we can get a certain conclusion, which is that the algorithms are correct and useful. At present, mobile video communication and IP video communication are future market goals for most companies in the communication field. At the same time, the error resilience technology is one of the keys in realizing a video communication system. So, you can see, the work of this thesis is meaningful and valuable.
引文
[1]黄孝建等．数字图像通信．北京：人民邮电出版社，1998．
[2]F.Kuo,W.Effelsberg,et al.著，龙晓苑译．多媒体通信协议与应用．北京：清华大学出版社，1997．
[3]Nelson.M，贾起东．数据压缩技术原理与范例．北京：科学出版社，1995．
[4]J.G.Proakis. Digital Communnication.北京：电子工业出版社(影印)，1998．
[5]T.S.Rappaprot. Wireless Communication.北京：电子工业出版社(影印)，1998．
[6]ITU-T Recommendation H.263 Version 2-1997, Video Coding for Low Bit Rate Communication.
[7]宋彬，常义林．基于H.263的甚低比特率视频软件编码器．西安电子科技大学学报，2期，2000．
[8]杨成军，全子一．H.263的数学模型及性能研究．电视技术，3期，2001．
[9]WANG Y, Zhu Q F. Error control and concealment for video communication. Proceedings of the IEEE, 5(86),1998.
[10]Wenwu Zhu, Yao Wang, Qin-Fan zhu. Second-Order Deriviative-Based Smoothness Measure for Error Concealment in DCT-Based Codecs. IEEE Transactions on Circuits and Systems for Video Technology. 6(8), 1998.
[11]张俊，余松煜，张文军．HDTV中MPEG-2实时解码器的错误处理实用新算法研究．电视技术，8期，2000．
[12]ISO/IEC 13813-1994,MPEG-2 Standard.
[13]罗明辉，朱秀昌，王厚大．基于H.263视频解码的误码掩盖技术研究．通信学报．5(20)，1999．
[14]Frank H.P. Span, Reginald L.Lagendijk,Jun Biemond, Error Robust Video Coding Based on H.263, International Conference Image Processing,1998.
[15]Jian Zhang, John F. Arnold. A Cell-Loss Concealment Technique for MPEG-2 Coded Video. IEEE Transactions on Circuits and Systems for Video Technology. 4(10), 2000.
[16]Ghanbari M, Seferidis V. Cell-loss Concealment in ATM Video Codecs.IEEE Trans on CSVT, 3(3), 1993.
[17]LAM W M, REIBMAN A R. An error concealment algorithm for images subject to channel errors. IEEE Trans on Image Processing. 5(4), 1995.
[18]LAM W M, REIBMAN A R. Self-synchronizing variable length codes for image transmission. Proc ICASSP. 3, 1992.
[19]高绍帅，涂国防．基于H.263视频解码的错误检测和掩盖．通信学报．12期，2001．
[20]曾玉净，张家谋．极低码率视频压缩编码算法的改进方法．通信学报．12期，2000．
[21]魏政刚．图像质量评价方法的历史、现状和未来．中国图像图形学报，3(5)，1998．
[22]陆杰，赵忠旭．图像质量评价的发展．计算机工程．11期，2000．
[23]陶振宁，吴德明．极低误码下的误码率估计问题．通信学报，7期，2000．
[24]Xu Wen. Picture Quality Evaluation Based on Error Segmentation.SPEIE, 1994.
[25]胡琳蓉，朱秀昌．一种适用于H.263的运动估计搜索算法．通信学报，6期，2000．
[26]郑磊，易波等．GSTN上多媒体通信终端的设计和实现．无线电通信技术，6(26)，2000．
[27]高剑林，余胜生，周敬利．H．263+中新的帧类型的研究．电视技术，(5)，1999．
[28]李建新等．现代通信系统分析与仿真．西安：西安电子科技大学出版社，2000．
[29]Eskiciogln A M. Image Quality Measures and Their Performance. IEEE Trans.Comm, 43(12), 1995.
[30]Align S,Fazel L. Temporal and Spatial Error Concealment Techniques for Hierarchical MPEG-2 Video Codec. In:Proc IEEE Int Conf on Commun,3期，1995.
[31]Kwok W, Sun H F. Multi-directional Interpolation for Spatial Error Concealment. IEEE Trans on Consumer Electronics, 39(3), 1993.
[32]Park J W et al. DCT Coeffcients Recovery-based Error Concealment Technique and Its Application to the MPEG-2 Bit Stream Error. IEEE Trans on CSVT, 7(6), 1997.
