图像编码与运动补偿算法研究
|
摘要
随着现代信息社会对通信业务要求的不断增长，图像通信与通信网容量的矛盾日益突出。特别是具有庞大数据量的数字图像通信，更难以传输与存储，极大地制约了图像通信的发展，成为图像通信发展中的“瓶颈”问题。图像压缩编码的目的就是要以尽量少的比特数表征图像，同时保持复原图像的质量，使它符合特定应用场合的要求。
本文主要研究基于小波的图像编码算法及快速DCT算法和钻石运动估值算法。研究了经典算法的不足，提出改进的算法，在相同信噪比的条件下提高了压缩比。
本文由三部分构成：基于小波的图像编码算法、DCT编码算法以及钻石运动估值算法。
第一部分在详细研究小波理论的基础上对小波系数的组织形式进行了讨论，提出改进型Shapiro编码算法，更有效的组织小波系数并对编码序列进行一次预编码，提高了压缩比。
第二部分采用JPEG、H261／H263、MPEG1／2等国际标准中所选择的量化矩阵，根据余弦函数的周期性提出一种基于查表的DCT编码算法。
第三部分针对当前一些运动补偿算法的特点提出了钻石搜索算法，可以提高运动矢量的准确性，并且根据大量实验证明，钻石算法的收敛性非常好。
In modern society, with continuous growth of the requirement for communication business, the contradiction between image communication and net capacity is becoming imminent. Especially for large data of digital picture, communication is so difficult to transmit and store that it confine the development of image communication and become the "bottle-neck". The target of image compression is to express image with less byte, as well as maintain the quality of the restore image to meet the requirement of special environment.
This thesis analyzes the image-coding algorithm based on wavelet, rapid DCT algorithm and diamond compensation algorithm. This paper analyzes the shortcoming of classical algorithm, puts forward the new algorithm and enhances the compression ratio on the condition of the same SNR.
This series is composed of three parts: image code algorithm based on wavelet, DCT coding algorithm and diamond movement compensation algorithm.
In the first part of this thesis, on the base of analyzing the wavelet theory and discussing the wavelet coefficient, a kind of Shapiro code algorithm is put forward. It efficiently organizes the wavelet coefficient and enhances the compression.
In the second part of this work, According to cosine function's cycle property and using the quantization square choosed by international standards such as JPEG, H261/H263,MPEG1/2, a kind of DCT coding algorithm that based on tabling is put forward.
In the third part of this thesis, according to the characteristic of some movement compensation algorithms, a kind of diamond search algorithm is put forward. It enhances the accuracy and according amount of experiment. According to amount of experiments, diamond algorithm's convergence property is very sound.
本文主要研究基于小波的图像编码算法及快速DCT算法和钻石运动估值算法。研究了经典算法的不足，提出改进的算法，在相同信噪比的条件下提高了压缩比。
本文由三部分构成：基于小波的图像编码算法、DCT编码算法以及钻石运动估值算法。
第一部分在详细研究小波理论的基础上对小波系数的组织形式进行了讨论，提出改进型Shapiro编码算法，更有效的组织小波系数并对编码序列进行一次预编码，提高了压缩比。
第二部分采用JPEG、H261／H263、MPEG1／2等国际标准中所选择的量化矩阵，根据余弦函数的周期性提出一种基于查表的DCT编码算法。
第三部分针对当前一些运动补偿算法的特点提出了钻石搜索算法，可以提高运动矢量的准确性，并且根据大量实验证明，钻石算法的收敛性非常好。
In modern society, with continuous growth of the requirement for communication business, the contradiction between image communication and net capacity is becoming imminent. Especially for large data of digital picture, communication is so difficult to transmit and store that it confine the development of image communication and become the "bottle-neck". The target of image compression is to express image with less byte, as well as maintain the quality of the restore image to meet the requirement of special environment.
This thesis analyzes the image-coding algorithm based on wavelet, rapid DCT algorithm and diamond compensation algorithm. This paper analyzes the shortcoming of classical algorithm, puts forward the new algorithm and enhances the compression ratio on the condition of the same SNR.
This series is composed of three parts: image code algorithm based on wavelet, DCT coding algorithm and diamond movement compensation algorithm.
In the first part of this thesis, on the base of analyzing the wavelet theory and discussing the wavelet coefficient, a kind of Shapiro code algorithm is put forward. It efficiently organizes the wavelet coefficient and enhances the compression.
