CCTV目标定位技术在VTS领域中的应用
|
摘要
VTS (Vessel Traffic Services)主要负责监管船舶的航行安全以及水域环境安全。以往的VTS系统中,主要依靠雷达获取目标的位置及动态信息。目前,我国各地的主要航道及重点水域已经建成CCTV (Closed Circuit Television)监控网,CCTV视频监控子系统已经成为VTS系统获取目标信息的重要途径,发挥越来越大的作用。
与雷达子系统相比,现有的CCTV监控子系统仅完成现场的实时监控与记录,无法从大量的平面图像中选取有用信息来对交通形势进行有效的判断与评估。技术相对成熟的雷达子系统由于受环境和水域条件影响,威力和性能常常难以发挥,导致雷达的投资效益比极低。因此,迫切需要一种改进的目标定位技术,在特定环境下弥补雷达不足。
本文的研究来源于宜昌海事局的“宜昌船舶交通管理系统工程枝城大桥站项目”,主要负责其中的CCTV监控技术改进方法的研究。文中提出一种基于CCTV子系统的双目立体定位理论,定位方法的确定受机器人视觉启发,以基于角度的测量方法为依据。首先,建立定位模型；其次,确定摄像机外参数矩阵；最终,将定位结果表示为统一的世界坐标。根据误差传递函数对定位误差进行了理论分析,分析结果显示在较好的图像处理技术支持下,CCTV目标定位技术对近距离目标的定位效果较好。
为了能够使CCTV目标定位技术不断的向实际应用环境发展,本文结合VisualC++开发平台,设计了一个基于简单背景的简单目标定位实验系统。实验系统方案设计中,以小型水域为背景分析如何选择所需设备,仅在实现过程中将其简化。
系统构成采用模块化结构,结构层次清晰。主要分为视频采集与处理模块(包括双目系统的目标识别与立体匹配等环节)、数码云台智能控制模块、实时定位模块、终端显示模块(显示目标图像信息与定位结果)。各模块之间相对独立工作、数据共享,实现联动控制。
Maritime traffic as an important transportation is supervised by the branch of maritime management with Vessel Traffic Services (VTS). At present, marine bureaus' VTS system in many provinces has been configured closed circuit television subsystem. As an important supervising manner, CCTV system can prolong the time of supervision; it also can widen the range of supervision.
Compared with radar subsystem, the existing CCTV video surveillance subsystem could only achieve the real-time supervising and recording of the locale. It couldn't select useful information from a lot of plane figures to judge and estimate the traffic situation effectively. Radar subsystem as a relatively mature technology, its power and performance is hardly to be exerted, due to the effect of environment and water conditions. So, we needs a kind of modified target location technology urgently, in order to compensate the weakness of radar in some specifically conditions.
This thesis, coming from "the branch project of Zhicheng bridge station of Yichang vessel traffic management system" for Yichang maritime authorities, is mainly researching on the improvement of the CCTV supervising technology in the project. This article advances a theory of binocular stereoscopic location based on CCTV subsystem. This kind of location method is enlightened by robot vision, according to the measurement based on angle measuring. First, establishes the model of location; second, obtains the extrinsic parameter matrix of camera; finally, the locating result is shown as a uniform coordinate. On the basis of error transfer function, this article analyses the source of error and its effects. Analytic result shows that CCTV targets location technology should brings better effect of close-in target location, if it is supported by a better image processing technology.
In order to make CCTV target location technology develop to actual application environment, this article combines Visual C++ development platform, has designed an experimental system for simple target in simple condition. In the process of project design, this article analysis how to choose the equipment on the background of maritime traffic environment. It is simplified just when it carried out.
System structure using module as basic units; hierarchy is very clearly. System structure is mainly divided into Video capture and process module(includes target identifying and stereoscopic matching of binocular system, digital PTZ intelligent control module, real time location module, terminal display module(shows the image of target and the result of location). Every module works independently, shares data continuously, and achieves linkage controlling.
