臺灣博碩士論文加值系統

相較於目前相當普遍的MPEG-2、MPEG-4（此指MPEG4-part2）影音壓縮技術，新一代的H.264/AVC無疑擁有更良好的壓縮比與較低的失真率。但在當前的環境下，H.264的普及率卻不及MPEG-2甚至MPEG-4，主因之ㄧ當然是前述的技術在市場上推廣多年，目前影音設備大多可沿用，另外H.264其強大效能所相對帶來的複雜運算量，壓縮時間亦較難達到即時的運算效果。現今市面上所開發出來能夠即時壓縮的H.264 Encoder產品，其價位較難被一般使用者所接受。在本研究中我們預計採用目前H.264演算法中效能/速度比最佳的x264為開發基礎，將x264平台在TI DM642 DSP上實作。在硬體架構的研究當中，TI DM642 DSP本身擁有良好的影像處理效能，妥善規劃記憶體配置並善用硬體指令，在系統架構上，有非常大的優化空間。在軟體上我們結合了多角度的演算法達到最佳化。在現今眾多的嵌入式H.264演算法當中，大部份都是使用硬體實現Codec的方式實作，雖然能得到良好的效能，卻有缺乏更改彈性、成本高、無法置換功能及版本更新不易等缺點。本研究所發展的H.264 Encoder Loopback架構，能在CIF達到real time的效率(超過40 FPS)，VGA解析度達到22.6 FPS的速度，不但兼具上述方式的優點，更有易維護、低成本與多功能的特性，未來擁有無限的衍生應用。

H.264/AVC is the most advanced compression technology which offers better compression ratio and lower distortion than the previous video compression standards such as MPEG-4 and MPEG-2. However, the computational cost is really high. x264 has the best performance/
cost time ratio in most of the H.264-based algorithms. The main objective of this thesis is to realize x264 on TI DM642 DSP framework. In this thesis, the complexity will be reduced by using algorithm optimization and programming structure rearrangement. We
also optimized part of the encoder with assembly code and new memory arrangement. The proposed and updated codec can achieve the speed of 22.6 FPS for VGA (640£480) size and realtime (more than 40 FPS) for CIF (352£288) size video sequence.

Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . i
Acknowledgements. . . . . . . . . . . . . . . . . . . . . . ii
Contents . . . . . . . . . . . . . . . . . . . . . . . . . iii

1 Introduction . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Motivation and Surveys . . . . . . . . . . . . . . . . . 1
1.2 Objective . . . . . . . . . . . . . . . . . . . . . . . 3
1.3 Organization . . . . . . . . . . . . . . . . . . . . . . 3

2 Overview of H.264/AVC. . . . . . . . . . . . . . . . . . . 5
2.1 Briefing of H.264/AVC . . . . . . . . . . . . . . . . . 5
2.2 Mode Decision . . . . . . . . . . . . . . . . . . . . . 8
2.3 Block Matching Algorithm . . . . . . . . . . . . . . . . 9
2.4 Transformation and Quantization . . . . . . . . . . . . 10
2.5 Profiles . . . . . . . . . . . . . . . . . . . . . . . 12

3 Brief Review of x264 . . . . . . . . . . . . . . . . . . 14
3.1 H.264 Algorithm Advantages . . . . . . . . . . . . . . 14
3.2 Main Program Structure . . . . . . . . . . . . . . . . 15
3.2.1 Motion Estimation of x264 . . . . . . . . . . . . . . 17
3.2.2 Mode Decision of x264 . . . . . . . . . . . . . . . . 19
3.3 The GNU GPL Licence . . . . . . . . . . . . . . . . . . 20

