JPEG2000软硬件优化技术的研究
摘要
JPEG2000是联合图片专家组(Joint Photographic Experts Group, JPEG)于2000年制定的一个静止图像压缩编码国际标准,与JPEG相比,它具有压缩率高、抗干扰性强、可选择有损和无损压缩、多分辨率表示、感兴趣区域访问等特点。但是,目前JPEG2000尚不成熟。首先,其标准并未完成,还在不断补充之中;其次,由于JPEG2000的核心编解码器采用小波变换、算术编码及嵌入式分层组织,较以往的静止图像压缩标准复杂度增加很多,目前的实现技术很难做到与JPEG同等的处理速度。
为了解决JPEG2000系统实现中的问题,本文对JPEG2000标准进行了深入分析,研究了其主要算法,找出了性能瓶颈,提出了软件和硬件两个方面的优化方法,并予以实现。
软件方面,首先利用Intel Pentium 4平台的SIMD扩展指令,研究了多分量颜色变换、离散小波变换的多种加速方法。其次,通过对该平台存储子系统的研究,提出了针对离散小波变换的分块和预取优化方法。
对于难以通过软件获得性能提升的算术编码部分,设计并实现了基于VLSI的硬件结构,达到了加速的目的。本文对于每种优化方法都给出了性能分析和实验结果。
JPEG2000 is a still image compression standard presented by Joint Photographic Experts Group in 2000. It provides many advantages over the previous JPEG, such as high compression efficiency, error resilience, lossy and lossless compression, multiple resolution representation, region-of-interest (ROI). But today JPEG2000 has not gain its success. The standard is not completely done and is still being complemented. Besides, the core codec implicates discrete wavelet transform, arithmetic coder and embedded bit-stream, which are more complicated than the previous standards. It is unlikely to achieve the same speed with JPEG based on today’s implementations.
Provided the problems above, this paper makes a research of the JPEG2000 standard and its encoding flow. Both software and hardware optimizations are made.
Software works are based upon Intel Pentium 4 platform. By using SIMD extensions, multi-component transform and discrete wavelet transform are accelerated. By researching the memory and cache subsystem, blocking and prefetching methods are proposed in the target of enhancing the proformance of discrete wavelet transform.
For the arithmetic coder which is hard to gain speedup by software, a VLSI design is presented, and comes to the purpose of acceleration. For each method in this paper, performance result is analyzed.
为了解决JPEG2000系统实现中的问题,本文对JPEG2000标准进行了深入分析,研究了其主要算法,找出了性能瓶颈,提出了软件和硬件两个方面的优化方法,并予以实现。
软件方面,首先利用Intel Pentium 4平台的SIMD扩展指令,研究了多分量颜色变换、离散小波变换的多种加速方法。其次,通过对该平台存储子系统的研究,提出了针对离散小波变换的分块和预取优化方法。
对于难以通过软件获得性能提升的算术编码部分,设计并实现了基于VLSI的硬件结构,达到了加速的目的。本文对于每种优化方法都给出了性能分析和实验结果。
JPEG2000 is a still image compression standard presented by Joint Photographic Experts Group in 2000. It provides many advantages over the previous JPEG, such as high compression efficiency, error resilience, lossy and lossless compression, multiple resolution representation, region-of-interest (ROI). But today JPEG2000 has not gain its success. The standard is not completely done and is still being complemented. Besides, the core codec implicates discrete wavelet transform, arithmetic coder and embedded bit-stream, which are more complicated than the previous standards. It is unlikely to achieve the same speed with JPEG based on today’s implementations.
Provided the problems above, this paper makes a research of the JPEG2000 standard and its encoding flow. Both software and hardware optimizations are made.
Software works are based upon Intel Pentium 4 platform. By using SIMD extensions, multi-component transform and discrete wavelet transform are accelerated. By researching the memory and cache subsystem, blocking and prefetching methods are proposed in the target of enhancing the proformance of discrete wavelet transform.
