臺灣博碩士論文加值系統

除了廣為人知的時間可調性（temporal scalability）、空間可調性（spatial scalability）與噪訊比可調性（SNR scalability）的特點外，隨著高畫質電視（high definition television）的發展，JSVM也開始廣邀各界投入於位元深度可調性之研究。
本論文提出三種能有效提升編碼效率的位元深度可調性之編碼架構（bit depth scalable video coding），方法一利用低位元深度影像為基底層（base layer），其透過反色調對映（inverse tone mapping）轉換的資訊可供高位元深度資訊參考，並且利用跨層級預測源以增加編碼效率，最後完成高位元深度層級的加強層（enhancement layer），此設計能讓位元流（bit-stream）達到嵌入式（embedded）的效果，傳送端根據不同的使用環境限制，可以選取部分或全部的位元流傳輸。
方法二之架構設計與目前可調性深度編碼研究的應用層面不盡相同。以高位元深度影像為基底層，經過色調對映轉換（tone mapping）後成為低位元深度影像的參考資訊，儘管提出的第二個架構編碼位元流並非嵌入式形式，不過透過方法二編碼架構的低位元深度影像品質比其單層編碼（single layer）還佳。假如傳送過程中不受頻寬限制的問題，但是接收端的設備無法顯示高位元影像，可以挑選方法二所提供的位元串來提升接收端所觀看的低位元影像品質。
方法三則是融合了方法一與方法二的特性，既能改進方法二無法提供嵌入式位元流的情況，還可以提供接收端品質較好的低位元深度影像。在不同測試條件下，透過實驗測試與數據統計有利於觀察出這三種方法之優點與特性。

Scalable video coding（SVC）is currently developed as an extension of H.264/AVC video coding standard. It aims at supporting any combination of spatial, temporal, and SNR scalability. Recently, images offering not only high definition quality, but also high dynamic range are desirable. Since the photographic color images with high dynamic range are more and more popular and easily acquired, JVT has issued a “Call for Proposals” to standardize bit-depth scalable video coding into SVC standard.
This thesis work proposed three H.264/AVC compliant bit-depth scalable video coding schemes: LH mode（Low Bit-depth to High Bit-depth）, HL mode（High Bit-depth to Low Bit-depth）and the merged LHHL mode for different applications. All of the schemes efficiently exploit the inter-layer relationship between the high bit-depth layer and the low bit-depth layer on Macroblock level
For LH mode, the bit-stream is generated embeddable; it means that the generated bitstream is backwards compatible to H.264 and it could be extended to a higher bit-depth enhancement layer. According to the channel condition, transmitter would truncate the bit-stream. For example, it is possible to delivery the low bit-depth bitstream（typically 8 bit-depth）only, or transmit all the whole bit-stream to reconstruct both bit-depth sequences, without any truncation. To improve the coding efficiency, information from low bit-depth sequence, such as residual data, reconstructed textures are processed by inverse tone mapping and can be regarded as inter-layer prediction information.
On the other hand, HL mode or LHHL mode contain distinct architecture. The inter-layer prediction is obtained through the process of tone mapping on high bit-depth information. According to the experimental results, HL mode and LHHL mode outperforms LH mode and traditional simulcast coding scheme.

致謝辭 II
中文摘要 III
英文摘要 IV
目錄 VI
圖目錄 IX
表目錄 XI
第一章 緒論 1
1.1 簡介 1
1.2 研究動機與目的 1
1.3 論文架構 2
第二章 高動態範圍影像簡介 3
2.1 高動態範圍影像（High Dynamic Range Image） 3
2.2 色調對映技術（Tone Mapping） 5
2.3 高動態範圍影像編碼之相關研究 7
第三章 可調性視訊編碼簡介及位元深度視訊編碼之相關研究 10
3.1 可調性視訊編碼（Scalable Video Coding） 10
3.1.1 空間可調性（Spatial Scalability） 11
3.1.2 時間可調性（Temporal Scalability） 12
3.1.3 品質可調性（SNR Scalability） 13
3.1.4 位元深度可調性（Bit-depth Scalability） 14
3.2 可調性視訊編碼架構介紹 15
3.2.1 JSVM編碼器的概觀結構 15
3.2.2 跨層級之估測（Inter-layer Prediction） 17
3.3 位元深度可調性編碼之相關研究 19
3.3.1 研究方法一 19
3.3.2 研究方法二 20
第四章 提出的位元深度可調性之視訊編碼 22
4.1 低位元深度影像為預測源（Low Bit-depth to High Bit-depth） 22
4.1.1 跨層級估測方法 22
4.1.2 位元串結構（Bit-stream） 24
4.2 高位元深度影像為預測源（High Bit-depth to Low Bit-depth） 25
4.2.1 跨層級估測方法 25
4.2.2 位元串結構 27
4.3 結合LH mode與 HL mode之整合型預測源（LHHL mode） 27
4.3.1 跨層級估測方法 27
4.3.2 位元串結構 27
4.4 色調對映技術與反色調對映技術 28
4.4.1 色調對映法 28
4.4.2 反色調對映法 30
第五章 實驗結果與數據討論 31
5.1 色調對映轉換成果 31
5.2 I-frame測試 32
5.3 LH mode之P-frame測試 34
5.4 HL mode之P-frame測試 38
5.5 LHHL mode之P-frame測試 39
5.6 LH mode與其他論文研究測試結果比較 40
5.7 不同高位元深度之編碼效果比較 42
5.8 低位元深度層之實驗結果比較 45
5.9 高、低位元深度層使用不同QP進行編碼測試 47
5.10 LH mode移動向量搜尋與Residual prediction順序先後之編碼測試 50
第六章 結論與未來展望 52
參考文獻 53

