基于DaVinci处理器的视频预处理DSP软件设计
|
摘要
随着人们生活水平的不断提高和以DSP处理器为核心的嵌入式系统的不断发展,人们对基于DSP处理器的视频产品的需求越来越大,而无论以DVR为代表的安防产品还是以高清晰度数字电视为代表的消费类电子产品,在视频采集获取过程中都需要对原始视频图像进行预处理,以得到理想的可以用于压缩和各种图像处理的高品质视频。基于上述原因,本文研究并实现了基于DaVinci处理器的视频预处理DSP软件。
第一章分析了选题意义,介绍了相关技术的发展现状,论述了数字视频和预处理技术的基础,DSP处理器的发展历史和现状,并阐述了课题的主要研究内容和本文章节结构。
第二章介绍了TI DSP处理器的结构和功能特点,以及集成开发环境CCS的功能。
第三章论述了视频采集和预处理技术,从色彩插值、色彩校正、自动聚焦、自动曝光、自动白平衡、伽马校正和图像增强几方面对视频预处理技术进行了详细的阐述。
第四章针对前沿的图像降噪和图像锐化算法,分别研究了改进型中值滤波和图像锐化新方法,重点阐述了算法流程和结构,并进行了仿真实验,与其它中值滤波和图像锐化算法进行比较。
第五章着重研究了基于DSP处理器的视频预处理软件,分为平台部分和算法部分。平台部分中对系统初始化模块、视频采集模块、视频预处理模块和通讯模块进行了详细研究；算法部分中研究了针对混合噪声的“两阶段”改进型滤波算法,并进行了算法仿真。
第六章研究了视频预处理的DSP软件优化,分为内存数据拷贝优化、帧间运动搜索优化和DMA搬运机制优化三部分。
第七章对整个研究进行了总结,并对视频预处理系统的发展和应用前景进行了展望。
With continuous improvements of people's living standard and developments of embedded systems based on DSP processors, demands for video products based on DSP processors are growing. This paper discusses a video pre-processing software design based on DaVinci DSP to address these demands.
Chapter 1 describes the progress of the video pre-processing system, analyzes the foundation of the digital video technology, and presents the evolution of DSP processors then puts forward the task of the subject and structure of this research paper.
Chapter 2 describes the structure and functional characteristics of TI TMS320DM6437 DSP processor as well as the function of integrated development environment of CCS.
Chapter 3 discusses the video capture and pre-processing technology in detail.
Chapter 4 presents the design of an improved median filtering, and an image sharpening method.
Chapter 5 divides the design into platform design and algorithm design, and focuses on video pre-processing software design based on DSP. Some related modules in platform design are discussed in detail, and an improved "two-stage" algorithm for filtering mixed noise is also discussed.
Chapter 6 describes the video pre-processing DSP software optimization in detail.
Chapter 7 summarizes the overall design and prospects for video pre-processing developments and applications.
第一章分析了选题意义,介绍了相关技术的发展现状,论述了数字视频和预处理技术的基础,DSP处理器的发展历史和现状,并阐述了课题的主要研究内容和本文章节结构。
第二章介绍了TI DSP处理器的结构和功能特点,以及集成开发环境CCS的功能。
第三章论述了视频采集和预处理技术,从色彩插值、色彩校正、自动聚焦、自动曝光、自动白平衡、伽马校正和图像增强几方面对视频预处理技术进行了详细的阐述。
第四章针对前沿的图像降噪和图像锐化算法,分别研究了改进型中值滤波和图像锐化新方法,重点阐述了算法流程和结构,并进行了仿真实验,与其它中值滤波和图像锐化算法进行比较。
第五章着重研究了基于DSP处理器的视频预处理软件,分为平台部分和算法部分。平台部分中对系统初始化模块、视频采集模块、视频预处理模块和通讯模块进行了详细研究；算法部分中研究了针对混合噪声的“两阶段”改进型滤波算法,并进行了算法仿真。
第六章研究了视频预处理的DSP软件优化,分为内存数据拷贝优化、帧间运动搜索优化和DMA搬运机制优化三部分。
第七章对整个研究进行了总结,并对视频预处理系统的发展和应用前景进行了展望。
With continuous improvements of people's living standard and developments of embedded systems based on DSP processors, demands for video products based on DSP processors are growing. This paper discusses a video pre-processing software design based on DaVinci DSP to address these demands.
