基于小波分析的视频压缩方法研究及实现
|
摘要
数字化后的信息，尤其是数字化后的视频和音频信息具有数据海量性，给信息的存储和传输造成较大的困难。因此，研究和开发新型有效的多媒体数据压缩编码方法是非常必要的。
小波变换编码属于图像处理中的变换编码方法。它能够遵从人类视觉生理和心理特点，将原始图像在频域内作多层分解，为压缩编码方法提供理想的变换域，本课题的目标是研究基于小波分析的各种视频压缩方法及其实时性能，并在DSP板上对有关部分进行实现。课题进行了以下的工作：
●编制出了在MATLAB下运行的基于小波分解和零树编码的单帧图像压缩解压程序。零树算法是一种效果比较好的小波系数扫描编码方法，可去除不同尺度间的小波系数幅值的强相关性。
●用小波分解和SPIHT编码方法进行单帧图像压缩。SPIHT(层次树中集合划分)方法是对零树编码方法的改进，具有更高的编码效率。用C语言编制出了相关程序。
●用小波分解域内运动估计方法进行多帧图像压缩。该方法可保证帧内、帧间编码过程始终在小波分解域内进行，运动估计运算量比块匹配全搜索法小得多。在MATLAB下编制相关程序进行了性能测试。
●在TMS320C6701评估板上用小波分解和SPIHT单帧图像进行实时压缩。为了提高编解码速度，图像小波分解与合成程序完全用C6701汇编指令编写，其中大量使用并行指令和流水线优化，使运算时间大大减少。主机和DSP评估板之间通过PCI接口进行高速图像数据交换。
从运行结果可以看出，当图像压缩率较高时，使用基于小波分析图像压缩方案的解码图像能保持清晰的轮廓，无DCT块编码压的方块效应。通过设置压缩程序输入参数可无级控制图像压缩比。经过精心调试，DSP上小波分解合成程序段运行速度已达C6701上限。由于时间限制，对小波域内运动预测方法仅做了仿真与性能测试。仿真结果表明，该方法重建的解码图像整体感好，无块匹配搜索法的块轮廓，而编码速度和块对数搜索法相当。在DSP上实现小波分解+SPIHT 编码+小波域内运动预测方法进行视频压缩具有较好的实用性，如果在系统中加入实时视频信号采集，可形成完整的视频采集压缩系统。
Digital information,especially digital video and audio information,always has large amount of data which makes information storage and transportation difficult. So it's necessary to research on and develop novel effective multimedia data compressing method to solve this problem.
Wavelet transform(WT) is a kind of domain transform method in image processing which takes human visual and mental systems into consideration. By decomposing original image to multi-resolutions in frequency domain,it provides ideal transform domain for image compression. The object of this article is to research on several wavelet transform based video compressing method and their running properties. We have done the following work:
We have designed wavelet transform and zero-tree coding based still image compressing/decompressing programs in MATLAB environment. Zero-coding method is an effective scanning method of wavelet decomposed coefficients,which can eliminate correlation of wavelet coefficients in different resolutions.
We have tested wavelet transform and SPIHT coding based still image compression method. SPIHT(Set Partitioning in Hierarchical Trees) method is a more effective wavelet coefficients scanning method originated from zero-tree method. We have designed programs in C language.
We have tested multi-resolution motion estimation(MRME) technique to achieve multi-frame compression in MATLAB. In MRME,coding of intra-frame and inter-frame are done in wavelet decomposition domain. MRME costs much less time to perform motion estimation operation compared to block matching motion full search algorithm(FSA).
We have applied wavelet transform and SPIHT coding based still image compressing programs on TMS320C6701 for hardware real-time compression. In order to enhance coding/decoding speed,we have rewritten image wavelet decomposition and composition program with C6701 assembly language in which we made use of parallel instructions and pipeline optimizing skill to lessen calculating time. Image data are exchanged between host and DSP evaluation board in a high speed through PCI interface.
Experimental results show when image compressing rate is high,decoded image of wavelet transform based compressing method can preserve objects outline well and has no block effect which may appear in DCT block coding technique. Image compression rate can be changed to any given number by modifying parameter of the program. After properly designing,wavelet transform program on DSP has reached C6701's speed limit. As to MRME method,we only did simulation to get its resultant statistics because of time limitation. Simulation results show reconstructed image of this technique has good overall visual effect and has no block profile which may exist in block matching motion estimation method. MRME's time requirement approximates that of block logarithm search algorithm. We concluded that designing DSP and
wavelet transform + SPIHT coding + MRME method based video compressing system is applicable. If real-time video data collection part is added,we can get a perfect real time video collecting and compressing system.
