基于小波域的视频压缩编码
|
摘要
目前，许多实用的图像编码算法都是基于空间域的运动估计和补偿、预测误差的DCT及量化以及变换系数的熵编码的混合编码方法，如有关静止图像和视频图像的国际标准JPEG，MPEG-1，MPEG-2等。但基于DCT的算法有其固有的缺点，即方块效应，在压缩比较高时，图像质量会很差，因此人们一直在努力研究寻找更为有效的编码方法。小波变换具有良好的空间一频率局域化等特性，非常适合描述非平稳图像信号，适应人的视觉系统特性，从而在视频编码领域受到越来越多的关注。
本文在介绍了视频编码基本原理的基础上，重点研究了基于小波域的视频编码系统的工作原理，分析了小波域的多分辨率运动估计和小波域的图像编码方法，总结了小波图像编码的经典算法，并对它们进行了性能比较，分析了多分辨率运动补偿后预测误差系数的分布特点，对量化后的预测误差系数的组合进行了改进，使之能更有效地编码，从而提高编码效率。编程实现了两个完整的编解码系统：传统的基于空间域运动估计、DCT变换的视频编码和基于小波域多分辨率运动估计的视频编码。论文还对利用三维小波变换进行视频编码以及小波变换存在的移变性进行了分析讨论，并介绍了克服这种移变性的低子带位移运动估计的方法，探索性地进行了计算机仿真。
计算机仿真结果表明，基于小波域的视频图像编解码系统无论从编码效率，还是从解码后的图像主观质量比基于空间域的编解码系统都有了一定程度的提高。
Recently, many applied image coding algorithms are mixed coding methods based on the spatial motion estimation and compensation, the DCT and quantification of displaced frame difference(DFD) and entropy coding such as some international standards about still images and video compression coding:JPEG,MPEG-1, MPEG-2 and so on. But the algorithms based on DCT have their inherent disadvantage that is "blocking effect". The quality of the restored images is too bad at the high compression ratio. So much more effective coding methods have been gone for. Having the good characteristics of temporal-frequency localization and so on, wavelet transform is very suitable for describing nonstationary signal and adapt to man's vision system. Thereby wavelet transform is increasingly concerned in the field of video compression coding.
In this thesis, the author emphatically discusses the theory of the video coding system based on the wavelet field following the basic theory of video coding. Then the multi-resolution motion estimation and image compression coding based on wavelet and are described and analyzed respectively. Make a summary and performance compare of the classical algorithms of image compression. Coding the displaced frame difference (DFD) which is the coefficient difference of the reference frame and the frame formed by multi-resolution motion compensation. The coefficients of DFD are recomposed according to their distributing traits in order to make them coded efficiently and increase the compression ratio. Programming two whole codec system one of which is conventional coding system based on spatial motion estimation and DCT, and the other is based on multi-resolution motion estimation in wavelet field. The shift-variant property of the wavelet transform and three-dimension wavelet transform are also discussed in the thes
is. And simulation experiment has been made exploringly for the motion estimation using Low-Band-Shift(LBS) method aim at overcoming the temporal variance of the wavelet transform.
The simulation results indicate that the performance of the video codec system based on wavelet field are superior to that based on spatial field not only in coding efficiency but in subjective quality of the decoded image.
本文在介绍了视频编码基本原理的基础上，重点研究了基于小波域的视频编码系统的工作原理，分析了小波域的多分辨率运动估计和小波域的图像编码方法，总结了小波图像编码的经典算法，并对它们进行了性能比较，分析了多分辨率运动补偿后预测误差系数的分布特点，对量化后的预测误差系数的组合进行了改进，使之能更有效地编码，从而提高编码效率。编程实现了两个完整的编解码系统：传统的基于空间域运动估计、DCT变换的视频编码和基于小波域多分辨率运动估计的视频编码。论文还对利用三维小波变换进行视频编码以及小波变换存在的移变性进行了分析讨论，并介绍了克服这种移变性的低子带位移运动估计的方法，探索性地进行了计算机仿真。
计算机仿真结果表明，基于小波域的视频图像编解码系统无论从编码效率，还是从解码后的图像主观质量比基于空间域的编解码系统都有了一定程度的提高。
Recently, many applied image coding algorithms are mixed coding methods based on the spatial motion estimation and compensation, the DCT and quantification of displaced frame difference(DFD) and entropy coding such as some international standards about still images and video compression coding:JPEG,MPEG-1, MPEG-2 and so on. But the algorithms based on DCT have their inherent disadvantage that is "blocking effect". The quality of the restored images is too bad at the high compression ratio. So much more effective coding methods have been gone for. Having the good characteristics of temporal-frequency localization and so on, wavelet transform is very suitable for describing nonstationary signal and adapt to man's vision system. Thereby wavelet transform is increasingly concerned in the field of video compression coding.