Chapter 1 describes the progress of the video pre-processing system, analyzes the foundation of the digital video technology, and presents the evolution of DSP processors then puts forward the task of the subject and structure of this research paper.
Chapter 2 describes the structure and functional characteristics of TI TMS320DM6437 DSP processor as well as the function of integrated development environment of CCS.
Chapter 3 discusses the video capture and pre-processing technology in detail.
Chapter 4 presents the design of an improved median filtering, and an image sharpening method.
Chapter 5 divides the design into platform design and algorithm design, and focuses on video pre-processing software design based on DSP. Some related modules in platform design are discussed in detail, and an improved "two-stage" algorithm for filtering mixed noise is also discussed.
Chapter 6 describes the video pre-processing DSP software optimization in detail.
Chapter 7 summarizes the overall design and prospects for video pre-processing developments and applications.
引文
[1]毕厚杰.新一代视频压缩编码标准——H.264/AVC[M].北京：人民邮电出版社,2005.
[2]陈航锋,基于DaVinci处理器的H.264解码器软件设计与优化实现[硕士学位论文],杭州,浙江大学,2007：1-2.
[3]Keith Jack, Video Demystified[M], LLH Technology Publishing, Eagle Rock, VA 24085,2001:21-23.
[4]Analog Device Inc, ADSP-BF561 Blackfin(?) Processor Hardware Reference[S], Analog Device Inc,2005:512-514.
[5]张雄伟,曹铁勇.DSP芯片的原理与开发应用(第2版)[M],北京,电子工业出版社,2000年09月：2-3.
[6]朱希志,罗良进.纵谈DSP的发展和应用[J],电脑知识于技术,2004(03)：65.
[7]郑小龙.达芬奇(DaVinci)技术融合之势[J],电子产品世界,2005(12)：151.
[8]郑小龙.达芬奇之旅-TI数字视频巡回研讨会的技术核心[J],电子产品世界,2007(01).
[9]德州仪器全新达芬奇技术开发平台加速数字视频产品开发[J],电子与电脑,2007(04)：72.
[10]代君利.通信、数字家庭应用市场驱动DSP技术发展[J].中国电子商情.2007(03).
[11]Texas Instruments, TMS320DM6437 Digital Media Processor[J],2007(03):5.
[12]Texas Instruments, TMS320C64x C64x+DSP CPU and Instruction Set Reference Guide[J],2006 (05):32.
[13]Texas Instruments,TMS320DM643x DMP DSP Subsystem Reference Guide[J], 2007(04):19.
[14]Texas Instruments, TMS320DM643x DMP Enhanced DMA (EDMA) Controller User's Guide[J],2007(01):20.
[15]李方慧等,TMS320C6000系列DSPs原理和应用(第二版)[M],北京,电子工业出版社,2003：86.
[16]Texas Instruments, TMS320C6000 Assembly Language Tools User's Guide[J], 2006(10):18.
[17]田黎育,何佩琨,朱梦宇,TMS320C6000系列DSP编程工具与指南[M],北京,清华大学出版社,2006年9月
[18]Yao Wang等著,侯正信,杨喜,王文全等译.视频处理与通信[M].北京：电子工业出版社,2003.
[19]X.Zhang, N.Fukuda等.A signal Processing System on Chip for Digital Cameras[C]. Industrial Electronics Society,26th Annual Conference of the IEEE 2000,1243-1248.
[20]孙景鳌,蔡安妮.彩色电视基础[M].北京：人民邮电出版社,1996.
[21]谢攀,张利,唐宗明等.一种基于尺度变化的DCT自动聚焦算法[J].清华大学学报(自然科学版),2003,1(43)：55-58.
[22]Kazushige Ooi, Keiji Izumi, M itsuyuki Nozaki等.An advanced autofocus system for video camera using quasi condition reasoning[J].IEEE transation on Consumer Elec.,1990,36:526-529.
[23]Gonzalez,Woods著,阮秋琦,阮宇智译.数字图像处理[M].北京：电子工业出版社,2008：132-137.
[24]霍俊彦,视频预处理关键技术研究[硕士学位论文],西安,西安电子科技大学,2006：10.
[25]J.W. Tukey, "Nonlinear methods for smoothing data,"in Conf. Rec., (Eascon), 1974.
[26]Li Jim S.,Randhawa Sharmil. Color filter array demosaicking using high-order interpolation techniques with a weighted median filter for sharp color edge preservation [J]. IEEE Transactions on Image Processing,2009,18(9):1946-1957.