In the second part of this work, According to cosine function's cycle property and using the quantization square choosed by international standards such as JPEG, H261/H263,MPEG1/2, a kind of DCT coding algorithm that based on tabling is put forward.
In the third part of this thesis, according to the characteristic of some movement compensation algorithms, a kind of diamond search algorithm is put forward. It enhances the accuracy and according amount of experiment. According to amount of experiments, diamond algorithm's convergence property is very sound.
引文
[1] 沈兰荪．图像编码与异步传输． 北京：人民邮电出版社，1998．5
[2] 吴乐南．数据压缩．北京：电子工业出版社，2000．4
[3] Kenneth.R.Castleman．数字图像处理．北京：电子工业出版社，1998．9
[4] 沈兰荪．智能仪器与信号处理技术．北京：科学出版社，1990
[5] 崔锦泰．小波分析导论．西安：西安交通大学出版社，1994
[6] 刘贵忠．小波分析理论的应用．西安：西安电子科技大学出版社，1990
[7] 秦前清，杨宗凯．小波分析理论的应用．西安：西安电子科技大学出版社，1992
[8] Meyer Y．小波与算子(Vol．1)．北京：世界图书出版社，1990
[9] Mallat S. Multifrequency channel decomposition of images and wavelet models. IEEE Trans on ICASSP, 1989,37(12):2091-2110
[10] Mallat S. A theory for multiresolution signal decomposition: the wavelet representation. IEEE Trans Pattern Anal Mach Intel, 1989, 11(7):674-693
[11] Burt P J. The laplacian pyramid as a compact image code. IEEE Trans Comm, 1983, 31(4):532-540
[12] Rioul O. A discrete-time multiresolution theory. IEEE Trans SP, 1993, 41(8): 2591-2606
[13] Zettler W R. Application of compactly supported wave to image compression. SPIE' 90, 1244:150-160
[14] 庄镇泉．神经网络与神经计算机．北京：科学出版社，1992
[15] Albanesi M G. Image compression by wavelet decomposition. Signal Processing, 1992,3(3):265-274
[16] 钟义信．智能理论与技术—人工智能与神经网络．北京：人民邮电出版社，1992
[17] Banham M R. A wavelet transform image coding technique with a quadtree structure, In:ICASSP' 92.
[18] Ohta M. Wavelet picture coding with transform coding approach. IEICE Trans Fund, 1992, e75-A(7):776-785
[19] Shapiro J. Embedded image coding using zerotrees of wavelt coefficients. IEEE Trans On SP, 1993, 41(12): 3445-3461
[20] Shapiro .J. An embedder wavelet hierarchical image coder. In:
ICASSP' 92.
[21] Witten H, Neal R, Cleary J G. Arithmetic coding for data compressing [J]. Comm. ACM, 1987, 30(6): 520-540
[22] A. Murat. Tekalp 著，崔之苦，江春译，数字视频处理，北京：电子工业出版社，1998，1
[23] 王琪，钟玉琢．一种结合量化的零树小波图像编码器．清华大学学报，2000，7(40)：29-33
[24] 夏勇，田捷．一种改进的零树小波图像压缩算法．软件学报，1999：626-630
[25] 余松煜．现代图像信息压缩技术．北京：科学出版社，1998
[26] Ahmed N, Rao K R. Orthogonal Transforms for Digital signal Processing. New York, Ny: Springer, 1975,30(Feb):200-204
[27] Ahmed N T, Natarajan, Rao K R, Discrete cosine transform. IEEE Trans Comput, 1974,23(Jan): 90-93
[28] 王福龙，余英林．改进的3-D DCT视频编码方案．通信学报，2001，22L：(1)：102-105
[29] 余英林．图像处理与模式识别[M]．广州：华南理工大学出版社，1991，108-120
[30] R, Chellappa and A. Jain, Marker Random Fields: Theory and Application, Academic Press, 1993:215-221
[31] 阿罕麦德N，罗K R．数字信号处理中的正交变换[M]．北京：人民邮电出版社，1979
[32] 吴红文，李久贤．一种新的二维离散余弦变换快速算法[J]．东南大学学报，1996，26(2)：7-15
[33] 郭太峰，钱源诚．用查表法实现的DCT方法探讨[J]．数据采集与处理，1997，12(4)：272-275
