MPEG-2到AVS视频转码器的研究与设计
|
摘要
现阶段视频通信领域，存在多种编码标准，由不同的组织，针对不同应用、场合制定，其中，ISO/IEC制定的MPEG-2标准在固定多媒体通信中得到了广泛的应用，在硬件、软件以及多媒体数据存储上都有巨大的积累，但在很多场合，MPEG-2编码数据并不适用，例如，无线移动网络、无线视频会议系统等。与MPEG-2相比，AVS具有更高的编码效率、能够支持更广泛的应用以及不受专利产权的困扰等优势。在这样的情况下，直接将现有大量的MPEG-2视频流直接变为AVS的视频流是有效的解决方案。
本论文首先简要介绍了视频转码技术的研究背景和研究现状，并阐述了本文的主要研究内容及主要工作。
其次，对MPEG-2和AVS的视频编码标准，尤其是MPEG-2的解码过程和AVS的编码过程进行了研究，通过比较分析，找到了两种标准的异同点，为MPEG-2到AVS视频转码器的设计做好了知识储备。
再次，简要分析与比较几种主流视频转码器的结构与性能。针对MPEG-2和AVS两种标准的异同点，提出了一种改进的基于像素域的级联式MPEG-2到AVS视频流转码方案。同时详细研究与分析了在实现该转码器过程中涉及的关键技术环节以及所用到的算法。
接着，为了缩短转码时间和提高转码质量，对AVS编码器的性能进行了优化。在分析了各模块运算耗时情况后，对其中运算量最大、最耗时的运动估计和模式选择提出了快速算法。
最后设计并实现了MPEG-2到AVS视频转码系统。实验表明本文的转码方案在兼顾转码后视频质量的同时，较大的加快了转码的速度，达到了实时转码的要求。
Currently there are a variety of encoding standards in different cases, aiming at different application evelopments occasion in video communications, in which MPEG-2 standard established by ISO/IEC has been widely applied in immobility multimedia communication and has huge accumulation in hardware, software and multimedia data storage. MPEG-2 data is not applicable in many cases, such as wireless mobile networks and wireless video conferencing system. Compared with MPEG-2, AVS has higher coding efficiency, can support wider use and property rights and doesn't have the trouble of patents right and so on. So, convert lots of MPEG-2 video stream directly to AVS Video stream is a most effective solution.
Firstly in this tesis, the video transcoding technology research background and current status of research is introduced, the main content and our main tasks is proposed.
Secondly, the MPEG-2 and AVS video coding standard, particularly the MPEG-2 decoding process and the AVS encoding process are researched. Through a comparative analysis similarities and differences between the two standards are found, which supplies knowledge reserves with the design of the transcoding device from MPEG-2 to AVS.
Thirdly, the structure and properties of several mainstream video transcoding devices are briefly analyzed and compared. Aiming at the similarities and differences between MPEG-2 and AVS standards, an improved, based on the pixel field and cascade transcoding program from MPEG-2 to AVS video stream. On the same time, the key technology sectors and the algorithm used in achieving transcoding process is detailed analyzed and discussed.
Fourthly, in order to shorten the time to transcode and improve code quality, the performance of the AVS encoder is optimized. After the analysis of the computational time of each module, aiming at the largest amount of computation, the most time-consuming and patterns choice, a fast algorithm is proposed.
Finally, the PEG-2 AVS video transcoding system is implemented in this paper, in which the research and improved algorithm are appplied reasonably. Through the test, the fact that the program of this paper greatly accelerates the rate of transcoding and takes into account the video quality after transcoding is demonstrated.
