臺灣博碩士論文加值系統

現今科技進步迅速，隨著行動通訊4G/LTE漸漸普及，帶動許多新媒體相關產業的發展，影音視訊的解析度與品質的需求也隨之提升，也因此視訊在做傳輸時所需要的頻寬與編碼壓縮時的資料量也越來越大，為了在有更好的壓縮效率下能夠保有視訊的質量，演算法固然會更加的複雜。在最新的視訊編碼標準H.265/High Efficiency Video Coding (HEVC)中，在編碼單元(Coding Unit, CU)上有了更加多元的模式，除了能夠對應不同解析度之外，也支援處理更高的解析度。對於行動裝置上的視訊影像，相對來說頻寬比較有限，所以對高解析度視訊降低頻寬的處理上會成為一個主要的目標。
本論文提出基於位元率失真最佳化(Rate-Distortion Optimization, RDO)的頻寬位元失真率最佳化(Bandwidth Rate-Distortion Optimization, BRDO)演算法；此演算法根據位元失真率成本(Rate-Distortion Cost, RDCost)之大小，來進行搜尋範圍(Search Range)的分類加以限制並降低頻寬的使用量，再搭配模擬兩種不同的搜尋演算法，全域搜尋(Full Search)和測試區域搜尋(Test Zone Search)，不僅保持原有的影像品質及位元率之外，整體下來FS頻寬平均減少了46%左右，TZS頻寬平均減少了42%左右。
在硬體架構設計上是使用Synopsys(Verilog、Design Compiler、Verdi、Synthesis、PrimeTime®、PrimePower®)和Cell Library(TSMC 90nm CLN90G)實作，在最差的情況(Worst case)下仍然可以達到1.064 GHz，且功率消耗為0.9108 mW。

Thanks to the fast developing high technology nowadays, mobile telecommunication 4G/LTE is popularized worldwide, and which makes a rapidly growing New Media related Industry.With the higher requirement for good quality an-d high resolution of Video/Webcam , the bandwidth and the amount of coding compressed data for transmitting Video have to be expanded. In order to keep high performance of video under efficient data compression, more complicated mathematical calculations is a must.In the newest HEVC, CU is quite diversified in order to match different resolution requirement as well as to support higher resolution. Since the bandwidth of audio and video on mobile internet device is limited, our major target is to settle bandwidth problem on high resolution video, that is , to narrow the bandwidth.
This thesis puts forward the algorithm of Bandwidth-Rate-Distortion Optimization (BRDO), which is on basis of Rate-Distortion Optimization. The algorithm distributes bandwidth and search area by size of Rate-Distortion Cost (RDCost). It not only lowering the usage of bandwidth but maintaining quality and rate. On average, more than 56% of bandwidth's usages were saved and more than 60% of encoding time decrease largely.
The hardware architecture was implemented by using Synopsys (Verilog, Verdi, Design Compiler, Synthesis, PrimeTime®, PrimePower®) and Cell Library (TSMC 90nm CLN90G). The speed of our design was 1.1GHz under the worst case simulation case, and the power consumption was 0.873mW.