[33]Lam W M, Reibmant A R, Liu B. Recovery of Lost or Erroneously Received Motion Vectors. In:Proc IEEE ICASSP, 5期，1993.
[34]http://www.itu.int.
[2]F.Kuo,W.Effelsberg,et al.著，龙晓苑译．多媒体通信协议与应用．北京：清华大学出版社，1997．
[3]Nelson.M，贾起东．数据压缩技术原理与范例．北京：科学出版社，1995．
[4]J.G.Proakis. Digital Communnication.北京：电子工业出版社(影印)，1998．
[5]T.S.Rappaprot. Wireless Communication.北京：电子工业出版社(影印)，1998．
[6]ITU-T Recommendation H.263 Version 2-1997, Video Coding for Low Bit Rate Communication.
[7]宋彬，常义林．基于H.263的甚低比特率视频软件编码器．西安电子科技大学学报，2期，2000．
[8]杨成军，全子一．H.263的数学模型及性能研究．电视技术，3期，2001．
[9]WANG Y, Zhu Q F. Error control and concealment for video communication. Proceedings of the IEEE, 5(86),1998.
[10]Wenwu Zhu, Yao Wang, Qin-Fan zhu. Second-Order Deriviative-Based Smoothness Measure for Error Concealment in DCT-Based Codecs. IEEE Transactions on Circuits and Systems for Video Technology. 6(8), 1998.
[11]张俊，余松煜，张文军．HDTV中MPEG-2实时解码器的错误处理实用新算法研究．电视技术，8期，2000．
[12]ISO/IEC 13813-1994,MPEG-2 Standard.
[13]罗明辉，朱秀昌，王厚大．基于H.263视频解码的误码掩盖技术研究．通信学报．5(20)，1999．
[14]Frank H.P. Span, Reginald L.Lagendijk,Jun Biemond, Error Robust Video Coding Based on H.263, International Conference Image Processing,1998.
[15]Jian Zhang, John F. Arnold. A Cell-Loss Concealment Technique for MPEG-2 Coded Video. IEEE Transactions on Circuits and Systems for Video Technology. 4(10), 2000.
[16]Ghanbari M, Seferidis V. Cell-loss Concealment in ATM Video Codecs.IEEE Trans on CSVT, 3(3), 1993.
[17]LAM W M, REIBMAN A R. An error concealment algorithm for images subject to channel errors. IEEE Trans on Image Processing. 5(4), 1995.
[18]LAM W M, REIBMAN A R. Self-synchronizing variable length codes for image transmission. Proc ICASSP. 3, 1992.
[19]高绍帅，涂国防．基于H.263视频解码的错误检测和掩盖．通信学报．12期，2001．
[20]曾玉净，张家谋．极低码率视频压缩编码算法的改进方法．通信学报．12期，2000．
[21]魏政刚．图像质量评价方法的历史、现状和未来．中国图像图形学报，3(5)，1998．
[22]陆杰，赵忠旭．图像质量评价的发展．计算机工程．11期，2000．
[23]陶振宁，吴德明．极低误码下的误码率估计问题．通信学报，7期，2000．
[24]Xu Wen. Picture Quality Evaluation Based on Error Segmentation.SPEIE, 1994.
[25]胡琳蓉，朱秀昌．一种适用于H.263的运动估计搜索算法．通信学报，6期，2000．
[26]郑磊，易波等．GSTN上多媒体通信终端的设计和实现．无线电通信技术，6(26)，2000．
[27]高剑林，余胜生，周敬利．H．263+中新的帧类型的研究．电视技术，(5)，1999．
[28]李建新等．现代通信系统分析与仿真．西安：西安电子科技大学出版社，2000．
[29]Eskiciogln A M. Image Quality Measures and Their Performance. IEEE Trans.Comm, 43(12), 1995.
[30]Align S,Fazel L. Temporal and Spatial Error Concealment Techniques for Hierarchical MPEG-2 Video Codec. In:Proc IEEE Int Conf on Commun,3期，1995.
[31]Kwok W, Sun H F. Multi-directional Interpolation for Spatial Error Concealment. IEEE Trans on Consumer Electronics, 39(3), 1993.
[32]Park J W et al. DCT Coeffcients Recovery-based Error Concealment Technique and Its Application to the MPEG-2 Bit Stream Error. IEEE Trans on CSVT, 7(6), 1997.
[33]Lam W M, Reibmant A R, Liu B. Recovery of Lost or Erroneously Received Motion Vectors. In:Proc IEEE ICASSP, 5期，1993.
[34]http://www.itu.int.