小波变换编码属于图像处理中的变换编码方法。它能够遵从人类视觉生理和心理特点，将原始图像在频域内作多层分解，为压缩编码方法提供理想的变换域，本课题的目标是研究基于小波分析的各种视频压缩方法及其实时性能，并在DSP板上对有关部分进行实现。课题进行了以下的工作：
●编制出了在MATLAB下运行的基于小波分解和零树编码的单帧图像压缩解压程序。零树算法是一种效果比较好的小波系数扫描编码方法，可去除不同尺度间的小波系数幅值的强相关性。
●用小波分解和SPIHT编码方法进行单帧图像压缩。SPIHT(层次树中集合划分)方法是对零树编码方法的改进，具有更高的编码效率。用C语言编制出了相关程序。
●用小波分解域内运动估计方法进行多帧图像压缩。该方法可保证帧内、帧间编码过程始终在小波分解域内进行，运动估计运算量比块匹配全搜索法小得多。在MATLAB下编制相关程序进行了性能测试。
●在TMS320C6701评估板上用小波分解和SPIHT单帧图像进行实时压缩。为了提高编解码速度，图像小波分解与合成程序完全用C6701汇编指令编写，其中大量使用并行指令和流水线优化，使运算时间大大减少。主机和DSP评估板之间通过PCI接口进行高速图像数据交换。
从运行结果可以看出，当图像压缩率较高时，使用基于小波分析图像压缩方案的解码图像能保持清晰的轮廓，无DCT块编码压的方块效应。通过设置压缩程序输入参数可无级控制图像压缩比。经过精心调试，DSP上小波分解合成程序段运行速度已达C6701上限。由于时间限制，对小波域内运动预测方法仅做了仿真与性能测试。仿真结果表明，该方法重建的解码图像整体感好，无块匹配搜索法的块轮廓，而编码速度和块对数搜索法相当。在DSP上实现小波分解+SPIHT 编码+小波域内运动预测方法进行视频压缩具有较好的实用性，如果在系统中加入实时视频信号采集，可形成完整的视频采集压缩系统。
Digital information,especially digital video and audio information,always has large amount of data which makes information storage and transportation difficult. So it's necessary to research on and develop novel effective multimedia data compressing method to solve this problem.
Wavelet transform(WT) is a kind of domain transform method in image processing which takes human visual and mental systems into consideration. By decomposing original image to multi-resolutions in frequency domain,it provides ideal transform domain for image compression. The object of this article is to research on several wavelet transform based video compressing method and their running properties. We have done the following work:
We have designed wavelet transform and zero-tree coding based still image compressing/decompressing programs in MATLAB environment. Zero-coding method is an effective scanning method of wavelet decomposed coefficients,which can eliminate correlation of wavelet coefficients in different resolutions.
We have tested wavelet transform and SPIHT coding based still image compression method. SPIHT(Set Partitioning in Hierarchical Trees) method is a more effective wavelet coefficients scanning method originated from zero-tree method. We have designed programs in C language.
We have tested multi-resolution motion estimation(MRME) technique to achieve multi-frame compression in MATLAB. In MRME,coding of intra-frame and inter-frame are done in wavelet decomposition domain. MRME costs much less time to perform motion estimation operation compared to block matching motion full search algorithm(FSA).
We have applied wavelet transform and SPIHT coding based still image compressing programs on TMS320C6701 for hardware real-time compression. In order to enhance coding/decoding speed,we have rewritten image wavelet decomposition and composition program with C6701 assembly language in which we made use of parallel instructions and pipeline optimizing skill to lessen calculating time. Image data are exchanged between host and DSP evaluation board in a high speed through PCI interface.
Experimental results show when image compressing rate is high,decoded image of wavelet transform based compressing method can preserve objects outline well and has no block effect which may appear in DCT block coding technique. Image compression rate can be changed to any given number by modifying parameter of the program. After properly designing,wavelet transform program on DSP has reached C6701's speed limit. As to MRME method,we only did simulation to get its resultant statistics because of time limitation. Simulation results show reconstructed image of this technique has good overall visual effect and has no block profile which may exist in block matching motion estimation method. MRME's time requirement approximates that of block logarithm search algorithm. We concluded that designing DSP and
wavelet transform + SPIHT coding + MRME method based video compressing system is applicable. If real-time video data collection part is added,we can get a perfect real time video collecting and compressing system.