本论文首先简要介绍了视频转码技术的研究背景和研究现状，并阐述了本文的主要研究内容及主要工作。
其次，对MPEG-2和AVS的视频编码标准，尤其是MPEG-2的解码过程和AVS的编码过程进行了研究，通过比较分析，找到了两种标准的异同点，为MPEG-2到AVS视频转码器的设计做好了知识储备。
再次，简要分析与比较几种主流视频转码器的结构与性能。针对MPEG-2和AVS两种标准的异同点，提出了一种改进的基于像素域的级联式MPEG-2到AVS视频流转码方案。同时详细研究与分析了在实现该转码器过程中涉及的关键技术环节以及所用到的算法。
接着，为了缩短转码时间和提高转码质量，对AVS编码器的性能进行了优化。在分析了各模块运算耗时情况后，对其中运算量最大、最耗时的运动估计和模式选择提出了快速算法。
最后设计并实现了MPEG-2到AVS视频转码系统。实验表明本文的转码方案在兼顾转码后视频质量的同时，较大的加快了转码的速度，达到了实时转码的要求。
Currently there are a variety of encoding standards in different cases, aiming at different application evelopments occasion in video communications, in which MPEG-2 standard established by ISO/IEC has been widely applied in immobility multimedia communication and has huge accumulation in hardware, software and multimedia data storage. MPEG-2 data is not applicable in many cases, such as wireless mobile networks and wireless video conferencing system. Compared with MPEG-2, AVS has higher coding efficiency, can support wider use and property rights and doesn't have the trouble of patents right and so on. So, convert lots of MPEG-2 video stream directly to AVS Video stream is a most effective solution.
Firstly in this tesis, the video transcoding technology research background and current status of research is introduced, the main content and our main tasks is proposed.
Secondly, the MPEG-2 and AVS video coding standard, particularly the MPEG-2 decoding process and the AVS encoding process are researched. Through a comparative analysis similarities and differences between the two standards are found, which supplies knowledge reserves with the design of the transcoding device from MPEG-2 to AVS.
Thirdly, the structure and properties of several mainstream video transcoding devices are briefly analyzed and compared. Aiming at the similarities and differences between MPEG-2 and AVS standards, an improved, based on the pixel field and cascade transcoding program from MPEG-2 to AVS video stream. On the same time, the key technology sectors and the algorithm used in achieving transcoding process is detailed analyzed and discussed.
Fourthly, in order to shorten the time to transcode and improve code quality, the performance of the AVS encoder is optimized. After the analysis of the computational time of each module, aiming at the largest amount of computation, the most time-consuming and patterns choice, a fast algorithm is proposed.
Finally, the PEG-2 AVS video transcoding system is implemented in this paper, in which the research and improved algorithm are appplied reasonably. Through the test, the fact that the program of this paper greatly accelerates the rate of transcoding and takes into account the video quality after transcoding is demonstrated.
引文
[1] Jun Xin, Ming-Ting Sun, Kangwook Chun, Byung Sun Choi. Motionre-estimation for HDTV and SDTV Transcoding. Circuits and Systems. 2002 (4) :715-718P
[2] P N Tudor, A.H Werner. Real-Time Transcoding of MPEG-2 Video Bit Streams. International Broadcasting Convention, 2002
[3] 范晓东，潘健，余松煌．基于频域的MPEG-2码率转码技术及其码率控制策略，电子学报．2001，29(1)：57-60页
[4] S. Dogan, A.H. Sadka, A.M. Kondoz. MPEG-4 Video Transcoder for Mobile Multimedia Traffic Planning. Proceedings of the IEE Second Tnternational Conference on 3G Mobile Communication Technologies, IEE Conference Publication No. X77, London, HK, March 2001:109-113P
[5] YongQing Liang, Yap-Peng fan. Methods and needs for transcoding MPEG-4 fine granularity scalahiiity video. Circuits and Systems, 2pp2 IEEE International Symposium on. 2002 (4):719-722P
[6] Safak DOGAN, Akin CELLATOGLU, Abdul H. SADKA, Ahmet M. KONDOZ. Error-Resilient MPEG-4 Video Transcoder for Bit Rate Regulation. Proceedings of the Fifth World Multi-Conference on Systemics, Cybernetics and Informatics, Orlando, Florida, USA, July 2001:312-317P
[7] Z. Lei, N.D. Georganas. H. 263 Video Transcoding for Spatial Resolution Downscaling. Proc. IEEE International Conference on Information Technology: Coding and Computing, Las Vegas, Nevada, April 2002:125-128P