In this thesis, the author emphatically discusses the theory of the video coding system based on the wavelet field following the basic theory of video coding. Then the multi-resolution motion estimation and image compression coding based on wavelet and are described and analyzed respectively. Make a summary and performance compare of the classical algorithms of image compression. Coding the displaced frame difference (DFD) which is the coefficient difference of the reference frame and the frame formed by multi-resolution motion compensation. The coefficients of DFD are recomposed according to their distributing traits in order to make them coded efficiently and increase the compression ratio. Programming two whole codec system one of which is conventional coding system based on spatial motion estimation and DCT, and the other is based on multi-resolution motion estimation in wavelet field. The shift-variant property of the wavelet transform and three-dimension wavelet transform are also discussed in the thes
is. And simulation experiment has been made exploringly for the motion estimation using Low-Band-Shift(LBS) method aim at overcoming the temporal variance of the wavelet transform.
The simulation results indicate that the performance of the video codec system based on wavelet field are superior to that based on spatial field not only in coding efficiency but in subjective quality of the decoded image.
引文
[1] ITU-T Recommendation H. 261. Video Codec for audiovisual servicees at p×64kbit/s. Mar.1993.
[2] ITU-T Recommendation H. 263. Video CodING for low rate. May. 1996.
[3] G.K. Wallace. The JPEG still picture compression standard. Commun. ACM Commun. ACM, 1991 Vol. 34, pp:30-44.
[4] MPEG-1 Committee Draft. ISO/IEC JTC1/SC29/WG11. ISO/IEC CD11172:Information technology, Dec. 1991.
[5] MPEG-2 Committee Draft. ISO/IEC JTC1/SC29/WG11.ISO/IEC CD13818:Information technology, Dec. 1991.
[6] MPEG-4 Video verification model version 14.0. ISO/IEC JTC1/SC29/WG111 n2932.1999:301-303.
[7] ISO/IEC JTC1/SC29/WG11. MPEG-7:Content and Objects(versionl1.0 Atlantic City).
[8] A. Murat Tekalp. Digital Video Processing.北京：清华大学出版社,1998
[9] J. Biemond, L. Looijenga, D. Boekee, and R.H.H.M. Plompen. A pel-recursive Viener-based displacement estimation algorithm. Sign. Proc., Dec. 1987, vol. 13, pp:399-412.
[10] Y.M. Erkam, M.I. Sezan, and A.T. Erdem, A hierarchical phase-correlation method for motion estimation. Proc. Conf. on Info. Scien. and Systems, Mar. 1993, pp: 419-424.
[11] W. Enkelmann. Investigations of multigrid algorithms for thesetimation of optical flow fields in image sequences. Comp. Vis. Graph Image Proc., 1988, vol. 43, pp:150-177.
[12] R. Abut. Full search and tree searched vector quantization of speech waveforms Gray. Acoustics, Speech, and Signal Processing, IEEE International Conference on ICASSP '82, May. 1982, Vol. 7, pp:593-596.
[13] J. Jain and A. Jain. Displacement measurement and its application in interframe image coding. IEEE Trans. Commum, 1981,29(12):1799-1808.
[14] T. Koga, K. Linuma, A. Hirano, Y. Lijima, and T. Ishiguro. Motion compensated interframe coding for video vonferencing, in Proc. NTCS1, Dec. 1981, pp:C9.6.1-9.6.5.
[15] M. Ghanbari. The cross-search algorithm for motion estimation. IEEE Trans. Commun. 1990 38(7), pp: 950-953.
[16] 许悦雷，毕笃彦，毛柏鑫．一种遗传搜索块运动估计算法．中国图形图像学报，Feb．2001，V01．6(A)，No．2．
[17] 洪波，庄健敏，余松煜．基于时空相关性的自适应运动估计．计算机工程，Aug．1998，26(8)：51-53．
[18] Halek S O, Sauer K..Temporal prediction of Block Motion Vectors with Local Ambiguity
-based Adaptivity. Proc Spie Visual Command Image Process, Boston, 1994 pp:1818-1825.