摘要 I
Abstract II
誌謝 III
目錄 IV
圖目錄 VI
表目錄 IX
第一章 導論 1
1.1 簡介 1
1.2 研究目的與動機 2
1.3 論文章節安排 2
第二章 視訊編碼系統介紹 3
2.1 HEVC視訊編碼介紹 3
2.1.1 編碼單元(Coding Unit, CU) 5
2.1.2 預測單元(Prediction Unit, PU) 6
2.1.3 轉換單元(Transform Unit, TU) 8
2.1.4 階層式編碼架構 9
2.1.5 位元率失真最佳化(Rate-Distortion Optimization) 9
2.2 內嵌式迴圈濾波器(In-loop Filter) 14
2.3 量化(Quantization) 14
2.4 熵編碼(Entropy Coding) 15
2.4.1 適應性可變長度編碼(Context Adaptive Variable Length Coding , CAVLC) ………………………………………………………………………………………………………………………….…..16
2.4.2 適應性二進位算術編碼(Context Adaptive Binary Arithmetic, CABAC)….16
2.5 畫面內預測(Intra Predicition) 17
2.6 畫面間預測(Inter Predicition) 19
2.6.1 移動估測(Motion Estimation, ME)………………………………………………………….…19
2.6.2 移動向量預測(Motion Vetor Prediction, MVP)………………………………………….21
2.6.3 全域搜尋演算法(Full search Algorithm, FS)……………………………………………..24
2.6.4 測試區搜尋演算法(Test Zone Search Algorithm, TZS)……………………………..25
2.6.5 多幅參考畫面(Multi-Reference Frame)……………………………………………………..29
2.6.6 非整數點移動估測…………………………………………………………………………………….29
第三章 頻寬分配演算法架構設計 32
3.1 記憶體頻寬分析…………………………………………………………………………………………32
3.2 頻寬位元失真率最佳化(Bandwidth-Rate-Distortion Optimization, BRDO)演算法…………………………………………………………………………………………………………………………..……..34
3.2.1 移動向量預測與頻寬計算(MVP & BW Calculation) 41
3.2.2 頻寬位元失真率分布對應RDCost之分析 43
第四章 演算法實驗結果分析 47
4.1 H.264中時域與空間域的選擇…………………………………………………………………..52
4.2 H.265之演算法實現………………………………………………………………………………….54
第五章 硬體架構設計與結果驗證 64
5.1 Cell-based Design Flow………………………………………………………………………………64
5.2 演算法之硬體架構規劃…………………………………………………………………………….65
5.3 演算法之硬體內部設計…………………………………………………………………………….68
5.3.1 RDCost Classifier模組區塊 69
5.3.2 Threshold Selection模組區塊 70
5.3.3 Search Range Allocation模組區塊 71
5.4 硬體架構實驗數據…………………………………………………………………………………….72
5.5 實驗數據比較………………………………………………………………………………………….…75
第六章 結論及延伸討論 76
參考文獻 77

[1] Draft ITU-T Recommendation and Final Draft International Standard of Joint Video Specification, 2003.
[2]G. J. Sullivan, J. R. Ohm, W. J. Han, and T. Wiegand, “Overview of the high efficiency video coding (HEVC) standard,” IEEE Trans. Circuits Syst. Video Technol., vol. 22, no. 12, pp. 1649-1668, Dec. 2012.
[3]J. Ohm , G. J. Sullivan, H. Schwarz , Thiow Keng Tan , and T. Wiegand, “Comparison of the Coding Efficiency of Video Coding Standards—Including High Efficiency Video Coding (HEVC) ,” IEEE Trans. Circuits Syst. Video Technol., vol. 22, no. 12, pp. 16691684, Dec. 2012.
[6]S. Kim , Ch. Park , H. Chun ,and J. Kim, “A novel fast and low-complexity Motion Estimation for UHD HEVC ,”IEEE Picture Coding Symposium (PCS), 2013.
[7]S. Lee “Fast motion estimation based on search range adjustment and matching point decimation , ” IET Image Process., vol. 4, no. 1, pp. 1-10, 2010.
[8]Z. He , Y. Liang , L. Chen , I. Ahmad and D. Wu “Power-rate-distortion analysis for wireless video communication under energy constraints,” IEEE Trans. Circuits Syst. Video Technol., vol. 15, no. 5, pp. 645-658, 2005.
[9] Eckart S., Fogg C. , “Iso/iec mpeg-2 software video codec, ” Proc. SPIE, vol. 2419, pp. 100– 118, 1995.
[10] A New Motion Estimation Algorithm for High Efficient Video Coding StandardIndia Conference(INDICON),2015Annual IEEE Karam Singh,Shaik Rafi Ahamed
[11] K. McCann, C. Rosewarne, B. Bross, M. Naccari, K. Sharman, G. Sullivan, “High Efficiency Video Coding (HEVC) Test Model 16 (HM 16) Improved Encoder Description,” ISO/IEC JTC1/SC29/WG11 N14970, Oct. 2014.
