道路交通事故现场摄影测量的研究
|
摘要
近年来，由于机动车辆剧增，道路交通事故的处理面临着新的挑战。以人工方式，手工勘查交通事故现场时间长、精度低，且容易发生漏测。这样就给后续的事故责任鉴定带来了不必要的麻烦。数字摄影测量与现代图像处理技术相结合，已成为交通现场间接测量的有效手段之一。
二维图像中的非线性畸变和拍摄图像的相机参数标定，是摄影测量的技术核心。如何充分利用先进的仪器设备，简化操作过程，提高测量精度，是本课题关心的主要问题。本文在分析了数码相机特点的基础上，提出将二维图像中的非线性畸变与相机参数标定分别求解的方案，减少了参数间的相互干扰。
在分析相机非线性畸变产生的主要原因之后，建立了非线性畸变数学模型，提出利用标准图形法求解相机镜头产生的非线性畸变的方法，并设计了求解方案。这样，可以在实施摄影测量之前，先对图像进行校正，消除图像中非线性畸变的影响，简化后续处理的过程。
由于道路交通事故现场的各种测量多是在路面上进行，故在本课题中，采用二维分析法。在校正图像中非线性畸变的基础上，只考虑路面这一二维平面上的尺寸测量，可以大大减少了所需标定点的数量，并简化了运算过程。
试验证明上述方案较为简便易行，且测量精度完全可以满足实际应用的要求。
In recent years, because the motor vehicle increases severely, the treatment of the traffic accident on the road faces the new challenge. Time is long to reconnoitre the scene of a traffic accident by hands. The precision is low and it's apt to leak the scales. These bring the unnecessary trouble to the accident of follow-up responsibility determine. Digital photogrammetric combining with modem image process technology has become one of effective means for indirect measure in scene of the traffic accident.
nonlinear distortion in 2-D image and calibration of camera's interior and external parameters are key techniques of photogrammetry. How to Fully utilize the advanced instrument and equipment, simplify the operation course and improve the precision of measuring is the main matter which the subject concerns. After analyzing the character of digital camera the paper brings forward the project which solves the nonlinear distortion and digital camera's parameters respectively. This method decreases the interference among the parameters.
After analyzing the main reasons which cause the nonlinear distortion, the mathematical model of image nonlinear distortion is built. The subject designs the project and brings forward the method which solves the nonlinear distortion by standard gridding. Before photogrammetry, correct the image first and dispel the influence of the nonlinear distortion. The method simplify the course of follow-up treatment.
Because the most of measures of the traffic accident are on the road, the subject adopt 2-D analysis method. This decreases the number of the calibration points greatly.
The test proves that above-mentioned schemes are comparatively simple , easy to do and the precision of measure can meet practical application.
二维图像中的非线性畸变和拍摄图像的相机参数标定，是摄影测量的技术核心。如何充分利用先进的仪器设备，简化操作过程，提高测量精度，是本课题关心的主要问题。本文在分析了数码相机特点的基础上，提出将二维图像中的非线性畸变与相机参数标定分别求解的方案，减少了参数间的相互干扰。
在分析相机非线性畸变产生的主要原因之后，建立了非线性畸变数学模型，提出利用标准图形法求解相机镜头产生的非线性畸变的方法，并设计了求解方案。这样，可以在实施摄影测量之前，先对图像进行校正，消除图像中非线性畸变的影响，简化后续处理的过程。
由于道路交通事故现场的各种测量多是在路面上进行，故在本课题中，采用二维分析法。在校正图像中非线性畸变的基础上，只考虑路面这一二维平面上的尺寸测量，可以大大减少了所需标定点的数量，并简化了运算过程。
试验证明上述方案较为简便易行，且测量精度完全可以满足实际应用的要求。
In recent years, because the motor vehicle increases severely, the treatment of the traffic accident on the road faces the new challenge. Time is long to reconnoitre the scene of a traffic accident by hands. The precision is low and it's apt to leak the scales. These bring the unnecessary trouble to the accident of follow-up responsibility determine. Digital photogrammetric combining with modem image process technology has become one of effective means for indirect measure in scene of the traffic accident.
nonlinear distortion in 2-D image and calibration of camera's interior and external parameters are key techniques of photogrammetry. How to Fully utilize the advanced instrument and equipment, simplify the operation course and improve the precision of measuring is the main matter which the subject concerns. After analyzing the character of digital camera the paper brings forward the project which solves the nonlinear distortion and digital camera's parameters respectively. This method decreases the interference among the parameters.
After analyzing the main reasons which cause the nonlinear distortion, the mathematical model of image nonlinear distortion is built. The subject designs the project and brings forward the method which solves the nonlinear distortion by standard gridding. Before photogrammetry, correct the image first and dispel the influence of the nonlinear distortion. The method simplify the course of follow-up treatment.
Because the most of measures of the traffic accident are on the road, the subject adopt 2-D analysis method. This decreases the number of the calibration points greatly.
The test proves that above-mentioned schemes are comparatively simple , easy to do and the precision of measure can meet practical application.
引文
[1] Kenneth R. Castleman . Digital Image Processing . Publishing House of Electronics Industry 2002
[2] Karlsruhe, Kurt Schwiedefsky, Friedrich Ackermann . Grundri β der Photogrammetrie 1984
[3] C.J. Hilditch . Linear skeletons from square cupboards in machine intell (B. Meltzer and D. Michie, Eds).New York: Amer. Elsevier. 1969. pp403-420. vol. 4.