[27]Lin, Tzu-Chao.A new adaptive center weighted median filter for suppressing impulsive noise in images[J]. Information Sciences,2007,177(4):1073-1087.
[28]Petrovic, Crnojevic. Universal impulse noise filter based on genetic programming [J]. IEEE Transactions on Image Processing,2008,17(7):1109-1120.
[29]Ng Pei-Eng,Ma Kai-Kuang. A switching median filter with boundary discriminative noise detection for extremely corrupted images [J]. IEEE Transactions on Image Processing,2006,15(6):1506-1516.
[30]Chang, Chin-Chen Hsiao, Ju-Yuan,et al. An adaptive median filter for image denoising[C]. Proceeding of 2nd International Symposium on Intelligent Information Technology Application, Shanghai,2008:346-350.
[31]Wang Ming-chang,Xing Li-xin,Yang Yi-heng, etal.Directional Weighted Two-Dimensional Multi-step Median Filtering in Image Processing [J]. Journal of Jinlin University(Informantion Science Edition),2006,24(1):18-21.
[32]Go Tanaka,Noriaki Suetake and Eiji Uchino. Random-valued impulse noise detector based on minimum spanning tree and its application to switching median filter[C].Proceeding of IEEE International Conference on Systems, Man and Cybernetics,Singapore,2008:1008-1013.
[33]Huang Zan. A median filter based on judging impulse noise by statistic and adaptive threshold[C]. Proceeding of 1st International Congress on Image and Signal Processing,Sanya,China,2008:207-210.
[34]Nie Yao,Barner Kenneth E.. The fuzzy transformation and its applications in image processing [J]. IEEE Transactions on Image Processing.2006,15(4): 910-927.
[35]Wong Kok Wai,Tikk Domonkos.Fuzzy rule interpolation for multidimensional input spaces with applications:A case study [J]. IEEE Transactions on Fuzzy Systems,2005,13(6):809-819.
[36]Kang, Chung-Chia.Fuzzy reasoning-based directional median filter design[J]. Signal Processing,2009,89(3):344-351.
[37]Lianghai Jin, Dehua Li, Zijian Lu. Improved adaptive spatial distance-weighted median filter[J]. Optical Engineering,2007,46(3):037003.
[38]Aysal Tuncer Can, Barner Kenneth E.Quadratic weighted median filters for edge enhancement of noisy images [J]. IEEE Transactions on Image Processing,2006,15(11):3294-3310.
[39]杜召彬 罗晓晖.自适应模板的图像锐化算法[J].工程地质计算机应用,2007.3：21-24.
[40]傅德胜,寿益禾.图形图像处理学[M].东南大学出版社,2001.193-199,204-224.
[41]Rafael C.Gonzalez,Richard E Woods.Digital Image Processing[M], Second Edition. Publishing House of Electronics Industry,98-108.
[42]Kenneth R.Castleman Digital Image Processing[M].Second Edition.Publishing House of Electronics Industry,2005:389-391.
[43]杨淑莹.Vc++图像处理程序设计(第二版)[M].清华大学出版社,北京交通大学出版社,2005：92-97,108-141.
[44]Texas Instruments, TMS320DM643x DMP Video Processing Front End user's Guide[S],2007(4):9.
[45]王国烽,基于DaVinci DSP的DVR编码软件设计[硕士学位论文],杭州,浙江大学,2008：49-51.
[46]Texas Instruments, TMS320DM643x DMP Host Port Interface (HPI) User's Guide [S],2007(2):8.
[47]R. Garnett, T. Huegerich, C. Chui, etal. A universal noise removal algorithm with an impulse detector[J].IEEE Transaction on Image Processing,2005,14 (11):1747-1754.
[48]Z. Ma, H. R. Wu, and B. Qiu. An robust structure adaptive hybrid vector filter for color image restoration [J]. IEEE Transaction on Image Processing,2005, 14(12):1900-2001.
[49]Jie Xiang Yang and Hong Ren Wu. Mixed Gaussian and uniform impulse noise analysis using robust estimation for digital images [C]. Proceeding of 16th International Conference on Digital Signal Processing, Santorini, Greece,2009: 1-5.
[50]Liwei Guo, Oscar C. Au and Mengyao Ma,et al. Fast multi-hypothesis motion compensated filter for video denoising[C]. Proceeding of IEEE Workshop on Signal Processing Systems, SiPS:Design and Implementation, Shanghai, China,2007:283-288.