[19] S. Zhu and K.K. Ma.A new diamond search algorithm for fast block-matching motion estimation. IEEE Trans. On Image Processing, 2000,29(2):287-290.
[20] R. Li, B. Zeng, M.L. Liou. A new three_step search algorithm for block motion estimation. IEEE Trans. CASVT. 1994,8,4(8):438-442.
[21] L.M. Po and W.C. Ma.A novel four-step search algorithm for fast block motion estimation. IEEE Trans. CASVT. 1996,6,6(6):313-317.
[22] Chow K H. K, Liou M L. Genetic motion search algorithm for video compression. IEEE Trans. Circuits Syst. Video Technol.,1993,3:440-445.
[23] Lin Chun-Hung, Wu Ja-Ling. A lightweight genetic block-matching algorithm for video coding. IEEE Trans. CircuitsSyst. Video Technol.,1998,8:386-392.
[24] Liu L K, Feig E.A block-based gradient descent search algorithm for block motion estimat-ion in video coding. IEEE Trans. CASVT, 1996,8,6(4):419-422.
[25] S. Zhu and K.K. Ma.A new diamond search algorithm for fast block-matching motion estimation. IEEE Trans. Image Processing. 2000,2,9(2):287-290.
[26] Draft ITU-T Recommendation H. 264(a.k.a "H. 26L").Joint video Team(JVT) of ISO/IEC MPEG and ITU-T VCEG, Document JVT-C167. May 2002.
[27] Chui C K. An introduction to wavelet. Academic Press, 1992.
[28] Daubechies I. The wavelet transform, time frequency localization and signal annalsys. IEEE Trans Information theory, 1990; 36(5):961-1005.
[29] S G. Mallat. A theory for multiresolution signal decomposition: the wavelet representat-ion. IEEE Trans Pattern Analysis and Machine Intelligence, 1989,11(7):674-693.
[30] 赵松年，熊小芸．子波变换和子波分析．电子工业出版社，1997．
[31] 吴乐南．数据压缩．电子工业出版社，2000．
[32] Jinwen Zan, Swamy, M.N.S.,Ahmad, M.O..Wavelet Filters in Multi-resolution Motion Estim-ation. Electrical and Computer Engineering, 2001, Vol:2, pp:1321-1326.
[33] J M. Shaprio. Embedded image coding using zerotree of wavelet coefficient. IEEE Trans. On Signal Processing, 1993;41(12):3445-3462.
[34] 章毓晋．图象处理和分析．清华大学出版社，2001．
[35] Saind A, Pearlman W.A new fast and efficient image codec based on set partitioning in hierarchical trees. IEEE Trans on Circuits and Systems for Video Technology 1996,6 (3):243-249.
[36] Y. Zhang and S. Zafar. Motion-compensated wavelet transform coding for color video compression. IEEE Trans On Circuits and System for Video technology Sept. 1992, vol. 2, pp:
285-296.
[37] Seongman Kim, Seunghryon Rhee, Jun Geun Jeon and Kyu Tae Park. Interframe Coding Using Two-Stage Variable Block-Size Multiresolution Motion Estimation and Wavelet Decompos-ition. IEEE Trans On Circuits and System for Video Technology, 1998;8(4):399-410.
[38] Tingrong Zhao, Tatsuo Ohtsuki. A fast MRME scheme for video compression based on coefficients masking in the wavelet domain. TENCON 99. Proceedings of the IEEE Region 10 Conference, 1999, Vol. 1, pp:127-130.
[39] Hyun-Wook Park, Hyung-Sun Kim. Motion estimation using low-band-shift method for wavele-t based moving-picture coding, IEEE Transactions on Image Processing, 2000, Vol. 9, No. 4, pp:577-587.
[40] Van der Auwera, G. Munteanu, A. Lafruit, G. Cornelis. Video coding based on motion estimation in the wavelet detail images. Acoustics, Speech, and Signal Processing, 1998. ICASSP'98. Proceedings of the 1998 IEEE International Conference on, May 1998 Vol. 5, pp:2801-2804.
