一种成像跟踪系统的图像处理模块的设计及算法研究
|
摘要
成像跟踪系统是利用景物的图像特征,对空中或地面的机动目标进行实时跟踪的系统。现代微电子技术、计算机技术的飞速发展和成像跟踪算法的优化设计,为成像跟踪系统的实现,奠定了良好的软硬件基础,使成像跟踪系统的实时性、智能性大大提高。本论文提出了成像跟踪系统中的核心模块—图像处理模块的功能模型、给出了具体的设计方案,并推导和实现了一种高效的成像跟踪算法。根据成像跟踪系统的要求,文中提出图像处理模块由波门电路和实时图像处理器系统两个功能模块组成,波门电路主要实现图像数据的采集和系统控制信号的产生,实时图像处理器系统主要对波门电路采集的图像数据进行算法处理并进行波门预置。针对图像处理模块的功能特点,论文给出了具体的设计方案,方案主要由两部分组成,即以定点数字信号处理器芯片为核心的实时图像处理器系统的设计和以复杂的可编程逻辑器件(CPLD)为核心的波门电路的设计,系统的整体硬件设计采用了数字信号处理器系统与专用集成电路(ASIC)相结合的结构,具备高速的信息处理能力。在成像跟踪算法方面,考虑了简单场景系统下弱机动远景目标成像跟踪的特点,设计了精度高、在结构形式上易于数字信号处理器芯片(DSP)处理的二值图像形心预测跟踪算法,其中二值门限和形心的提取及形心的预测是算法的关键,算法通过验证并在TI公司的定点数字信号处理器芯片上实现。
Image tracking system is a real-time tracking system, which takes advantage of characters of image to track moving targets in the air or on the ground. It is the quick development of micro-electronic technology、computer technology and the optimum design of image tracking algorithm that build the software and hardware foundation of the implement of image tracking system. They also greatly improve the real-time and intellective attributes of the system. The paper presents the functional framework and the concrete design method of an image processing module which is the key part of an image tracking system. An efficient algorithm of image tracking is deduced and implemented in real-time DSP. On the requirement of image tracking system, this paper issues that image processing module is composed by two functional modules, which are wavegate circuit and real-time image processing system. The main function of wavegate is the collection of image data and the production of control signal of the system; that of real-time image processing system is to process the image data acquired by wavegate circuit and set the coordinate of wavegate in advance. Focused on the functional characteristic of image processing module, this paper present a concrete method of design. There are two parts of this project, they are the design of real-time image processing system with the core of the Fixed-Point DSP and the design of wavegate with the core of CPLD. The whole hardware design adopt the architecture which is combined of digital signal processing system and ASIC. It has high-speed signal processing ability. On the image tracking algorithm, this paper design a tracking algorithm of centroid to binary image, which is high precision and with the structure easily be processed by DSP. The Binary threshold and the abstract and forecast of centroid is the key to this algorithm. The algorithms have been validated and implemented in the TI Fixed-Point DSP.
Image tracking system is a real-time tracking system, which takes advantage of characters of image to track moving targets in the air or on the ground. It is the quick development of micro-electronic technology、computer technology and the optimum design of image tracking algorithm that build the software and hardware foundation of the implement of image tracking system. They also greatly improve the real-time and intellective attributes of the system. The paper presents the functional framework and the concrete design method of an image processing module which is the key part of an image tracking system. An efficient algorithm of image tracking is deduced and implemented in real-time DSP. On the requirement of image tracking system, this paper issues that image processing module is composed by two functional modules, which are wavegate circuit and real-time image processing system. The main function of wavegate is the collection of image data and the production of control signal of the system; that of real-time image processing system is to process the image data acquired by wavegate circuit and set the coordinate of wavegate in advance. Focused on the functional characteristic of image processing module, this paper present a concrete method of design. There are two parts of this project, they are the design of real-time image processing system with the core of the Fixed-Point DSP and the design of wavegate with the core of CPLD. The whole hardware design adopt the architecture which is combined of digital signal processing system and ASIC. It has high-speed signal processing ability. On the image tracking algorithm, this paper design a tracking algorithm of centroid to binary image, which is high precision and with the structure easily be processed by DSP. The Binary threshold and the abstract and forecast of centroid is the key to this algorithm. The algorithms have been validated and implemented in the TI Fixed-Point DSP.