For the arithmetic coder which is hard to gain speedup by software, a VLSI design is presented, and comes to the purpose of acceleration. For each method in this paper, performance result is analyzed.
引文
[1] ISO/IEC JPEG committee, JPEG 2000 image coding system—ISO/IEC 15444-1: 2000, Dec 2000
[2] M. W. Marcellin, M. J. Gormish, A. Bilgin, and M. P.Boliek, An Overview of JPEG 2000, In Proc. DataCompression Conf., March 2000
[3] J.M.Shapiro, Embedded image coding using zerotrees of wavelets coefficients, IEEE Trans, Signal Processing, Dec 1993
[4] A.Said, W.Pearlman, A new fast and efficient image codec based on set partitioning in hierarchical trees, IEEE Trans. Circ. Syst. Video Tech, 1996
[5] John L.Hennessy and David A. Patterson, Computer Architecture: A Quantitave Approach, Third Edition, Elsvier science (USA), 2003
[6] David S.Taubman and Michael W.Marcellin, JPEG2000 Image Compression Fundamentals, Standards and Practice, Kluwer Academic Publishers, 2001
[7] Michael D.Ciletti, Advanced Digital Design with the Verilog HDL, Prentice Hall, 2003
[8] I.Daubechies and W.Sweldens, Factoring Wavelet Transforms into Lifting Steps, Journal of Fourier Analysis and Applications, 1998
[9] W.Sweldens, The Lifting Scheme: A Custom-Design Construction of Biorthogonal Wavelets. Journal of Applied and Computational Harmonic Analysis, 1996
[10] D.Chaver, C.Tenllado, L.Pinuel, M.Prieto and F.Tirado, Wavelet Transform for Large Scale Image Processing on Modern Microprocessors
[11] D. Chaver, M. Prieto, L. Pi?uel, F. Tirado, Parallel Wavelet Transform for Large ScaleImage Processing, Proceedings of the International Parallel and Distributed Processing Symposium (IPDPS'2002). Florida, USA, April 2002
[12] S. Chatterjee and C. D. Brooks, Cache-Efficient Wavelet Lifting in JPEG 2000, Proc. IEEE Int. Conf. on Multimedia, pages 797-800, August 2002
[13] Intel Corporation, IA-32 Intel Architecture Software Developer’s Manual volume 3: System Programming Guide, 2005, Order Number: 248966-012
[14] Intel Corporation, IA-32 Intel Architecture Optimization Reference Manual, 2005, Order Number: 253668-016
[15] Intel Corporation, IA-32 Intel Architecture Software Developer’s Manual volume 2:Instruction Set Reference, 2005, Order Number: 253666-016, 253667-016
[16] Intel Corporation, Desktop Performance and Optimization for Intel Pentium 4 processor, 2001, Order Number: 249438-01
[17] Rade Kutil, A Single-Loop Approach to SIMD Parallelization of 2-D Wavelet Lifting, Porceedings of the 14th Euromicro Int. Conf. on Parallel, Distributed, and Network-Based Processing, 2006
[18] R.Kutil, P.Eder and M.Watzl, SIMD parallelization of common wavelet filters, Parallel Numerics ’05, Apr, 2005
[19] P.Meerwald, R.Norcen and A.Uhl, Cache isuus with JPEG2000 wavelet lifting, Visual Communications and Image Processing 2002
[20] D.Taubman, High performance scalable image compression with EBCOT, IEEE Trans. on Image Processing, 2000
[21] D. Chaver, C. Tenllado, L. Pinuel, M. Prieto, and F. Tirado, 2-D Wavelet Transform Enhancement on General-Purpose Microprocessors: Memory Hierarchy and SIMD Parallelism Exploitation. In Proc. Int. Conf. on the High Performance Computing, December 2002