[8] Soam acharya, Brian Smith. Compressed Domain Transcoding of MPEG. TEEE International Conference on Multimedia Computing and Systems, 1998 (4):298-304P
[9] Shanableh T. , Ghanhari M. Transcoding of video into different encoding formats. Acoustics, Speech, and Signal Processing, 2000 (4) :1927-1930P
[10] Susie J. Lee, Bhaskaram Vasudev, Sam Liu. Transcoding MPEG Video Streams in the Compressed Domain. http://www. hpl. hp. co. m/per. sonai. /Susie Wee/PAPERS/pidc97/hp idc97. html
[11] Dogan S., Sadka A. H, Kondoz A.M. Efficient MPEG-4/H. 263 video transcoder for interoperability of heterogeneous multimedia networks. Electtunics Letters, 1999, 35 (11) :863-864P
[12] Masaru Sugano, An efficient Yasuyuki Nakajima, Hiromasa Yanagihara, Akio Yoneyama. Transcoding from MPEG-2 to MPEG-1. Image Processing, 2001. Proceedings. 2001 Internat Tona Conference on. Volume 1, 2001:417-420P
[13] Susie J. Wee, John G. apostolopoulos, Nick Feamster. Field-to-frame Transcoding with Spatial and Temporal Downsampling. IEEE International Conference on Image Processing, Kobe, Japan, 1999
[14] Kwang-deok Seo, Seong-Gheol Heo, Jae-kyoon Kim. A rate control algorithm based on adaptve R-Q madel for MPEG-1 to MPEG-4 transcoding in DCT domain, 2002:109-113P
[15] Nakajima J., Tsujl H., Yashima Y., Kobayashi N. Motion vectorre-estimation for fastvideo transcoding from MPEG-2 to MPEG-4. 2001 Proceedings of Workshop and Exhibition on, 2001:87-90P
[16] Jun Xin, Ming-Ting Sun, Kangwook Chun. Motion Re-estimation for MPEG-2 to MPEG-4 Simple Profile Transcoding. http://students, washinton. edu/.jxin/pub/02pv. pdf
[17] Kwang-deok Seo, Jae-kyoon Kim. Fast motion vector refinement for MPEG-1 to MPEG-4 transcoding with spatial down-sampling in DCT domain. 2001 International Conference, 2001, 1:469—472P
[18] Wang Xing Guo, Liu Ji Lin, Wei Guo Zheng. Mismatch MB Retrieval for MPEG-2 to MPEG—4 Transcoding. IEEE Pacific Rim Conference on Multimedia. 2001:86-93P
[19] 王兴国，解蓉，刘济林．压缩域MPEG-2到MPEG-4视频转码中不匹配宏块的复原算法．电子学报．2002，30(9)：1405-1408页
[20] Chang, S.F., Messerschmitt, D.G. Manipulation and Composting of MC-DCT Compressed Video. IEEE Journal on Selected Areas in Communication: Special issueon Intelligent Signal Processing. 1995, 13:1-11P
[21] Jeongnam Youn, Ming-Ting Sun. Video transcoding for multiple clients. Proc. SPIE: Visual Communication and Image Processing 2000. Sydney, Australia, Aug, 2000:76-85P
[22] Youn J, Sun M T, Xin J. Video transcoder architectures for bit rate scaling of H. 263 bit streams. ACM Multimedia, Nov, 1999: 243-250P
[23] Shanableh T., Ghanbari M. Heterogeneous video transcoding to lower spatio-temporal resolutions and different encoding formats. Multimedia, IEEE Transactions on. 2000, 2 (2):101-110P
[24] Shih-Fu Chang, Messerschmitt D.C. Manipulation and compositing of MC-DCT compressed video. Selected Areas in Communications, IEEE Journal on, Jan. 1995, 13 (1):1-11P
[25] Donveon Kim, Bumsik Yoon. Conversion between DV and MPEG-2 intra coding. International Conference on Consumer Electronics, 2001:34-35P
[26] Minhua Zhou. Video Block Transform. U.S. Patent: No. 2003-OO93452A1
[27] PL Hosur, K.K. Ma. Motion Vector Field Adaptive Fast Motion Estimation. Int. Conf. Information, Communications and Signal Processing, 1999
[28] A.M. Tourapis, O.C. Au, M.L. Liou. Fast Block-matching Motion Estimation Using Predictive Motion Vector Field Adaptive Search Technique. Noordwijkerhout, NL, Mar, 2000
[29] A.M. Tourapis, O.C. Au, andM. L. Liou. Highly Efficient Predictive ZonalAlgorithms for Fast Block Matching Motion Estimation. Circuits Syst. Video Technol, Oct, 2002, 12 (10): 934-947P
[30] Yu-Kung Tu, Jar-Ferr Yang, Yi-Nung Shen, Ming-Ting Stan. Fast Variable-Size Block Motion Estimation Using Merging Procedure with an Adaptive Threshold, 2003.