引文
1.杨宜禾,周维真.成像跟踪系统导论.西安电子科技大学出版社,1991
2.杨宜禾,岳敏.红外系统.国防工业出版社,1985
3.罗第才.导引头智能跟踪方案设计.西安电子科技大学硕士论文,1989
4.黄如,李正明,徐光辉.一种成像跟踪系统图像处理模块的设计与实现.第7届全国青年通信学术会议论文集:P76-80,电子工业出版社,2001
5.TMS320C54x DSP Reference Set Volume 1:CPU and Peripherals,1997
6.徐志军,徐光辉.CPLD/FPGA的开发与应用.电子工业出版社,2002
7.Nader Kazor.Targer Tracking Based Scene Analysis,university of Maryland,1984
8.Thomas Meier,kng N.Ngan.Automatic Segmentation of Moving Objects for Video Object Plane Generation IEEE TRANSACTIONS ON CIRCUITS AND SYSTEM FOR VIDEO TECHNOLOGY,VOL.8,NO.5,1998
9.Liang Hwa Chen et al.A video tracking system with adaptive predictor,Pattern Recongnition Vol.25,1991
10.MASAHIKO YACHIDA,MINORU ASADAAND SABURO TSUJI.Automatic Analysis of Moving Image IEEE TRANSACTIONB ON PATTERN ANALYSIS AND MACHINGINTELLIGENCE,VOL.PAMI-1,NO.1.JAN 1981
11.W.K.Pratt.Digital Image Processing,John Wiley&SonsInc,1978
12.R.J.Schalkoff,etal.Motion Estimation,Spriger-Verlag,1981
13.D.H.Ballard and C.M.Brown.Computer Vision,Prentice,1982
14.Tracking Multiple Target With Correlated Measurements and Maneuver IEEE TRANSCATION ON AEROSPACE AND ELECTRONIC SYSTEM VOL.24.NO.3MAY 1988
15.CHAW-BING CHANG.Application of State Estimation to Target Tracking,IEEE TRANSACTION ON AUTOMATIC CONTROL,VOL.AC-29,NO.2,FEB 1984
16.E.S.Movey et al.A perturbation method for obtaining control signals in an image tracking system,Proc.Joint Automation Control,1979
17.国澄明,李铁盘.用DSP实现图像的直接捕获与显示.电子学报.1995(3):119-120
18.余祖俊.基于FPGA的多路视频图像高速同步采集系统.北方交通大学学报.1998(4):66-69
19.彭鹏,王宏远.用FPGA实现图像采集的控制.华中理工大学学报.1998(11):14-15
20.高永英,张利,吴国威.一种基于灰度期望值的图像二值化算法.中国图像图形学报.1999(6):524-528
21.江捧岳.图像跟踪算法分析.火力与指挥控制.1998(2):48-52
22.付忠良.图像阀值选取方法的构造.中国图像图形学报.2000(6):466-469
23.王春平,朱元昌.基于DSP的电视跟踪系统算法设计.军械工程学院学报.1999(2):33-39
24.张杰,王加存.两种跟踪机动目标的方法研究.火力与指挥控制.1998(2):69-73
25.林广艳,曹晓英.利用双口RAM实现的高速数据采集系统.天津纺织工学院学报.1999(5):78-82
26.蔡靖,王金刚,丁润涛.高速数据采集输出接口控制.天津通信技术.1999(2):4-7
27.徐向辉,陶然,王越.视频图像采集系统的设计与实现.系统工程与电子技术.2001(10):13-16
28.高西全,丁玉美.数字信号处理.西安电子科技大学出版社,2001
29.潘松,王国栋.VHDL实用教程.2000
30.戴明帧,周建江.TMS320C54X DSP结构、原理及应用.北京航空航天大学出版社,2001
31.程军.Intel 80C196单片机应用实践与C语言开发.北京航空航天大学出版社,2000
32.张雄伟,曹铁勇.DSP芯片的原理与开发应用.电子工业出版社,2000
33.高光天.模数转换器应用技术.科学出版社,2001
34.孙仲康.数字图像处理及应用,国防工业出版社,1985
35.Sahoo P K et al.A survey of thresholding technique,computer Vision,1988
36.王念旭.DSP基础与应用系统设计,北京航空航天大学出版社,2001
37.李广军.可编程ASIC设计及应用,电子科技大学出版社,2000
38.李刚.数字信号微处理器的原理及其开发应用,天津大学出版社,2000
39.申敏.DSP原理及其在移动通信中的应用,人民邮电出版社,2001
40.伯晓晨、李涛.Matlab工具箱应用指南—信息工程篇,电子工业出版社,2000