[22] C. Chakrabarti and C. Mumford, Efficient realizations of encoders and decoders based on the 2-D discrete wavelet transform, IEEE Trans. VLSI Syst., pp. 289-298, September 1999
[23] Asadollah Shahbahrami, Ben Juurlink & Stamatis Vassiliadis, Improving the Memory Behavior of Vertical Filtering in the Discrete Wavelet Transform, CF06, May, 2006
[24] Hironori Komi and Antonio Ortega, Analysis of Cache Efficiency in 2D Wavelet Transform, IEEE Int. Conf. on Multimedia and Expo, 2001
[25] S.Coleman and K.S.McKinley, Tile size selection using cache orgnization and data layout, ACM SIG-PLAN, 1995
[26] An Introduction to Arighmetic Coding, IBM J. Res. Develop, vol28, 1984
[27] Optimal hardware and software arithmetic coding procedures for the Q-coder, IBM J. Res. Develop, vol. 32, Nov. 1998
[28] Grzegorz Pastuszak, A novel architecture of arithmetic coder in JPEG2000 based on parallel symbol encoding, Proceedings of the Int. Conf. on Parallel Computing in Electrical Engineering, 2004
[29] G.K.Wallace, The JPEG still-picture compression standard, Communications of ACM, Vol.34, April 1991
[30] 吴金, 齐欢,Pentium 4 处理器的内存层次分析, 微机发展, 2004
[31] 夏宇闻, Verilog 数字系统设计教程, 北京航空航天大学出版社, 2003
[32] 胡栋, 静止图像编码的基本方法与国际标准, 北京邮电大学出版社, 2003
[33] 钟广军, 成礼智, 陈火旺, 基于提升方法的简单 9/7 小波滤波器, 计算机工程与科学, 2001
[34] 吴乐南, 数据压缩, 电子工业出版社,2000
[35] 张旭东, 卢建国, 冯健, 图像编码基础和小波压缩技术—原理、算法和标准, 清华大学出版社, 2004
[36] Michael D.Adams, ISO/IEC JTC 1/SC 29/WG1(ITU-T SG 16) N2415-Jasper Software Reference Manual, http://www.ece.uvic.ca/~mdadams, 2003
[37] Yannick Verschueren, OpenJPEG Documentation, http://www.openjpeg.org, Dec. 2003
[2] M. W. Marcellin, M. J. Gormish, A. Bilgin, and M. P.Boliek, An Overview of JPEG 2000, In Proc. DataCompression Conf., March 2000
[3] J.M.Shapiro, Embedded image coding using zerotrees of wavelets coefficients, IEEE Trans, Signal Processing, Dec 1993
[4] A.Said, W.Pearlman, A new fast and efficient image codec based on set partitioning in hierarchical trees, IEEE Trans. Circ. Syst. Video Tech, 1996
[5] John L.Hennessy and David A. Patterson, Computer Architecture: A Quantitave Approach, Third Edition, Elsvier science (USA), 2003
[6] David S.Taubman and Michael W.Marcellin, JPEG2000 Image Compression Fundamentals, Standards and Practice, Kluwer Academic Publishers, 2001
[7] Michael D.Ciletti, Advanced Digital Design with the Verilog HDL, Prentice Hall, 2003
[8] I.Daubechies and W.Sweldens, Factoring Wavelet Transforms into Lifting Steps, Journal of Fourier Analysis and Applications, 1998
[9] W.Sweldens, The Lifting Scheme: A Custom-Design Construction of Biorthogonal Wavelets. Journal of Applied and Computational Harmonic Analysis, 1996
[10] D.Chaver, C.Tenllado, L.Pinuel, M.Prieto and F.Tirado, Wavelet Transform for Large Scale Image Processing on Modern Microprocessors
[11] D. Chaver, M. Prieto, L. Pi?uel, F. Tirado, Parallel Wavelet Transform for Large ScaleImage Processing, Proceedings of the International Parallel and Distributed Processing Symposium (IPDPS'2002). Florida, USA, April 2002