4 Optimization Approach . . . . . . . . . . . . . . . . . 22
4.1 Complexity Analysis of x264 . . . . . . . . . . . . . . 23
4.2 The Proposed 2-D Mode Decision . . . . . . . . . . . . 23
4.3 Motion Estimation Optimization . . . . . . . . . . . . 27
4.4 Quantization and DCT Optimization . . . . . . . . . . . 29
4.5 System Level Optimization . . . . . . . . . . . . . . . 33
4.5.1 Optimization using Assembly Code . . . . . . . . . . 33
4.5.2 Fixed-Point Architecture Design . . . . . . . . . . . 35
4.5.3 Other Methods . . . . . . . . . . . . . . . . . . . . 39

5 Implementation and Optimization Results . . . . . . . . . 42
5.1 The procedure of development and implementation . . . . 42
5.2 Optimization Results and Performance Comparison . . . . 47

6 Conclusion and Future Work . . . . . . . . . . . . . . . 57

A Execution Cycle Analysis . . . . . . . . . . . . . . . . 65

[1] In-Stat. MPEG video ICs : H.264 in the limelight. 2007.
[2] T. Wiegand, G. J. Sullivan, G. Bjntegaard, and A. Luthra. Overview of the H.264/AVC video coding standard. Circuits and Systems for Video Technology, IEEE Transactions on, 13(7):560–576, 2003.
[3] H. Malvar, A. Hallapuro, M. Karczewicz, and L. Kerofsky. Low-complexity transform and quantization in H.264/AVC. Circuits and Systems for Video Technology, IEEE Transactions on, 13(7):598–603, 2003.
[4] Texas Instrument. TMS320C64x Technical Overview. 2001.
[5] Texas Instrument. TMS320C6000 DSP/BIOS User’s Guide. 2000.
[6] Texas Instrument. TMS320C6000 Optimization Workshop. 2005.
[7] Texas Instrument. IQmath Library. 2003.
[8] L. Zhuo, Q.Wang, D.-D. Feng, and L. Shen. Optimization and Implementation of H.264 Encoder on DSP Platform. Multimedia and Expo, 2007 IEEE International Conference on, pages 232–35, 2007.
[9] H.-J. Wang, Y.-J. Huang, and H. Li. H. 264/avc video encoder implementation based on TI TMS320DM642. Intelligent Information Hiding and Multimedia Signal Processing, 2006. IIH-MSP’06. International Conference on, pages 503–506, 2006.
[10] In-Stat. http://http://www.instat.com.
[11] Fujitsu. The Fujitsu H.264 Format Video-Processing IC MB86H50, 2007.
[12] Fujitsu. Fujitsu to Release H.264 Format Video-Processing LSI Chip Supporting High-Definition and Low Power Consumption, 2007.
[13] K. Goto, A. Hatabu, H. Nishizuka, K. Matsunaga, R. Nakamura, Y. Mochizuki, and T. Miyazaki. H.264 video encoder implementation on a low-power DSP with low and stable computational complexity. Signal Processing Systems Design and Implementation, 2006. SIPS’06. IEEE Workshop on, pages 101–106, 2006.
[14] T.-C. Chen, Y.-H. Chen, S.-F. Tsai, S.-Y. Chien, and L.-G. Chen. Fast algorithm and architecture design of low-power integer motion estimation for H.264/AVC. Circuits and Systems for Video Technology, IEEE Transactions on, 17(5):568–577, 2007.
[15] L. Zhuo, Q. Wang, D.-D. Feng, and L. Shen. Optimization and implementation of H.264 encoder on DSP platform. Multimedia and Expo, 2007 IEEE International Conference on, pages 232–235, 2007.
[16] Z. Li, Q. Xing, and X. Zhu. H.264 video encoder implementation and optimization based on DM642 DSP. Networking, Sensing and Control, 2008. ICNSC 2008. IEEE International Conference on, pages 891–894, 2008.
[17] Z. Wei and C. Cai. Realization and optimization of DSP based H.264 encoder. Circuits and Systems, 2006. ISCAS 2006. Proceedings. 2006 IEEE International Symposium on, page 4, 2006.