与雷达子系统相比,现有的CCTV监控子系统仅完成现场的实时监控与记录,无法从大量的平面图像中选取有用信息来对交通形势进行有效的判断与评估。技术相对成熟的雷达子系统由于受环境和水域条件影响,威力和性能常常难以发挥,导致雷达的投资效益比极低。因此,迫切需要一种改进的目标定位技术,在特定环境下弥补雷达不足。
本文的研究来源于宜昌海事局的“宜昌船舶交通管理系统工程枝城大桥站项目”,主要负责其中的CCTV监控技术改进方法的研究。文中提出一种基于CCTV子系统的双目立体定位理论,定位方法的确定受机器人视觉启发,以基于角度的测量方法为依据。首先,建立定位模型；其次,确定摄像机外参数矩阵；最终,将定位结果表示为统一的世界坐标。根据误差传递函数对定位误差进行了理论分析,分析结果显示在较好的图像处理技术支持下,CCTV目标定位技术对近距离目标的定位效果较好。
为了能够使CCTV目标定位技术不断的向实际应用环境发展,本文结合VisualC++开发平台,设计了一个基于简单背景的简单目标定位实验系统。实验系统方案设计中,以小型水域为背景分析如何选择所需设备,仅在实现过程中将其简化。
系统构成采用模块化结构,结构层次清晰。主要分为视频采集与处理模块(包括双目系统的目标识别与立体匹配等环节)、数码云台智能控制模块、实时定位模块、终端显示模块(显示目标图像信息与定位结果)。各模块之间相对独立工作、数据共享,实现联动控制。
Maritime traffic as an important transportation is supervised by the branch of maritime management with Vessel Traffic Services (VTS). At present, marine bureaus' VTS system in many provinces has been configured closed circuit television subsystem. As an important supervising manner, CCTV system can prolong the time of supervision; it also can widen the range of supervision.
Compared with radar subsystem, the existing CCTV video surveillance subsystem could only achieve the real-time supervising and recording of the locale. It couldn't select useful information from a lot of plane figures to judge and estimate the traffic situation effectively. Radar subsystem as a relatively mature technology, its power and performance is hardly to be exerted, due to the effect of environment and water conditions. So, we needs a kind of modified target location technology urgently, in order to compensate the weakness of radar in some specifically conditions.
This thesis, coming from "the branch project of Zhicheng bridge station of Yichang vessel traffic management system" for Yichang maritime authorities, is mainly researching on the improvement of the CCTV supervising technology in the project. This article advances a theory of binocular stereoscopic location based on CCTV subsystem. This kind of location method is enlightened by robot vision, according to the measurement based on angle measuring. First, establishes the model of location; second, obtains the extrinsic parameter matrix of camera; finally, the locating result is shown as a uniform coordinate. On the basis of error transfer function, this article analyses the source of error and its effects. Analytic result shows that CCTV targets location technology should brings better effect of close-in target location, if it is supported by a better image processing technology.
In order to make CCTV target location technology develop to actual application environment, this article combines Visual C++ development platform, has designed an experimental system for simple target in simple condition. In the process of project design, this article analysis how to choose the equipment on the background of maritime traffic environment. It is simplified just when it carried out.
System structure using module as basic units; hierarchy is very clearly. System structure is mainly divided into Video capture and process module(includes target identifying and stereoscopic matching of binocular system, digital PTZ intelligent control module, real time location module, terminal display module(shows the image of target and the result of location). Every module works independently, shares data continuously, and achieves linkage controlling.
引文
[1]李平."CCTV"水上交通安全视频监控系统在海事信息化管理中的运用与展望.科技信息.2009年第29期.
[2]杨立波,王旺.海事电视监控系统工程方案设计.水运工程.2003年10月总第357期.
[3]刘谦,梁兆棋,林丽.CCTV视频监控系统在广东内河水域的应用.中国水运.2009年6月.
[4]钱正锋.江苏海事局远程视频监控系统建设思考.中国水运.2008年1月.
[5]厦门海事局CCTV系统介绍.http://www.xmmsa.gov.cn/html/xmhsj/hsywzx/cctv/index.html
[6]丁鹭飞,耿富录,雷达原理.西安电子科技大学出版社.2002年6月.
[7]Fagueras O D, Toscani G. The calibration problem for stereo. IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Minmi Beach, Florida, June22-26,1986:15-20.