[41] Podilchuk Christine I, Jayant Nikil S, Nariman Farvardin. Three Dimentional Subband Coding of Video. IEEE Trans. Image Processing, 1995,4(2):125-139.
[42] Chang Y, Pearlman W.A. Three-dimensional subband coding using the zero-tree method, Proc SPIE, Mar, 1997:1302-1312.
[43] Kim B J, Pearlman W.A. An embedded wavelet video coder using three-dimensional set Partitioning in hierarchical trees(SPIHT).Proc Data Compression Conf, 1997:251-260.
[44] 林刚，刘泽民．基于运动补偿的三维小波视频编码．中国图象图形学报，Nov，1998，Vol．3，pp：926-929
[45] Martuci S A, SodagarI, ChiangT, Zhang Y Q.A zerotree wavelet videocoder. IEEE Trans on circuits and systems for video technology, 1997;7(1):85-94.
[2] ITU-T Recommendation H. 263. Video CodING for low rate. May. 1996.
[3] G.K. Wallace. The JPEG still picture compression standard. Commun. ACM Commun. ACM, 1991 Vol. 34, pp:30-44.
[4] MPEG-1 Committee Draft. ISO/IEC JTC1/SC29/WG11. ISO/IEC CD11172:Information technology, Dec. 1991.
[5] MPEG-2 Committee Draft. ISO/IEC JTC1/SC29/WG11.ISO/IEC CD13818:Information technology, Dec. 1991.
[6] MPEG-4 Video verification model version 14.0. ISO/IEC JTC1/SC29/WG111 n2932.1999:301-303.
[7] ISO/IEC JTC1/SC29/WG11. MPEG-7:Content and Objects(versionl1.0 Atlantic City).
[8] A. Murat Tekalp. Digital Video Processing.北京：清华大学出版社,1998
[9] J. Biemond, L. Looijenga, D. Boekee, and R.H.H.M. Plompen. A pel-recursive Viener-based displacement estimation algorithm. Sign. Proc., Dec. 1987, vol. 13, pp:399-412.
[10] Y.M. Erkam, M.I. Sezan, and A.T. Erdem, A hierarchical phase-correlation method for motion estimation. Proc. Conf. on Info. Scien. and Systems, Mar. 1993, pp: 419-424.
[11] W. Enkelmann. Investigations of multigrid algorithms for thesetimation of optical flow fields in image sequences. Comp. Vis. Graph Image Proc., 1988, vol. 43, pp:150-177.
[12] R. Abut. Full search and tree searched vector quantization of speech waveforms Gray. Acoustics, Speech, and Signal Processing, IEEE International Conference on ICASSP '82, May. 1982, Vol. 7, pp:593-596.
[13] J. Jain and A. Jain. Displacement measurement and its application in interframe image coding. IEEE Trans. Commum, 1981,29(12):1799-1808.
[14] T. Koga, K. Linuma, A. Hirano, Y. Lijima, and T. Ishiguro. Motion compensated interframe coding for video vonferencing, in Proc. NTCS1, Dec. 1981, pp:C9.6.1-9.6.5.
[15] M. Ghanbari. The cross-search algorithm for motion estimation. IEEE Trans. Commun. 1990 38(7), pp: 950-953.
[16] 许悦雷，毕笃彦，毛柏鑫．一种遗传搜索块运动估计算法．中国图形图像学报，Feb．2001，V01．6(A)，No．2．
[17] 洪波，庄健敏，余松煜．基于时空相关性的自适应运动估计．计算机工程，Aug．1998，26(8)：51-53．
[18] Halek S O, Sauer K..Temporal prediction of Block Motion Vectors with Local Ambiguity
-based Adaptivity. Proc Spie Visual Command Image Process, Boston, 1994 pp:1818-1825.
[19] S. Zhu and K.K. Ma.A new diamond search algorithm for fast block-matching motion estimation. IEEE Trans. On Image Processing, 2000,29(2):287-290.
[20] R. Li, B. Zeng, M.L. Liou. A new three_step search algorithm for block motion estimation. IEEE Trans. CASVT. 1994,8,4(8):438-442.
[21] L.M. Po and W.C. Ma.A novel four-step search algorithm for fast block motion estimation. IEEE Trans. CASVT. 1996,6,6(6):313-317.
[22] Chow K H. K, Liou M L. Genetic motion search algorithm for video compression. IEEE Trans. Circuits Syst. Video Technol.,1993,3:440-445.