41.陈亚勇.Matlab信号处理详解,人民邮电出版社,2001
2.杨宜禾,岳敏.红外系统.国防工业出版社,1985
3.罗第才.导引头智能跟踪方案设计.西安电子科技大学硕士论文,1989
4.黄如,李正明,徐光辉.一种成像跟踪系统图像处理模块的设计与实现.第7届全国青年通信学术会议论文集:P76-80,电子工业出版社,2001
5.TMS320C54x DSP Reference Set Volume 1:CPU and Peripherals,1997
6.徐志军,徐光辉.CPLD/FPGA的开发与应用.电子工业出版社,2002
7.Nader Kazor.Targer Tracking Based Scene Analysis,university of Maryland,1984
8.Thomas Meier,kng N.Ngan.Automatic Segmentation of Moving Objects for Video Object Plane Generation IEEE TRANSACTIONS ON CIRCUITS AND SYSTEM FOR VIDEO TECHNOLOGY,VOL.8,NO.5,1998
9.Liang Hwa Chen et al.A video tracking system with adaptive predictor,Pattern Recongnition Vol.25,1991
10.MASAHIKO YACHIDA,MINORU ASADAAND SABURO TSUJI.Automatic Analysis of Moving Image IEEE TRANSACTIONB ON PATTERN ANALYSIS AND MACHINGINTELLIGENCE,VOL.PAMI-1,NO.1.JAN 1981
11.W.K.Pratt.Digital Image Processing,John Wiley&SonsInc,1978
12.R.J.Schalkoff,etal.Motion Estimation,Spriger-Verlag,1981
13.D.H.Ballard and C.M.Brown.Computer Vision,Prentice,1982
14.Tracking Multiple Target With Correlated Measurements and Maneuver IEEE TRANSCATION ON AEROSPACE AND ELECTRONIC SYSTEM VOL.24.NO.3MAY 1988
15.CHAW-BING CHANG.Application of State Estimation to Target Tracking,IEEE TRANSACTION ON AUTOMATIC CONTROL,VOL.AC-29,NO.2,FEB 1984
16.E.S.Movey et al.A perturbation method for obtaining control signals in an image tracking system,Proc.Joint Automation Control,1979
17.国澄明,李铁盘.用DSP实现图像的直接捕获与显示.电子学报.1995(3):119-120
18.余祖俊.基于FPGA的多路视频图像高速同步采集系统.北方交通大学学报.1998(4):66-69
19.彭鹏,王宏远.用FPGA实现图像采集的控制.华中理工大学学报.1998(11):14-15
20.高永英,张利,吴国威.一种基于灰度期望值的图像二值化算法.中国图像图形学报.1999(6):524-528
21.江捧岳.图像跟踪算法分析.火力与指挥控制.1998(2):48-52
22.付忠良.图像阀值选取方法的构造.中国图像图形学报.2000(6):466-469
23.王春平,朱元昌.基于DSP的电视跟踪系统算法设计.军械工程学院学报.1999(2):33-39
24.张杰,王加存.两种跟踪机动目标的方法研究.火力与指挥控制.1998(2):69-73
25.林广艳,曹晓英.利用双口RAM实现的高速数据采集系统.天津纺织工学院学报.1999(5):78-82
26.蔡靖,王金刚,丁润涛.高速数据采集输出接口控制.天津通信技术.1999(2):4-7
27.徐向辉,陶然,王越.视频图像采集系统的设计与实现.系统工程与电子技术.2001(10):13-16
28.高西全,丁玉美.数字信号处理.西安电子科技大学出版社,2001
29.潘松,王国栋.VHDL实用教程.2000
30.戴明帧,周建江.TMS320C54X DSP结构、原理及应用.北京航空航天大学出版社,2001
31.程军.Intel 80C196单片机应用实践与C语言开发.北京航空航天大学出版社,2000
32.张雄伟,曹铁勇.DSP芯片的原理与开发应用.电子工业出版社,2000
33.高光天.模数转换器应用技术.科学出版社,2001
34.孙仲康.数字图像处理及应用,国防工业出版社,1985
35.Sahoo P K et al.A survey of thresholding technique,computer Vision,1988
36.王念旭.DSP基础与应用系统设计,北京航空航天大学出版社,2001
37.李广军.可编程ASIC设计及应用,电子科技大学出版社,2000
38.李刚.数字信号微处理器的原理及其开发应用,天津大学出版社,2000
39.申敏.DSP原理及其在移动通信中的应用,人民邮电出版社,2001
40.伯晓晨、李涛.Matlab工具箱应用指南—信息工程篇,电子工业出版社,2000
41.陈亚勇.Matlab信号处理详解,人民邮电出版社,2001