[8]Tsai R Y. A versatile camera calibration technique for high-accuracy 3D machine vision metrology using off-the-shelf cameras and lens. IEEE Transactions on Robotics and Automation,1987,3(4):323-344.
[9]Zhang Z. A flexible new technique for camera calibration. IEEE Transactions on Pattern Analysis and Machine Intelligence,2000,22(11):1330-1334.
[10]Kim J S, Kim H W, Kweon I S. A camera calibration method using concentric circles for vision applications. The 5th Asian Conference on Computer Vision,2002:515-520.
[11]Basu A. Active calibration:alternatibe strategy and analysis. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition,1993:495-500.
[12]Du F, Brady M. Self-calibration of the intrinsic parameters of camera for active vision system. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition,1993:477-482.
[13]Ma S D. A Self-calibration technique for active vision system. IEEE Transaction on Robotics and Automation,1996,12(1):114-120.
[14]Hartley R. Self-calibration of stationary cameras. International Journal of Computer Vision,1997,229(1):2-5.
[15]杨长江,孙凤梅,胡占义.基于二次曲面的纯旋转摄像机自标定.自动化学报,2001,27(3)：1130-1139.
[16]吴福朝,李华,胡占义.基于主动视觉系统的摄像机自标定研究.自动化学报,2001,27(6)：752-762.
[17]雷成,吴福朝,胡占义.一种新的基于主动视觉系统的摄像机自标定方法.计算机学报,2000,23(11)：1130-1139.
[18]扬长江,孙凤梅,胡占义.基于平面二次曲线的摄像机标定.计算机学报,2000,23(5)：541-547.
[19]李华,吴福朝,胡占义.一种新的线性摄像机自标定方法.计算机学报,2000,23(11)：1121-1129.
[20]傅丹等.一种摄像机自标定的线性方法.光电工程,2008,35(1)：71-75.
[21]Carvalho P C P, Szenberg F, Gattass M. Image-based modeling using a two-step camera calibration method. Proceedings of International Symposium on Computer Graphics, Image Processing and Vision,1998:388-395.
[22]Benallal M, Meunier J. Camera calibration with simple geometry. The Processing of 2003 International Conference on Image and Singal,2003.
[23]Yang A Y, Hong W, Ma Y. Structure and pose from single images of symmetric objects with applications to robot navigation. Proceeding of IEEE International Conference on Robotics & Automation, Taiper,2003:1013-1020.
[24]Wu Y H, Zhu H, Wu F C. Camera calibration from the quasi-affine invariance of two parallel circles. The 8th European Conference on Computer Vision(ECCV), 2004:190-202.
[25]贾云得.机器视觉.北京：科学出版社,2000.
[26]杨少荣,吴迪靖,段德山.机器视觉算法与应用.北京：清华大学出版社双语版.2008.
[27]马松德,张正友.计算机视觉.北京：科学出版社,1998.
[28]韩冰,林明星,丁凤华.机器视觉技术及其应用分析.农业装备与车辆工程.2008年第10期.
[29]王军,李星野.机器人视觉系统的组成及其工作原理.机器工程师.2007年第3期.
[30]张广军.视觉测量.北京：科学出版社.2008.
[31]徐德,谭民,李原.机器人视觉测量与控制==Visual measurement and control for robots.北京：国防工业出版社.2008.
[32]刘嵩鹤.基于双目立体视觉的安全车距测量技术研究：(硕士学位论文).长沙：武汉理工大学,2008.
[33]张义悦.基于机器视觉的三位检测定位理论研究：(硕士学位论文).昆明：昆明理工大学,2008.
[34]肖占春.基于双目立体视觉的机器人导航研究：(硕士学位论文).长春：东北师范大学,2008.
[35]崔岩.基于双目立体视觉的距离测量：(硕士学位论文).长春：长春理工大学,2006.
[36]常征.完善我国VTS服务的思考.中国海洋.2006第3期.
[37]才建胜.对国内船舶交通管制系统(VTS)现状的思考.天津航海.2006年第2期.
[38]何立居.基于VTS的海事信息发布系统.《航海技术》2006年第1期.
[39]赛迪.智能视频监控粉墨登场.中华建筑报2006年4月6日第007版电气智能化.