[23] Lin Chun-Hung, Wu Ja-Ling. A lightweight genetic block-matching algorithm for video coding. IEEE Trans. CircuitsSyst. Video Technol.,1998,8:386-392.
[24] Liu L K, Feig E.A block-based gradient descent search algorithm for block motion estimat-ion in video coding. IEEE Trans. CASVT, 1996,8,6(4):419-422.
[25] S. Zhu and K.K. Ma.A new diamond search algorithm for fast block-matching motion estimation. IEEE Trans. Image Processing. 2000,2,9(2):287-290.
[26] Draft ITU-T Recommendation H. 264(a.k.a "H. 26L").Joint video Team(JVT) of ISO/IEC MPEG and ITU-T VCEG, Document JVT-C167. May 2002.
[27] Chui C K. An introduction to wavelet. Academic Press, 1992.
[28] Daubechies I. The wavelet transform, time frequency localization and signal annalsys. IEEE Trans Information theory, 1990; 36(5):961-1005.
[29] S G. Mallat. A theory for multiresolution signal decomposition: the wavelet representat-ion. IEEE Trans Pattern Analysis and Machine Intelligence, 1989,11(7):674-693.
[30] 赵松年，熊小芸．子波变换和子波分析．电子工业出版社，1997．
[31] 吴乐南．数据压缩．电子工业出版社，2000．
[32] Jinwen Zan, Swamy, M.N.S.,Ahmad, M.O..Wavelet Filters in Multi-resolution Motion Estim-ation. Electrical and Computer Engineering, 2001, Vol:2, pp:1321-1326.
[33] J M. Shaprio. Embedded image coding using zerotree of wavelet coefficient. IEEE Trans. On Signal Processing, 1993;41(12):3445-3462.
[34] 章毓晋．图象处理和分析．清华大学出版社，2001．
[35] Saind A, Pearlman W.A new fast and efficient image codec based on set partitioning in hierarchical trees. IEEE Trans on Circuits and Systems for Video Technology 1996,6 (3):243-249.
[36] Y. Zhang and S. Zafar. Motion-compensated wavelet transform coding for color video compression. IEEE Trans On Circuits and System for Video technology Sept. 1992, vol. 2, pp:
285-296.
[37] Seongman Kim, Seunghryon Rhee, Jun Geun Jeon and Kyu Tae Park. Interframe Coding Using Two-Stage Variable Block-Size Multiresolution Motion Estimation and Wavelet Decompos-ition. IEEE Trans On Circuits and System for Video Technology, 1998;8(4):399-410.
[38] Tingrong Zhao, Tatsuo Ohtsuki. A fast MRME scheme for video compression based on coefficients masking in the wavelet domain. TENCON 99. Proceedings of the IEEE Region 10 Conference, 1999, Vol. 1, pp:127-130.
[39] Hyun-Wook Park, Hyung-Sun Kim. Motion estimation using low-band-shift method for wavele-t based moving-picture coding, IEEE Transactions on Image Processing, 2000, Vol. 9, No. 4, pp:577-587.
[40] Van der Auwera, G. Munteanu, A. Lafruit, G. Cornelis. Video coding based on motion estimation in the wavelet detail images. Acoustics, Speech, and Signal Processing, 1998. ICASSP'98. Proceedings of the 1998 IEEE International Conference on, May 1998 Vol. 5, pp:2801-2804.
[41] Podilchuk Christine I, Jayant Nikil S, Nariman Farvardin. Three Dimentional Subband Coding of Video. IEEE Trans. Image Processing, 1995,4(2):125-139.
[42] Chang Y, Pearlman W.A. Three-dimensional subband coding using the zero-tree method, Proc SPIE, Mar, 1997:1302-1312.
[43] Kim B J, Pearlman W.A. An embedded wavelet video coder using three-dimensional set Partitioning in hierarchical trees(SPIHT).Proc Data Compression Conf, 1997:251-260.
[44] 林刚，刘泽民．基于运动补偿的三维小波视频编码．中国图象图形学报，Nov，1998，Vol．3，pp：926-929
[45] Martuci S A, SodagarI, ChiangT, Zhang Y Q.A zerotree wavelet videocoder. IEEE Trans on circuits and systems for video technology, 1997;7(1):85-94.