[34] 柳海泉，郑建宏．矢量量化的一种快速编码算法[J]．重庆邮电学院学报，1998，10(1)：1-3
[35] 王新成，卢颉，朱维乐．二维离散余弦变换的一种新的快速[J]，电子科技大学学报，1993，22(6)：561-568
[36] Conway J, et al. Fast quantization and decoding algorithms for lattice quantizers and codes. IEEE Trans IT, 1982,28(2):227-231
[37] A.K. Jain, Fundamentals of Digital Image Processing, Englewood Cliffs, NJ:Prentice Hall, 1989:462-470
[38] 吴乐南 编著，徐孟侠 审，数据压缩的原理与应用 北京：电子工业出版社，1994
[39] 毕厚杰、陈延标，图像通信工程，北京：人民邮电出版社，1995
[40] A. Verri and T. Poggio, "Motion field and optical flow:Qualitative properties," IEEE Trans. Patt. Anal. Mach. Intel., 1989, Vol. PAMI-11: 490-498
[41] M. bertero, T.A. Poggio and V. Torre, "Ill-posed problems in early vision," Proc. IEEE, 1988, vol. 76: 869-889
[42] P. Anandan, J. R. Bergen, K. J Hanna, and R. Hingorani, "Hierarchical model-based motion estimation," in Motion Analysis and Image Sequence Processing, M. I. Sezan and R. L. Lagendijk, eds,. Norwell, MA: kluwer, 1993
[43] J. K. Aggarawal and N.Nandhakumar, "On the computation of motion from sequences of images," Proc. IEEE, I988, vol. 76:917-935
[44] N. Ohta, "Optical flow detection by color images" NEC Res. and Dev., 1990,no. 97:78-84,
[45] J. R. Jain and A. K. Jain, "Displacement measurement and its application in interframe image coding," IEEE Trans. Commun. 1981, vol. 29:1799-1808,
[46] H. Gharavi and M. Mills, "Block-matching motion estimation algorithms: New results,"IEEE Trans. Circ. And Syst.,1990, vol. 37, pp. 649-651
[47] M. Ghanbari, "The cross-search algorithm for motion-estimation," IEEE Trans. Commun.,1990, vol. 38, pp. 950-953
[48] T. Koga, K. Iinuma, A. Hirano, Y. Iijima, and T. Ishiguro, "Motion compensated interframe image coding," IEEE Trans. Commun,. Vol. COMM-29,:G5.3.1-5.3.5
[49] David J.Kruglinski著，潘爱民，王国印 译．VC++技术内幕．北京：清华出版社，1998
[2] 吴乐南．数据压缩．北京：电子工业出版社，2000．4
[3] Kenneth.R.Castleman．数字图像处理．北京：电子工业出版社，1998．9
[4] 沈兰荪．智能仪器与信号处理技术．北京：科学出版社，1990
[5] 崔锦泰．小波分析导论．西安：西安交通大学出版社，1994
[6] 刘贵忠．小波分析理论的应用．西安：西安电子科技大学出版社，1990
[7] 秦前清，杨宗凯．小波分析理论的应用．西安：西安电子科技大学出版社，1992
[8] Meyer Y．小波与算子(Vol．1)．北京：世界图书出版社，1990
[9] Mallat S. Multifrequency channel decomposition of images and wavelet models. IEEE Trans on ICASSP, 1989,37(12):2091-2110
[10] Mallat S. A theory for multiresolution signal decomposition: the wavelet representation. IEEE Trans Pattern Anal Mach Intel, 1989, 11(7):674-693
[11] Burt P J. The laplacian pyramid as a compact image code. IEEE Trans Comm, 1983, 31(4):532-540
[12] Rioul O. A discrete-time multiresolution theory. IEEE Trans SP, 1993, 41(8): 2591-2606
[13] Zettler W R. Application of compactly supported wave to image compression. SPIE' 90, 1244:150-160
[14] 庄镇泉．神经网络与神经计算机．北京：科学出版社，1992
[15] Albanesi M G. Image compression by wavelet decomposition. Signal Processing, 1992,3(3):265-274
[16] 钟义信．智能理论与技术—人工智能与神经网络．北京：人民邮电出版社，1992
[17] Banham M R. A wavelet transform image coding technique with a quadtree structure, In:ICASSP' 92.