[40]张明友,汪学刚.雷达系统.北京：电子工业出版社.第二版：第1—28页.
[41]丁鹭飞,耿富录.雷达原理.西安：西安电子科技大学出版社.第三版：第1—24页.
[42]孙巍等.移动定位技术和移动定位系统.计算机系统应用.2003.10.
[43]夸拉特龙,萨莱里,李敏波.MATLAB科学计算.北京：清华大学出版社.2005.
[44]张志涌.精通MATLAB 6.5版.北京：北京航空航天大学出版社2003.3.
[45]杨磊等.闭路电视监控系统.北京：机械工业出版社,2003.5.
[46]http://www.net114.com/862817624/product/1284982.html
[47]http://www.net114.com/862908448/product/1236978.html
[48]http://www.net114.com/862904801/product/1176057.html
[49]宋坤等.Visual C++开发技术大全.北京：人民邮电出版社,2007.
[50]宋坤.Visual C++视频技术方案宝典.北京：人民邮电出版社.2008.
[51]阮秋琦,阮宇智.冈萨雷斯数字图像处理.北京：电子工业出版社第二版.2003.3.
[52]杨淑莹.VC++图像处理程序设计.北京：清华大学出版社第二版.2005：205-206页.
[53]何斌,马天予,王运坚.Visual C++数字图像处理.北京：人民邮电出版社.2002.
[54]求是科技.Visual C++数字图像处理典型算法及实现.北京：人民邮电出版社.2006.
[55]张筠莉,刘书智.Visual C++实践与提高串口通信与工程应用篇.北京：中国铁道出版社.2006.
[56]沈阳里奥科技电气部.LION-2007V205数控云台设备控制协议.2009年6月.
[2]杨立波,王旺.海事电视监控系统工程方案设计.水运工程.2003年10月总第357期.
[3]刘谦,梁兆棋,林丽.CCTV视频监控系统在广东内河水域的应用.中国水运.2009年6月.
[4]钱正锋.江苏海事局远程视频监控系统建设思考.中国水运.2008年1月.
[5]厦门海事局CCTV系统介绍.http://www.xmmsa.gov.cn/html/xmhsj/hsywzx/cctv/index.html
[6]丁鹭飞,耿富录,雷达原理.西安电子科技大学出版社.2002年6月.
[7]Fagueras O D, Toscani G. The calibration problem for stereo. IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Minmi Beach, Florida, June22-26,1986:15-20.
[8]Tsai R Y. A versatile camera calibration technique for high-accuracy 3D machine vision metrology using off-the-shelf cameras and lens. IEEE Transactions on Robotics and Automation,1987,3(4):323-344.
[9]Zhang Z. A flexible new technique for camera calibration. IEEE Transactions on Pattern Analysis and Machine Intelligence,2000,22(11):1330-1334.
[10]Kim J S, Kim H W, Kweon I S. A camera calibration method using concentric circles for vision applications. The 5th Asian Conference on Computer Vision,2002:515-520.
[11]Basu A. Active calibration:alternatibe strategy and analysis. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition,1993:495-500.
[12]Du F, Brady M. Self-calibration of the intrinsic parameters of camera for active vision system. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition,1993:477-482.
[13]Ma S D. A Self-calibration technique for active vision system. IEEE Transaction on Robotics and Automation,1996,12(1):114-120.
[14]Hartley R. Self-calibration of stationary cameras. International Journal of Computer Vision,1997,229(1):2-5.
[15]杨长江,孙凤梅,胡占义.基于二次曲面的纯旋转摄像机自标定.自动化学报,2001,27(3)：1130-1139.
[16]吴福朝,李华,胡占义.基于主动视觉系统的摄像机自标定研究.自动化学报,2001,27(6)：752-762.
[17]雷成,吴福朝,胡占义.一种新的基于主动视觉系统的摄像机自标定方法.计算机学报,2000,23(11)：1130-1139.
[18]扬长江,孙凤梅,胡占义.基于平面二次曲线的摄像机标定.计算机学报,2000,23(5)：541-547.
[19]李华,吴福朝,胡占义.一种新的线性摄像机自标定方法.计算机学报,2000,23(11)：1121-1129.