[12]B. Montrucchio and D. Quaglia, “New sorting-based lossless motion estimation algorithm and a partial distortion elimination performance analysis’, IEEE Trans. CSVT, vol.15, no.2, pp. 210–220, Feb. 2005.
[13] S. Zhu and K.-K. Ma, “A new diamond search algorithm for fast block matching
motion estimation,” IEEE Trans. Image Process., vol.9, no.2, pp. 287 –290, Feb.2000.
[14]Reoxiang Li, Bing Zeng, M.L. Liou, “A new three-step search algorithm for block motion estimation,” IEEE Trans. Circuits Syst. Video Technol., vol.4, no.4, pp.438-442, Aug. 1994.
[15] C. Zhu, X. Lin, and L.-P. Chau, “Hexagon-based search pattern for fast
block motion estimation,” IEEE Trans. Circuits. Syst. Video Technol., vol.12, no.5,pp.349 –355,May. 2002.
[17] N. Purnachand, L. Alves, and A. Navarro, “Fast motion estimation algorithm for hevc,” IEEE Int. Conf. Consumer Electronics – Berlin (ICCE-Berlin), 2012, pp. 34–37, Sept. 2012.
[18] J.-L. Lin , Y.-W. Chen ,Y.-P. Tsai , Y.-W. Huang, S. Lei, “Motion vector coding techniques for HEVC,” in Proc. IEEE MMSP, pp. 1-6, Oct. 2011.
[20]Q. Yu , L. Zhao , S. Ma, “Parallel AMVP candidate list construction for HEVC,” in IEEE VCIP, Nov. 2012.
[21]L. Zhao, X. Guo , S. Lei , S. Ma , D. Zhao, “Simplified AMVP for High Efficiency Video Coding,” in IEEE VCIP, Nov. 2012.
[22]M. E. Sinangil, A. P. Chandrakasan, V. Sze, M. Zhou, “Memory cost vs. coding efficiency trade-offs for HEVC motion estimation engine,” in IEEE ICIP, Oct. 2012.
[23]J.-C. Tuan, T.-S. Chang, C.-W. Jen, “ On the Data Reuse and Memory Bandwidth Analysis for Full-Search Block-Matching VLSI Architecture” IEEE Trans. Circuits Syst. Video Technol.,vol.12, no.1, pp.61-72, Jan.2002.
[24] J.-H. Hsieh, J.-H. Huang, H.-R. Wang, “DVFS-aware motion estimation design scheme based on bandwidth–rate–distortion optimization in application processor systems,” INTEGRATION the VLSI journal, Elsevier, vol.57, pp.74–80, Mar. 2017.
[25] K.-I. Lee, A.-C. Tsai, J.-F. Wang, and J.-F. Yang, "Efficient Inter Mode Prediction Based on Model Selection and Rate Feedback for H.264/AVC," IEEE Trans. Circuits Syst. Video Technol., vol. 21, pp.708-716, Jun. 2011.
[26] C.-Y. Tsai, C.-H. Chung, Y.-H. Chen, T.-C. Chen, and L.-G. Chen, “Low power cache algorithm and architecture design for fast motion estimation in H.264/AVC encoder system,” in Proc. IEEE Int. Conf. on Acoust., Speech, and Signal Process., pp. 97-100. Apr. 2007
[27] Joint Video Team Reference Software JM15.0, ITU-T Available: http://iphome.hhi.de/suehring/tml/download/
[28] JCT-VC, HM 16.6, Reference Software. [Online]. Available: https://hevc.hhi.fraunhofer.de/svn/svn_HEVCSoft ware/ tags/HM-16.6/
[29] G. Bjøntegaard, "Calculation of average PSNR differences between RDcurves," Technical Report VCEG-M33, ITU-T SG16/Q6, 2001.
[30] “Computationally Efficient HEVC/H.265 Motion Estimation Algorithm for Low Power Applications” Information and Communication(ICICT),2015 International Conference. Fauzia Yasir, Shehzad Hasan