[18] Ohta M. Wavelet picture coding with transform coding approach. IEICE Trans Fund, 1992, e75-A(7):776-785
[19] Shapiro J. Embedded image coding using zerotrees of wavelt coefficients. IEEE Trans On SP, 1993, 41(12): 3445-3461
[20] Shapiro .J. An embedder wavelet hierarchical image coder. In:
ICASSP' 92.
[21] Witten H, Neal R, Cleary J G. Arithmetic coding for data compressing [J]. Comm. ACM, 1987, 30(6): 520-540
[22] A. Murat. Tekalp 著，崔之苦，江春译，数字视频处理，北京：电子工业出版社，1998，1
[23] 王琪，钟玉琢．一种结合量化的零树小波图像编码器．清华大学学报，2000，7(40)：29-33
[24] 夏勇，田捷．一种改进的零树小波图像压缩算法．软件学报，1999：626-630
[25] 余松煜．现代图像信息压缩技术．北京：科学出版社，1998
[26] Ahmed N, Rao K R. Orthogonal Transforms for Digital signal Processing. New York, Ny: Springer, 1975,30(Feb):200-204
[27] Ahmed N T, Natarajan, Rao K R, Discrete cosine transform. IEEE Trans Comput, 1974,23(Jan): 90-93
[28] 王福龙，余英林．改进的3-D DCT视频编码方案．通信学报，2001，22L：(1)：102-105
[29] 余英林．图像处理与模式识别[M]．广州：华南理工大学出版社，1991，108-120
[30] R, Chellappa and A. Jain, Marker Random Fields: Theory and Application, Academic Press, 1993:215-221
[31] 阿罕麦德N，罗K R．数字信号处理中的正交变换[M]．北京：人民邮电出版社，1979
[32] 吴红文，李久贤．一种新的二维离散余弦变换快速算法[J]．东南大学学报，1996，26(2)：7-15
[33] 郭太峰，钱源诚．用查表法实现的DCT方法探讨[J]．数据采集与处理，1997，12(4)：272-275
[34] 柳海泉，郑建宏．矢量量化的一种快速编码算法[J]．重庆邮电学院学报，1998，10(1)：1-3
[35] 王新成，卢颉，朱维乐．二维离散余弦变换的一种新的快速[J]，电子科技大学学报，1993，22(6)：561-568
[36] Conway J, et al. Fast quantization and decoding algorithms for lattice quantizers and codes. IEEE Trans IT, 1982,28(2):227-231
[37] A.K. Jain, Fundamentals of Digital Image Processing, Englewood Cliffs, NJ:Prentice Hall, 1989:462-470
[38] 吴乐南 编著，徐孟侠 审，数据压缩的原理与应用 北京：电子工业出版社，1994
[39] 毕厚杰、陈延标，图像通信工程，北京：人民邮电出版社，1995
[40] A. Verri and T. Poggio, "Motion field and optical flow:Qualitative properties," IEEE Trans. Patt. Anal. Mach. Intel., 1989, Vol. PAMI-11: 490-498
[41] M. bertero, T.A. Poggio and V. Torre, "Ill-posed problems in early vision," Proc. IEEE, 1988, vol. 76: 869-889
[42] P. Anandan, J. R. Bergen, K. J Hanna, and R. Hingorani, "Hierarchical model-based motion estimation," in Motion Analysis and Image Sequence Processing, M. I. Sezan and R. L. Lagendijk, eds,. Norwell, MA: kluwer, 1993
[43] J. K. Aggarawal and N.Nandhakumar, "On the computation of motion from sequences of images," Proc. IEEE, I988, vol. 76:917-935
[44] N. Ohta, "Optical flow detection by color images" NEC Res. and Dev., 1990,no. 97:78-84,
[45] J. R. Jain and A. K. Jain, "Displacement measurement and its application in interframe image coding," IEEE Trans. Commun. 1981, vol. 29:1799-1808,
[46] H. Gharavi and M. Mills, "Block-matching motion estimation algorithms: New results,"IEEE Trans. Circ. And Syst.,1990, vol. 37, pp. 649-651
[47] M. Ghanbari, "The cross-search algorithm for motion-estimation," IEEE Trans. Commun.,1990, vol. 38, pp. 950-953
[48] T. Koga, K. Iinuma, A. Hirano, Y. Iijima, and T. Ishiguro, "Motion compensated interframe image coding," IEEE Trans. Commun,. Vol. COMM-29,:G5.3.1-5.3.5
[49] David J.Kruglinski著，潘爱民，王国印 译．VC++技术内幕．北京：清华出版社，1998