[18] H.-.J Wang, Y.-.Y Hou, and H. Li. H.264/AVC video encoder algorithm optimization based on TI TMS320DM642. Intelligent Information Hiding and Multimedia Signal Processing, 2007. IIHMSP 2007. Third International Conference on, 1, 2007.
[19] I. Werda, H. Chaouch, A. Samet, M. A. B. Ayed, N. Masmoudi, E. Akbal, B. Ergen, H. Muljadi, H. Takeda, K. Ando, et al. Optimal DSP-based motion estimation tools implementation for H.264/AVC baseline encoder. IJCSNS, 7(5):141, 2007.
[20] E. Petit M. Chen J. Clay M. Rullgard R. Czyz C. Heine A. Izvorski L. Aimar, L. Merritt and A. Wright. VideoLAN. X.264 group.
[21] LFree Software Foundation. GNU operating system. http://www.gnu.org/.
[22] J.V. Team. Draft ITU-T recommendaation and final draft international standard of joint video specification. March 2003.
[23] A. Ahmad, N. Khan, S. Masud, and MA Maud. Efficient block size selection in H.264 video coding standard. Electronics Letters, 40(1):19–21, 2004.
[24] B. Zeng R. Li and M.-L. Liou. A new three-step search algorithm for block motion estimation. In Circuits and Systems for Video Technology, IEEE Transactions on, volume 4, pages 438–442, August 1997.
[25] L.-M. Po and W. C. Ma. A novel four-step search algorithm for fast block motion estimation. Circuits and Systems for Video Technology, IEEE Transactions on, 6(3):313–317, 1996.
[26] A. Puri, H.-M. Hang, and D. Schilling. An efficient block-matching algorithm for motioncompensated coding. Acoustics, Speech, and Signal Processing, IEEE International Conference on ICASSP’87., 12, 1987.
[27] S. Zhu and K.-K. Ma. A new diamond search algorithm for fast block-matching motionestimation. Image Processing, IEEE Transactions on, 9(2):287–290, 2000.
[28] C. Zhu, X. Lin, and L.-P. Chau. Hexagon-based search pattern for fast block motion estimation. Circuits and Systems for Video Technology, IEEE Transactions on, 12(5):349–355, 2002.
[29] G. Bjontegaard. Responce to call for H.26L. In ITU-T SG16, volume 12, pages 349–355,February 2002.
[30] Joint Model reference software. http://www.iphome.hhi.de/suehring/tml/index.htm.
[31] Google Video. http://video.google.com/.
[32] MobileASL. http://mobileasl.cs.washington.edu/.
[33] Speed Demos Archive. http://speeddemosarchive.com/.
[34] TASvideos. http://tasvideos.org/.
[35] Avail Media. http://www.availmedia.com/.
[36] CIF sample. http://www.tkn.tu-berlin.de/research/evalvid/cif.html.
[37] Texas Instrument. TMS320C6000 Assembly Language Tools v6.0 Beta. 2005.
[38] Texas Instrument. Code Composer Studio User’s Guide. 2000.
[39] Texas Instrument. TMS320C64x/C64x+ DSP CPU and instruction set reference guide.
[40] Texas Instrument. Video Encoding Optimization on TMS320DM64x/C64x. 2004.
[41] Texas Instrument. TMS320C64x DSP Two-Level Internsl Memory Rfference Guide.2006.
[42] C.-Y. Chang, C.-H. Pan, and H. Chen. Fast mode decision for P-frames in H. 264. Picture Coding Symposium (PCS), 2004.
[43] Texas Instruments. http://www.ti.com/.
[44] Texas Instrument. TMS320DM642 Video Port Mini-Driver. 2003.
[45] H. Schwarz, D. Marpe, and T. Wiegand. Overview of the Scalable Video Coding Extension of the H. 264/AVC Standard. IEEE Transactions on Circuits and Systems for Video Technology, 17(9):1103–1129, 2007.