臺灣博碩士論文加值系統

隨著網際網路的蓬勃發展與數位相機的廣泛使用，靜態數位影像大量的使用於傳輸與儲存媒介。JPEG-2000是一個新制訂完成的靜態影像壓縮標準，它比目前通用的JPEG標準有更高的壓縮效能，並且可以支援許多新且實用的功能。因此，JPEG-2000的硬體架構設計已成為數位相機的關鍵技術。
在這本論文中，我們提出了一個高效能的JPEG-2000 EBCOT方塊編碼器硬體架構。在這個硬體架構之中，我們根據EBCOT演算法的特性，設計了兩種加速方法並使用管線化的技巧。這些加速方式使得方塊編碼器所需的運算週期數可以降低到原本的40%。方塊編碼器可以分為兩個主要部分：context產生器與算術編碼器。在context產生器中，主要的加速想法是將不需要壓縮的點快速的跳過。我們使用以「行」為基礎的運算，並配合兩種加速方式來達成。而在算術編碼器中，則使用了一個四級的管線來降低最長路徑所需的運算時間，並且使用look-ahead的技巧，以達到每一個週期壓縮一組context。
我們完成了一個原型晶片以驗證架構的正確性。此晶片以COMPASS 0.35um標準單元合成，使用台積電0.35um 1P4M製程下線製作，面積是3.67x3.67mm。模擬速度可達50MHz，每秒可處理四百六十萬像素，相當於每秒壓縮一張2400x1800的影像，或是每秒壓縮30張452x340的影片。

As the rapid increase of Internet demanding and the use of digital still camera (DSC), still image is broadly used as a storage and transmission media. JPEG-2000 is a new still image compression standard. It has better compression performance than conventional JPEG standard, and it provides many useful features. The hardware implementation of JPEG-2000, therefore, becomes essential technique of digital still camera.
In this thesis, a high performance hardware architecture design of EBCOT block coder for JPEG-2000 is proposed. Speedup methods and pipelining technique are adopted according to the characteristic of EBCOT block coding algorithm. By using this architecture, process time can be reduced to about 40% of previous work. There are two major parts in this block coder: context formation (CF) and arithmetic encoder (AE). The main idea to improve the process speed of context formation is to skip no-operation samples. We achieve this by column-based coding architecture and two speedup methods. In arithmetic encoder, a four-stage pipeline is used to reduce the clock cycle time. Besides, a look-ahead technique is used in probability table look up to process two identical contexts continuously inputted.
A prototyping chip is implemented to verify the proposed architecture. The technology used is CMOS 0.35um 1P4M technology. The area of this chip is 3.67x3.67 mm2. The clock frequency is 50 MHz. It can encode 4.6 million pixels image within 1 second, corresponding to 2400x1800 image size, or 452x340 video sequence with 30 frames per second. Therefore, this chip is compliant with the upcoming motion JPEG-2000 standard.

CHAPTER 1.INTRODUCTION1
1.1.STILL IMAGE COMPRESSION OVERVIEW1
1.2.TRADITIONAL JPEG SYSTEM3
1.3.JPEG-2000 OVERVIEW5
1.3.1.Discrete Wavelet Transform7
1.3.2.Embedded Block Coding with Optimized Truncation (EBCOT)10
1.4.JPEG-2000 VS. JPEG14
1.5.THESIS ORGANIZATION18
CHAPTER 2.OVERVIEW OF EBCOT BLOCK CODER19
2.1.CONTEXT FORMATION OF BLOCK CODING20
2.1.1.Three coding passes23
2.1.2.Four types of coding operation (primitive)27
2.2.ADAPTIVE CONTEXT-BASED ARITHMETIC ENCODER OF BLOCK CODING31
CHAPTER 3.ANALYSIS AND PROFILING OF JPEG-200036
3.1.JPEG-2000 ANALYSIS AND PROFILING36
3.2.DETAIL ANALYSIS OF EBCOT BLOCK CODER38
3.3.PREVIOUS WORK42
3.3.1.Proposal by David Taubman43
3.3.2.Proposal by Analog Devices Inc.44
3.4.COLUMN-BASED OPERATION AND TWO SPEEDUP METHODS45
3.4.1.Column-based Operation45
3.4.2.Speedup Method 1: Sample Skipping (SS)48
3.4.3.Speedup Method 2: Group-Of-Column Skipping (GOCS)50
CHAPTER 4.ARCHITECTURE DESIGN OF EBCOT BLOCK CODER56
4.1.BLOCK CODING SYSTEM56
4.2.CONTEXT FORMATION57
4.2.1.Column-based Operation and Significant Register PE59
4.2.2.Sample Skipping Speedup Method62
4.2.3.Group-of-Column Skipping63
4.2.4.Memory Arrangements65
4.3.ARITHMETIC ENCODER67
4.3.1.Architecture Comparison70
4.4.PERFORMANCE EVALUATION71
CHAPTER 5.CHIP IMPLEMENTATION76
5.1.DESIGN FLOW76
5.2.TESTING CONSIDERATION78
5.3.CHIP LAYOUT AND SPECIFICATIONS79
CHAPTER 6.CONCLUSION83
BIBLIOGRAPHY85
INDEX88