引文
1．闫敬文，卢刚，王超，郭子祺，基于形态学小波变换的SAR图像数据海洋目标提取，《工程图学学报》，2001年增刊
2．卢刚，闫敬文，用C542 DSK实现信号倒谱计算，《厦门大学学报(自然科学版)》，Vol．40(5)，PP．1056-1061，2001
3．郭子祺，卢刚，王超，潘广东，海洋SAR图像小波Speckle滤波及边缘信息提取，《遥感学报》，Vol．5(6)，pp．428-433，2001
4．闫敬文，王超，卢刚，郭子祺，一种基于小波变换的SAR海洋图像数据增强系统，《海洋学报》，Vol．23(5)，pp．130-135，2001
5．(TMS320C6000系列DSPs的原理与应用》，任丽香、马淑芬、李方慧编著，电子工业出版社
6．《基于对象的多媒体数据压缩编码国际标准MPEG-4及其校验模型》，钟玉琢、王琪、贺玉文编著，科学出版社
7．《WT6201／6701PA板用户使用说明书》，北京闻亭科技发展有限责任公司编著
8. Yan Jingwen, Lu Gang, Lu Hanqing, SAR image enhancement technique based on morphological wavelet transformation, SPIE, Vol.4548
9. Guo Ziqi, Lu Gang, Using wavelet for filtering speckle and extracting edge from SAR image,SPIE, Vol.4550
10. Yan Jingwen, Shen Guiming, Lu Gang, Improved bi-block zero tree coding for video data compression, SPIE, Vol.4551
11. Y. Q. Zhang and S. Zafar, Motion-Compensated Wavelet Transform Coding for Color Video Compression, IEEE Transactions on Circuits and Systems for Video Technology, Vol.2, No.3,pp.285-296(Sept 1992)
12. TMS320C6000 Peripherals, Texas Instruments Incorporated, April 1999
13. TMS320C62x/C67x CPU and Instruction Set, Texas Instruments Incorporated, Mar. 1998
14. TMS320C6000 Optimizing C Compiler, Texas Instruments Incorporated, Feb. 1999
15. TMS320C6x Assembly Language Tools, Texas Instruments incorporated, Feb. 1998
16. TMS320C62x/C67x Programmer's Guide, Texas Instruments Incorporated, Feb. 1998
17. TMS320C6x Peripheral Support Library, Texas Instruments Incorporated, Mar. 1998
2．卢刚，闫敬文，用C542 DSK实现信号倒谱计算，《厦门大学学报(自然科学版)》，Vol．40(5)，PP．1056-1061，2001
3．郭子祺，卢刚，王超，潘广东，海洋SAR图像小波Speckle滤波及边缘信息提取，《遥感学报》，Vol．5(6)，pp．428-433，2001
4．闫敬文，王超，卢刚，郭子祺，一种基于小波变换的SAR海洋图像数据增强系统，《海洋学报》，Vol．23(5)，pp．130-135，2001
5．(TMS320C6000系列DSPs的原理与应用》，任丽香、马淑芬、李方慧编著，电子工业出版社
6．《基于对象的多媒体数据压缩编码国际标准MPEG-4及其校验模型》，钟玉琢、王琪、贺玉文编著，科学出版社
7．《WT6201／6701PA板用户使用说明书》，北京闻亭科技发展有限责任公司编著
8. Yan Jingwen, Lu Gang, Lu Hanqing, SAR image enhancement technique based on morphological wavelet transformation, SPIE, Vol.4548
9. Guo Ziqi, Lu Gang, Using wavelet for filtering speckle and extracting edge from SAR image,SPIE, Vol.4550
10. Yan Jingwen, Shen Guiming, Lu Gang, Improved bi-block zero tree coding for video data compression, SPIE, Vol.4551
11. Y. Q. Zhang and S. Zafar, Motion-Compensated Wavelet Transform Coding for Color Video Compression, IEEE Transactions on Circuits and Systems for Video Technology, Vol.2, No.3,pp.285-296(Sept 1992)
12. TMS320C6000 Peripherals, Texas Instruments Incorporated, April 1999
13. TMS320C62x/C67x CPU and Instruction Set, Texas Instruments Incorporated, Mar. 1998
14. TMS320C6000 Optimizing C Compiler, Texas Instruments Incorporated, Feb. 1999
15. TMS320C6x Assembly Language Tools, Texas Instruments incorporated, Feb. 1998
16. TMS320C62x/C67x Programmer's Guide, Texas Instruments Incorporated, Feb. 1998
17. TMS320C6x Peripheral Support Library, Texas Instruments Incorporated, Mar. 1998