[12] S. Chatterjee and C. D. Brooks, Cache-Efficient Wavelet Lifting in JPEG 2000, Proc. IEEE Int. Conf. on Multimedia, pages 797-800, August 2002
[13] Intel Corporation, IA-32 Intel Architecture Software Developer’s Manual volume 3: System Programming Guide, 2005, Order Number: 248966-012
[14] Intel Corporation, IA-32 Intel Architecture Optimization Reference Manual, 2005, Order Number: 253668-016
[15] Intel Corporation, IA-32 Intel Architecture Software Developer’s Manual volume 2:Instruction Set Reference, 2005, Order Number: 253666-016, 253667-016
[16] Intel Corporation, Desktop Performance and Optimization for Intel Pentium 4 processor, 2001, Order Number: 249438-01
[17] Rade Kutil, A Single-Loop Approach to SIMD Parallelization of 2-D Wavelet Lifting, Porceedings of the 14th Euromicro Int. Conf. on Parallel, Distributed, and Network-Based Processing, 2006
[18] R.Kutil, P.Eder and M.Watzl, SIMD parallelization of common wavelet filters, Parallel Numerics ’05, Apr, 2005
[19] P.Meerwald, R.Norcen and A.Uhl, Cache isuus with JPEG2000 wavelet lifting, Visual Communications and Image Processing 2002
[20] D.Taubman, High performance scalable image compression with EBCOT, IEEE Trans. on Image Processing, 2000
[21] D. Chaver, C. Tenllado, L. Pinuel, M. Prieto, and F. Tirado, 2-D Wavelet Transform Enhancement on General-Purpose Microprocessors: Memory Hierarchy and SIMD Parallelism Exploitation. In Proc. Int. Conf. on the High Performance Computing, December 2002
[22] C. Chakrabarti and C. Mumford, Efficient realizations of encoders and decoders based on the 2-D discrete wavelet transform, IEEE Trans. VLSI Syst., pp. 289-298, September 1999
[23] Asadollah Shahbahrami, Ben Juurlink & Stamatis Vassiliadis, Improving the Memory Behavior of Vertical Filtering in the Discrete Wavelet Transform, CF06, May, 2006
[24] Hironori Komi and Antonio Ortega, Analysis of Cache Efficiency in 2D Wavelet Transform, IEEE Int. Conf. on Multimedia and Expo, 2001
[25] S.Coleman and K.S.McKinley, Tile size selection using cache orgnization and data layout, ACM SIG-PLAN, 1995
[26] An Introduction to Arighmetic Coding, IBM J. Res. Develop, vol28, 1984
[27] Optimal hardware and software arithmetic coding procedures for the Q-coder, IBM J. Res. Develop, vol. 32, Nov. 1998
[28] Grzegorz Pastuszak, A novel architecture of arithmetic coder in JPEG2000 based on parallel symbol encoding, Proceedings of the Int. Conf. on Parallel Computing in Electrical Engineering, 2004
[29] G.K.Wallace, The JPEG still-picture compression standard, Communications of ACM, Vol.34, April 1991
[30] 吴金, 齐欢,Pentium 4 处理器的内存层次分析, 微机发展, 2004
[31] 夏宇闻, Verilog 数字系统设计教程, 北京航空航天大学出版社, 2003
[32] 胡栋, 静止图像编码的基本方法与国际标准, 北京邮电大学出版社, 2003
[33] 钟广军, 成礼智, 陈火旺, 基于提升方法的简单 9/7 小波滤波器, 计算机工程与科学, 2001
[34] 吴乐南, 数据压缩, 电子工业出版社,2000
[35] 张旭东, 卢建国, 冯健, 图像编码基础和小波压缩技术—原理、算法和标准, 清华大学出版社, 2004
[36] Michael D.Adams, ISO/IEC JTC 1/SC 29/WG1(ITU-T SG 16) N2415-Jasper Software Reference Manual, http://www.ece.uvic.ca/~mdadams, 2003
[37] Yannick Verschueren, OpenJPEG Documentation, http://www.openjpeg.org, Dec. 2003