[51]Ismail, Bayoumi, Magdy A. Enhanced efficient diamond search algorithm for fast block motion estimation [C]. Proceeding of IEEE International Symposium on Circuits and Systems, Taipei, Taiwan,2009:3198-3201.
[52]雷玉堂,标准式DVR及未来的智能式DVR[J],中国公共安全,2007(7)：82-83.
[2]陈航锋,基于DaVinci处理器的H.264解码器软件设计与优化实现[硕士学位论文],杭州,浙江大学,2007：1-2.
[3]Keith Jack, Video Demystified[M], LLH Technology Publishing, Eagle Rock, VA 24085,2001:21-23.
[4]Analog Device Inc, ADSP-BF561 Blackfin(?) Processor Hardware Reference[S], Analog Device Inc,2005:512-514.
[5]张雄伟,曹铁勇.DSP芯片的原理与开发应用(第2版)[M],北京,电子工业出版社,2000年09月：2-3.
[6]朱希志,罗良进.纵谈DSP的发展和应用[J],电脑知识于技术,2004(03)：65.
[7]郑小龙.达芬奇(DaVinci)技术融合之势[J],电子产品世界,2005(12)：151.
[8]郑小龙.达芬奇之旅-TI数字视频巡回研讨会的技术核心[J],电子产品世界,2007(01).
[9]德州仪器全新达芬奇技术开发平台加速数字视频产品开发[J],电子与电脑,2007(04)：72.
[10]代君利.通信、数字家庭应用市场驱动DSP技术发展[J].中国电子商情.2007(03).
[11]Texas Instruments, TMS320DM6437 Digital Media Processor[J],2007(03):5.
[12]Texas Instruments, TMS320C64x C64x+DSP CPU and Instruction Set Reference Guide[J],2006 (05):32.
[13]Texas Instruments,TMS320DM643x DMP DSP Subsystem Reference Guide[J], 2007(04):19.
[14]Texas Instruments, TMS320DM643x DMP Enhanced DMA (EDMA) Controller User's Guide[J],2007(01):20.
[15]李方慧等,TMS320C6000系列DSPs原理和应用(第二版)[M],北京,电子工业出版社,2003：86.
[16]Texas Instruments, TMS320C6000 Assembly Language Tools User's Guide[J], 2006(10):18.
[17]田黎育,何佩琨,朱梦宇,TMS320C6000系列DSP编程工具与指南[M],北京,清华大学出版社,2006年9月
[18]Yao Wang等著,侯正信,杨喜,王文全等译.视频处理与通信[M].北京：电子工业出版社,2003.
[19]X.Zhang, N.Fukuda等.A signal Processing System on Chip for Digital Cameras[C]. Industrial Electronics Society,26th Annual Conference of the IEEE 2000,1243-1248.
[20]孙景鳌,蔡安妮.彩色电视基础[M].北京：人民邮电出版社,1996.
[21]谢攀,张利,唐宗明等.一种基于尺度变化的DCT自动聚焦算法[J].清华大学学报(自然科学版),2003,1(43)：55-58.
[22]Kazushige Ooi, Keiji Izumi, M itsuyuki Nozaki等.An advanced autofocus system for video camera using quasi condition reasoning[J].IEEE transation on Consumer Elec.,1990,36:526-529.
[23]Gonzalez,Woods著,阮秋琦,阮宇智译.数字图像处理[M].北京：电子工业出版社,2008：132-137.
[24]霍俊彦,视频预处理关键技术研究[硕士学位论文],西安,西安电子科技大学,2006：10.
[25]J.W. Tukey, "Nonlinear methods for smoothing data,"in Conf. Rec., (Eascon), 1974.
[26]Li Jim S.,Randhawa Sharmil. Color filter array demosaicking using high-order interpolation techniques with a weighted median filter for sharp color edge preservation [J]. IEEE Transactions on Image Processing,2009,18(9):1946-1957.
[27]Lin, Tzu-Chao.A new adaptive center weighted median filter for suppressing impulsive noise in images[J]. Information Sciences,2007,177(4):1073-1087.
[28]Petrovic, Crnojevic. Universal impulse noise filter based on genetic programming [J]. IEEE Transactions on Image Processing,2008,17(7):1109-1120.