[1]Y.Wang, J.Ostermann and Y.Zhang., Video processing and communications. prentice hall. 2002.
[2]J.Reichel, H.Schwarz, and M.Wien (eds.), “Scalable Video Coding – Joint Draft 9,” Joint Video Team, Doc. JVT-V201, Marrakech, Jan. 2007.
[3]J.Reichel, H.Schwarz, and M.Wien (eds.), “Joint Scalable Video Model JSVM-9,” Joint Video Team, Doc. JVT-V202, Klagenfurt, Marrakech, Jan. 2007.
[4]ISO/IEC and ISO/IEC JTC1, “JSVM 9 Software,” JVT-V203, Klagenfurt, Marrakech, Jan. 2007.
[5]H.Schwarz, and M.Wien, J.Vieron (eds.), “JSVM Software Manual,” including in [4].
[6]ITU-T Rec. H.264 & ISO/IEC 14496-10 AVC, “Advance Video Coding for Generic Audiovisual Services,” 2003.
[7]H. Seetzen, W. Heidrich, W. Stuezlinger, G. Ward, L. Whitehead, M. Trentacoste, A. Ghosh, A. Vorozcovs, “High Dynamic Range Display Systems,” ACM Transactions on Graphics, Vol. 23, No. 3, August 2004.
[8]H. Seetzen, L.Whitehead and G. Ward, “High Dynamic Range Displays Using Low and High Resolution Modulators,” The Society for Information Display International Symposium, Baltimore, MD May 2003.
[9]Thomson images & beyond website, http://www.thomsongrassvalley.com.
[10]Greg Ward, “Real Pixels,” Graphics Gems II, Ed. by James Arvo, Academic Press, pp.80-83, 1991.
[11]Greg Ward, “The RADIANCE Lighting Simulation and Rendering System,” ACM Computer Graphics Proceedings of SIGGRAPH 94, 1994.
[12]Greg Ward Larson, “The LogLuv Encoding for Full Gamut, High Dynamic Range Images,” Journal of Graphics Tools, Vol. 3, No. 1, pp.15-31,1998.
[13]Greg Ward, “High Dynamic Range Image Examples,” Dec. 14, 2003, http://www.anywhere.com/gward/hdrenc/pages/originals.html.
[14]K. Devlin, “A Review of Tone Reproduction Techniques,” Technical Report CSTR-02-005, Computer Science, Univ. of Bristol, 2002.
[15]Sumanta N. Pattanaik, James A. Ferwerda, Mark D. Fairchild, and Donald P. Greenberg, “A Multiscale Model of Adaptation and Spatial Vision for Realistic Image Display,” In Proceedings of SIGGRAPH 98, pp.287–298, July 1998.
[16]Jack Tumblin, Jessica K. Hodgins, and Brian K. Guenter, “Two Methods for Display of High Contrast Images,” ACM Transactions on Graphics, Vol. 18, No. 1, pp.56-94, January 1999.
[17] M. Ashikhmin, “A Tone Mapping Algorithm for High Contrast Images,” In 13th Eurographics Workshop on Rendering. Eurographics, June 2002.
[18]Erik Reinhard, Michael Stark, Peter Shirley, and Jim Ferwerda, “Photographic Tone Reproduction for Digital Images,” In Proceedings of ACM SIGGRAPH, July 2002.
[19]Greg Ward, Maryann Simmons, “JPEG-HDR: A Backwards-compatible, High Dynamic Range Extension to JPEG,” In Proceedings of the 13th Color Imaging Conference, pp.283-290, May 2005 .
[20]Martin Winken, Detlev Marpe, Heiko Schwarz, and Thomas Wiegand, “Bit-depth Scalable Video Coding ,” In IEEE International Conference on Image Processing, 2007.
[21]Andrew Segall, “Scalable Coding of High Dynamic Range Video,” In IEEE International Conference on Image Processing, 2007.
[22]Yuwen Wu, Yongying Gao, and Ying Chen, “Bit Depth Scalable Coding,” In IEEE ICME, pp.1139-1142 July 2007.
[23]ISO/IEC and ISO/IEC JTC1, “Donation of Tone Mapped Image Sequences,” JVT-Y072, Shenzhen, China, October, 2007.