[31] 李学明，李继．H．264视频编码算法性能分析．中国图象图形学报，2003(12)：1450-1456页
[32] 刘海峰，郭宝龙．基于块匹配运动估值的正方形一菱形搜索算法．计算机学报，2002，25(7)：747-752页
[33] 丁贵广，郭宝龙．基于线性搜索的快速运动估计算法．西安交通大学学报，2004，38(2)：1182-1186页
[34] 向友君，郭宝龙．基于起点预测的快速运动估计算法．西安电子科技大学学报，2003，30(3)：386-390页
[35] 陈兆龙．AVS的发展现状及应用前景．安庆师范学院学报，2005，11(4)：109-112页
[36] 钟玉琢，王琪，赵黎等．MPEG-2运动图像压缩编码国际标准及MPEG的新发展．清华大学出版社，2002
[37] 袁禄军．视频转码技术的研究及其应用，中科院研究生院博士学位论文，2005
[38] 高文．流媒体数据压缩技术．电子工业出版社，1994
[39] 张益贞，刘滔．Visual C++实现MPEG／JPEG编解码技术．人民邮电出版，2002
[40] 侯自强．发展IPTV应该关注的几个问题．世界电信，2006，11(3)：21-24页
[2] P N Tudor, A.H Werner. Real-Time Transcoding of MPEG-2 Video Bit Streams. International Broadcasting Convention, 2002
[3] 范晓东，潘健，余松煌．基于频域的MPEG-2码率转码技术及其码率控制策略，电子学报．2001，29(1)：57-60页
[4] S. Dogan, A.H. Sadka, A.M. Kondoz. MPEG-4 Video Transcoder for Mobile Multimedia Traffic Planning. Proceedings of the IEE Second Tnternational Conference on 3G Mobile Communication Technologies, IEE Conference Publication No. X77, London, HK, March 2001:109-113P
[5] YongQing Liang, Yap-Peng fan. Methods and needs for transcoding MPEG-4 fine granularity scalahiiity video. Circuits and Systems, 2pp2 IEEE International Symposium on. 2002 (4):719-722P
[6] Safak DOGAN, Akin CELLATOGLU, Abdul H. SADKA, Ahmet M. KONDOZ. Error-Resilient MPEG-4 Video Transcoder for Bit Rate Regulation. Proceedings of the Fifth World Multi-Conference on Systemics, Cybernetics and Informatics, Orlando, Florida, USA, July 2001:312-317P
[7] Z. Lei, N.D. Georganas. H. 263 Video Transcoding for Spatial Resolution Downscaling. Proc. IEEE International Conference on Information Technology: Coding and Computing, Las Vegas, Nevada, April 2002:125-128P
[8] Soam acharya, Brian Smith. Compressed Domain Transcoding of MPEG. TEEE International Conference on Multimedia Computing and Systems, 1998 (4):298-304P
[9] Shanableh T. , Ghanhari M. Transcoding of video into different encoding formats. Acoustics, Speech, and Signal Processing, 2000 (4) :1927-1930P
[10] Susie J. Lee, Bhaskaram Vasudev, Sam Liu. Transcoding MPEG Video Streams in the Compressed Domain. http://www. hpl. hp. co. m/per. sonai. /Susie Wee/PAPERS/pidc97/hp idc97. html
[11] Dogan S., Sadka A. H, Kondoz A.M. Efficient MPEG-4/H. 263 video transcoder for interoperability of heterogeneous multimedia networks. Electtunics Letters, 1999, 35 (11) :863-864P
[12] Masaru Sugano, An efficient Yasuyuki Nakajima, Hiromasa Yanagihara, Akio Yoneyama. Transcoding from MPEG-2 to MPEG-1. Image Processing, 2001. Proceedings. 2001 Internat Tona Conference on. Volume 1, 2001:417-420P
[13] Susie J. Wee, John G. apostolopoulos, Nick Feamster. Field-to-frame Transcoding with Spatial and Temporal Downsampling. IEEE International Conference on Image Processing, Kobe, Japan, 1999
[14] Kwang-deok Seo, Seong-Gheol Heo, Jae-kyoon Kim. A rate control algorithm based on adaptve R-Q madel for MPEG-1 to MPEG-4 transcoding in DCT domain, 2002:109-113P