[1].H. Witten, R. M. Neal, and J. G. Cleary, “Arithmetic coding for data compression,” Communications of the ACM, vol. 30, no. 6, pp. 520-540, 1987.
[2].M. J. Slattery and J. L. Mitchell, “The Qx-coder,” IBM Journal of Research and Development, vol. 42, No. 6. 1998.
[3].Digital Compression and Coding of Continuous-Tone Still Images, ISO/IEC, International Standard DIS 10918.
[4].W. Pennebaker and J. Mitchell, JPEG: still image data compression standard, Van Nostrand Reinhold, NY 1992.
[5].JPEG-2000 Verification Model Software Version 7.2, ISO/IEC JTC1/SC29/WG1 N1699.
[6].“JPEG-2000 Verification Model Version 7.0,” ISO/IEC JTC1/SC29/WG1 N1684.
[7].“JPEG-2000 Part I Final Committee Draft Version 1.0,” ISO/IEC JTC1/SC29/WG1 N1646R.
[8].Michael D. Adams, Hong Man, Faouzi Kossentini, Touradj Ebrahimi, “JPEG 2000: The Next Generation Still Image Compression Standard,” ISO/IEC JTC1/SC29/WG1 N1734.
[9].M. Gormish, D. Lee, M. Marcellin, “JPEG 2000: Overview, Architecture, and Applications,” Proc. of IEEE International Conference on Image Processing (ICIP), Vancouver, Canada, Sep. 2000.
[10].M. Boliek, J. Houchin, G. Wu, “JPEG 2000 Next Generation Image Compression System Features and Syntax,” Proc. of IEEE International Conference on Image Processing (ICIP), Vancouver, Canada, Sep. 2000.
[11].C. Christopoulos, J. Askelof, M. Larsson, “Efficient Region Of Interest Coding Techniques in the Upcoming JPEG2000 Still Image Coding Standard, “Proc. of IEEE International Conference on Image Processing (ICIP), Vancouver, Canada, Sep. 2000.
[12].I. Moccagatta, S. Soudagar, J. Liang, H. Chen, “Error Resilient Coding in JPEG-2000 and MPEG-4,” IEEE Journal of selected areas in Commun., vol. 18, No. 6, 2000, pp. 899-914
[13].T. C. Yang, K. S, Y. Bao, C. J. Kuo, “Robust coding of images using EBCOT and RVLC,” Multimedia Signal Processing, 1999 IEEE 3rd Workshop, 1999, pp. 395 —400
[14].I. Moccagatta, M. Z. Coban, H. H. Chen, “Wavelet-Based Image Coding: Comparison of MPEG-4 and JPEG-2000,” Proc. of thirty-third Asilomar Conference on Signals, Systems, and Computers, pp. 1178-1182, 1999.
[15].“MPEG-4 Video Verification Model Version 8.0,” ISO/IEC JTC1/SC29/WG11/N1796, July 1997.
[16]. Information Technology — Lossless and Near-Lossless Compression of Continuous-Tone Still Images: Baseline, ISO/IEC, 14495-1, Dec, 1999.
[17].“PNG (Portable Network Graphics) Specification,” W3C, http://www.w3.org/TR/REC-png, Oct, 1996.
[18].P. P. Vaidyanathan, Multirate Systems and Filter Banks, Prentice Hall, Inc., 1993
[19].Khalid Sayood, Introduction to Data Compression, Morgan Kaufmann Publishers
[20].Y. Q. Shi, H. Sun, Image and Video Compression for Multimedia Engineering: Fundamentals, Algorithms, and Standards, CRC Press
[21].D. Santa-Cruz, T. Ebrahimi, “An Analytical Study of JPEG 2000 Functionalities,” Proc. of IEEE International Conference on Image Processing (ICIP), Vancouver, Canada, Sep. 2000.
[22].D. Taubman, “EBCOT: Embedded Block Coding with Optimized Truncation,” ISO/IEC JTC1/SC29/WG1 N1020R.
[23].D. Taubman, “High Performance Scalable Image Compression With EBCOT,” Proc. of IEEE International Conference on Image Processing (ICIP), Kobe, Japan, 1999, vol. 3, pp. 344—348.
[24].D. Taubman, “High Performance Scalable Image Compression With EBCOT,” IEEE Trans. On Image Processing, July, 2000.
[25].D. Taubman, E. Ordentlich, I. Ueno, “Embedded Block Coding in JPEG2000,” Proc. of IEEE International Conference on Image Processing (ICIP), Vancouver, Canada, Sep. 2000.
[26].J. M. Shapiro, “An Embedded Hierarchical Image Coder Using Zerotrees of Wavelet Coefficients,” Proc. of IEEE Data Compression Conference (Snowbird, Utah). Pp. 214-223, 1993.
[27].A. Said and W. Pearlman, “A New, Fast and Efficient Image Codec Based on Set Partitioning Iin Hierarchical Trees,” IEEE Trans. Circuits and Systems for Video Technology. pp. 243-250, June 1996.
[28].B. Fu, K. K. Parhi, “Two VLSI Design Advances in Arithmetic Coding,” IEEE Circuits and Systems Society, Proc. of ISCAS, 1995, vol. 2, pp. 1440 —1443.
[29].Po Tong, Peng Ang, “A JBIG Arithmetic Coder-Decoder Chip,” Proc. of Fifth Annual IEEE International, 1992, pp. 189—192.
[30].M. Tarui, M. Oshita, T. Onoye, I. Shirakawa, “High-Speed Implementation of JBIG Arithmetic Coder,” Proc. of the IEEE Region 10 Conference, 1999, vol. 2, pp. 1291 —1294.
[31].“Motion JPEG-2000 Verification Model Version 0.1,” ISO/IEC JTC1/SC29/WG1 N1680.
[32].T. Fukuhara, K. Katoh, S. Kimura, K. Hosaka, A. Leung, “Montion-JPEG 2000 standarization and target market,” Proc. of IEEE International Conference on Image Processing (ICIP), Vancouver, Canada, Sep. 2000.
[33].J. Yoshida, “JPEG2000 Wave Rises As ADI Shows Still-Camera Accelerator,” EETimes.Com, May 11, 2001.