[20]傅丹等.一种摄像机自标定的线性方法.光电工程,2008,35(1)：71-75.
[21]Carvalho P C P, Szenberg F, Gattass M. Image-based modeling using a two-step camera calibration method. Proceedings of International Symposium on Computer Graphics, Image Processing and Vision,1998:388-395.
[22]Benallal M, Meunier J. Camera calibration with simple geometry. The Processing of 2003 International Conference on Image and Singal,2003.
[23]Yang A Y, Hong W, Ma Y. Structure and pose from single images of symmetric objects with applications to robot navigation. Proceeding of IEEE International Conference on Robotics & Automation, Taiper,2003:1013-1020.
[24]Wu Y H, Zhu H, Wu F C. Camera calibration from the quasi-affine invariance of two parallel circles. The 8th European Conference on Computer Vision(ECCV), 2004:190-202.
[25]贾云得.机器视觉.北京：科学出版社,2000.
[26]杨少荣,吴迪靖,段德山.机器视觉算法与应用.北京：清华大学出版社双语版.2008.
[27]马松德,张正友.计算机视觉.北京：科学出版社,1998.
[28]韩冰,林明星,丁凤华.机器视觉技术及其应用分析.农业装备与车辆工程.2008年第10期.
[29]王军,李星野.机器人视觉系统的组成及其工作原理.机器工程师.2007年第3期.
[30]张广军.视觉测量.北京：科学出版社.2008.
[31]徐德,谭民,李原.机器人视觉测量与控制==Visual measurement and control for robots.北京：国防工业出版社.2008.
[32]刘嵩鹤.基于双目立体视觉的安全车距测量技术研究：(硕士学位论文).长沙：武汉理工大学,2008.
[33]张义悦.基于机器视觉的三位检测定位理论研究：(硕士学位论文).昆明：昆明理工大学,2008.
[34]肖占春.基于双目立体视觉的机器人导航研究：(硕士学位论文).长春：东北师范大学,2008.
[35]崔岩.基于双目立体视觉的距离测量：(硕士学位论文).长春：长春理工大学,2006.
[36]常征.完善我国VTS服务的思考.中国海洋.2006第3期.
[37]才建胜.对国内船舶交通管制系统(VTS)现状的思考.天津航海.2006年第2期.
[38]何立居.基于VTS的海事信息发布系统.《航海技术》2006年第1期.
[39]赛迪.智能视频监控粉墨登场.中华建筑报2006年4月6日第007版电气智能化.
[40]张明友,汪学刚.雷达系统.北京：电子工业出版社.第二版：第1—28页.
[41]丁鹭飞,耿富录.雷达原理.西安：西安电子科技大学出版社.第三版：第1—24页.
[42]孙巍等.移动定位技术和移动定位系统.计算机系统应用.2003.10.
[43]夸拉特龙,萨莱里,李敏波.MATLAB科学计算.北京：清华大学出版社.2005.
[44]张志涌.精通MATLAB 6.5版.北京：北京航空航天大学出版社2003.3.
[45]杨磊等.闭路电视监控系统.北京：机械工业出版社,2003.5.
[46]http://www.net114.com/862817624/product/1284982.html
[47]http://www.net114.com/862908448/product/1236978.html
[48]http://www.net114.com/862904801/product/1176057.html
[49]宋坤等.Visual C++开发技术大全.北京：人民邮电出版社,2007.
[50]宋坤.Visual C++视频技术方案宝典.北京：人民邮电出版社.2008.
[51]阮秋琦,阮宇智.冈萨雷斯数字图像处理.北京：电子工业出版社第二版.2003.3.
[52]杨淑莹.VC++图像处理程序设计.北京：清华大学出版社第二版.2005：205-206页.
[53]何斌,马天予,王运坚.Visual C++数字图像处理.北京：人民邮电出版社.2002.
[54]求是科技.Visual C++数字图像处理典型算法及实现.北京：人民邮电出版社.2006.
[55]张筠莉,刘书智.Visual C++实践与提高串口通信与工程应用篇.北京：中国铁道出版社.2006.
[56]沈阳里奥科技电气部.LION-2007V205数控云台设备控制协议.2009年6月.