[15] Nakajima J., Tsujl H., Yashima Y., Kobayashi N. Motion vectorre-estimation for fastvideo transcoding from MPEG-2 to MPEG-4. 2001 Proceedings of Workshop and Exhibition on, 2001:87-90P
[16] Jun Xin, Ming-Ting Sun, Kangwook Chun. Motion Re-estimation for MPEG-2 to MPEG-4 Simple Profile Transcoding. http://students, washinton. edu/.jxin/pub/02pv. pdf
[17] Kwang-deok Seo, Jae-kyoon Kim. Fast motion vector refinement for MPEG-1 to MPEG-4 transcoding with spatial down-sampling in DCT domain. 2001 International Conference, 2001, 1:469—472P
[18] Wang Xing Guo, Liu Ji Lin, Wei Guo Zheng. Mismatch MB Retrieval for MPEG-2 to MPEG—4 Transcoding. IEEE Pacific Rim Conference on Multimedia. 2001:86-93P
[19] 王兴国，解蓉，刘济林．压缩域MPEG-2到MPEG-4视频转码中不匹配宏块的复原算法．电子学报．2002，30(9)：1405-1408页
[20] Chang, S.F., Messerschmitt, D.G. Manipulation and Composting of MC-DCT Compressed Video. IEEE Journal on Selected Areas in Communication: Special issueon Intelligent Signal Processing. 1995, 13:1-11P
[21] Jeongnam Youn, Ming-Ting Sun. Video transcoding for multiple clients. Proc. SPIE: Visual Communication and Image Processing 2000. Sydney, Australia, Aug, 2000:76-85P
[22] Youn J, Sun M T, Xin J. Video transcoder architectures for bit rate scaling of H. 263 bit streams. ACM Multimedia, Nov, 1999: 243-250P
[23] Shanableh T., Ghanbari M. Heterogeneous video transcoding to lower spatio-temporal resolutions and different encoding formats. Multimedia, IEEE Transactions on. 2000, 2 (2):101-110P
[24] Shih-Fu Chang, Messerschmitt D.C. Manipulation and compositing of MC-DCT compressed video. Selected Areas in Communications, IEEE Journal on, Jan. 1995, 13 (1):1-11P
[25] Donveon Kim, Bumsik Yoon. Conversion between DV and MPEG-2 intra coding. International Conference on Consumer Electronics, 2001:34-35P
[26] Minhua Zhou. Video Block Transform. U.S. Patent: No. 2003-OO93452A1
[27] PL Hosur, K.K. Ma. Motion Vector Field Adaptive Fast Motion Estimation. Int. Conf. Information, Communications and Signal Processing, 1999
[28] A.M. Tourapis, O.C. Au, M.L. Liou. Fast Block-matching Motion Estimation Using Predictive Motion Vector Field Adaptive Search Technique. Noordwijkerhout, NL, Mar, 2000
[29] A.M. Tourapis, O.C. Au, andM. L. Liou. Highly Efficient Predictive ZonalAlgorithms for Fast Block Matching Motion Estimation. Circuits Syst. Video Technol, Oct, 2002, 12 (10): 934-947P
[30] Yu-Kung Tu, Jar-Ferr Yang, Yi-Nung Shen, Ming-Ting Stan. Fast Variable-Size Block Motion Estimation Using Merging Procedure with an Adaptive Threshold, 2003.
[31] 李学明，李继．H．264视频编码算法性能分析．中国图象图形学报，2003(12)：1450-1456页
[32] 刘海峰，郭宝龙．基于块匹配运动估值的正方形一菱形搜索算法．计算机学报，2002，25(7)：747-752页
[33] 丁贵广，郭宝龙．基于线性搜索的快速运动估计算法．西安交通大学学报，2004，38(2)：1182-1186页
[34] 向友君，郭宝龙．基于起点预测的快速运动估计算法．西安电子科技大学学报，2003，30(3)：386-390页
[35] 陈兆龙．AVS的发展现状及应用前景．安庆师范学院学报，2005，11(4)：109-112页
[36] 钟玉琢，王琪，赵黎等．MPEG-2运动图像压缩编码国际标准及MPEG的新发展．清华大学出版社，2002
[37] 袁禄军．视频转码技术的研究及其应用，中科院研究生院博士学位论文，2005
[38] 高文．流媒体数据压缩技术．电子工业出版社，1994
[39] 张益贞，刘滔．Visual C++实现MPEG／JPEG编解码技术．人民邮电出版，2002
[40] 侯自强．发展IPTV应该关注的几个问题．世界电信，2006，11(3)：21-24页