[29]Ng Pei-Eng,Ma Kai-Kuang. A switching median filter with boundary discriminative noise detection for extremely corrupted images [J]. IEEE Transactions on Image Processing,2006,15(6):1506-1516.
[30]Chang, Chin-Chen Hsiao, Ju-Yuan,et al. An adaptive median filter for image denoising[C]. Proceeding of 2nd International Symposium on Intelligent Information Technology Application, Shanghai,2008:346-350.
[31]Wang Ming-chang,Xing Li-xin,Yang Yi-heng, etal.Directional Weighted Two-Dimensional Multi-step Median Filtering in Image Processing [J]. Journal of Jinlin University(Informantion Science Edition),2006,24(1):18-21.
[32]Go Tanaka,Noriaki Suetake and Eiji Uchino. Random-valued impulse noise detector based on minimum spanning tree and its application to switching median filter[C].Proceeding of IEEE International Conference on Systems, Man and Cybernetics,Singapore,2008:1008-1013.
[33]Huang Zan. A median filter based on judging impulse noise by statistic and adaptive threshold[C]. Proceeding of 1st International Congress on Image and Signal Processing,Sanya,China,2008:207-210.
[34]Nie Yao,Barner Kenneth E.. The fuzzy transformation and its applications in image processing [J]. IEEE Transactions on Image Processing.2006,15(4): 910-927.
[35]Wong Kok Wai,Tikk Domonkos.Fuzzy rule interpolation for multidimensional input spaces with applications:A case study [J]. IEEE Transactions on Fuzzy Systems,2005,13(6):809-819.
[36]Kang, Chung-Chia.Fuzzy reasoning-based directional median filter design[J]. Signal Processing,2009,89(3):344-351.
[37]Lianghai Jin, Dehua Li, Zijian Lu. Improved adaptive spatial distance-weighted median filter[J]. Optical Engineering,2007,46(3):037003.
[38]Aysal Tuncer Can, Barner Kenneth E.Quadratic weighted median filters for edge enhancement of noisy images [J]. IEEE Transactions on Image Processing,2006,15(11):3294-3310.
[39]杜召彬 罗晓晖.自适应模板的图像锐化算法[J].工程地质计算机应用,2007.3：21-24.
[40]傅德胜,寿益禾.图形图像处理学[M].东南大学出版社,2001.193-199,204-224.
[41]Rafael C.Gonzalez,Richard E Woods.Digital Image Processing[M], Second Edition. Publishing House of Electronics Industry,98-108.
[42]Kenneth R.Castleman Digital Image Processing[M].Second Edition.Publishing House of Electronics Industry,2005:389-391.
[43]杨淑莹.Vc++图像处理程序设计(第二版)[M].清华大学出版社,北京交通大学出版社,2005：92-97,108-141.
[44]Texas Instruments, TMS320DM643x DMP Video Processing Front End user's Guide[S],2007(4):9.
[45]王国烽,基于DaVinci DSP的DVR编码软件设计[硕士学位论文],杭州,浙江大学,2008：49-51.
[46]Texas Instruments, TMS320DM643x DMP Host Port Interface (HPI) User's Guide [S],2007(2):8.
[47]R. Garnett, T. Huegerich, C. Chui, etal. A universal noise removal algorithm with an impulse detector[J].IEEE Transaction on Image Processing,2005,14 (11):1747-1754.
[48]Z. Ma, H. R. Wu, and B. Qiu. An robust structure adaptive hybrid vector filter for color image restoration [J]. IEEE Transaction on Image Processing,2005, 14(12):1900-2001.
[49]Jie Xiang Yang and Hong Ren Wu. Mixed Gaussian and uniform impulse noise analysis using robust estimation for digital images [C]. Proceeding of 16th International Conference on Digital Signal Processing, Santorini, Greece,2009: 1-5.
[50]Liwei Guo, Oscar C. Au and Mengyao Ma,et al. Fast multi-hypothesis motion compensated filter for video denoising[C]. Proceeding of IEEE Workshop on Signal Processing Systems, SiPS:Design and Implementation, Shanghai, China,2007:283-288.
[51]Ismail, Bayoumi, Magdy A. Enhanced efficient diamond search algorithm for fast block motion estimation [C]. Proceeding of IEEE International Symposium on Circuits and Systems, Taipei, Taiwan,2009:3198-3201.
[52]雷玉堂,标准式DVR及未来的智能式DVR[J],中国公共安全,2007(7)：82-83.