[4] Clive S. Fraser . Automated processes in digital photogrammetric calibration, orientation, and triangulation . Digital Signal Processing 8, 277-283 (1998)
[5] M. Pollefeys, R. Koch, M. Vergauwen, L. Van Gool. Automated reconstruction of 3D rcenes from sequences of images. Photogrammetry & Remote Sensing55 (2000) 251-267
[6] Heping Pan. A Direct closed-Form solution to general relative orientation of two stereo views . Digital Signal Processing 9, 195-221 (1999)
[7] Qiang Ji, Mauro S. Costa, Robert M. Haralick, Linda G. Shapiro. A Robust linear least-squares estimation of camera exterior orientation using multiple geometric features . Photogrammetry & Remote Sensing55 (2000) 75-93
[8] Dimitrios Ioannou, Walter Huda, Andrew F. Laine . Circle recognition through a 2D hough transform and radius histogramming. Imageand Vision Computing 17 (1999) 15-26
[9] Petros Patias. Medical imaging challenges photogrammetry. ISPRS Journal of Photogrammetry & Remote Sensing 56 (2002) 295- 310
[10] H.L. Mitchell, I. Newton. Medical photogrammetric measurement: overview and prospects. ISPRS Journal of Photogrammetry & Remote Sensing 56 (2002) 286 - 294
[11] 崔屹．数字图像处理技术及应用．电子工业出版社1997
[12] 傅以诺．交通事故处理与车辆保险．人民交通出版社
[13] 姜大志．计算机视觉中三维重构的研究与应用2001
[14] 李江、江国宪、李洪才．摄影测量技术在交通事故再现中的应用1995
[15] 王澍权．道路交通事故分析与处理方法．人民交通出版社 1998
[16] 张新生、黄小波、朱述龙．道路交通事故现场快速测绘系统．解放军测绘学院学报1999年第3期
[17] 张凯丽，刘辉．边缘检测技术的发展研究．昆明理工大学学报 2000年第10期
[18] 王丰元、孙壮志、周一鸣．计算机视觉摄影测量三维控制场设计．农业机械学报1998年第3期
[19] 王丰元等．计算机视觉测量中CCD摄像机的标定．中国农业大学学报 1996，1(6)：99～103
[20] 王丰元、宇仁德、杜守旭．道路交通事故现场摄影测量原理与模型试验．中国公路学报2001年第1期
[21] 侯宇．圆和椭圆边缘检测的快速方法．中国计量学院学报2000年第2期
[22] 王成儒、胡正平、练秋生．一种高效的混合圆／椭圆检测方法．贵州工业大学学报2002年第4期
[2] Karlsruhe, Kurt Schwiedefsky, Friedrich Ackermann . Grundri β der Photogrammetrie 1984
[3] C.J. Hilditch . Linear skeletons from square cupboards in machine intell (B. Meltzer and D. Michie, Eds).New York: Amer. Elsevier. 1969. pp403-420. vol. 4.
[4] Clive S. Fraser . Automated processes in digital photogrammetric calibration, orientation, and triangulation . Digital Signal Processing 8, 277-283 (1998)
[5] M. Pollefeys, R. Koch, M. Vergauwen, L. Van Gool. Automated reconstruction of 3D rcenes from sequences of images. Photogrammetry & Remote Sensing55 (2000) 251-267
[6] Heping Pan. A Direct closed-Form solution to general relative orientation of two stereo views . Digital Signal Processing 9, 195-221 (1999)
[7] Qiang Ji, Mauro S. Costa, Robert M. Haralick, Linda G. Shapiro. A Robust linear least-squares estimation of camera exterior orientation using multiple geometric features . Photogrammetry & Remote Sensing55 (2000) 75-93
[8] Dimitrios Ioannou, Walter Huda, Andrew F. Laine . Circle recognition through a 2D hough transform and radius histogramming. Imageand Vision Computing 17 (1999) 15-26
[9] Petros Patias. Medical imaging challenges photogrammetry. ISPRS Journal of Photogrammetry & Remote Sensing 56 (2002) 295- 310
[10] H.L. Mitchell, I. Newton. Medical photogrammetric measurement: overview and prospects. ISPRS Journal of Photogrammetry & Remote Sensing 56 (2002) 286 - 294
[11] 崔屹．数字图像处理技术及应用．电子工业出版社1997
[12] 傅以诺．交通事故处理与车辆保险．人民交通出版社
[13] 姜大志．计算机视觉中三维重构的研究与应用2001
[14] 李江、江国宪、李洪才．摄影测量技术在交通事故再现中的应用1995
[15] 王澍权．道路交通事故分析与处理方法．人民交通出版社 1998
[16] 张新生、黄小波、朱述龙．道路交通事故现场快速测绘系统．解放军测绘学院学报1999年第3期
[17] 张凯丽，刘辉．边缘检测技术的发展研究．昆明理工大学学报 2000年第10期
[18] 王丰元、孙壮志、周一鸣．计算机视觉摄影测量三维控制场设计．农业机械学报1998年第3期
[19] 王丰元等．计算机视觉测量中CCD摄像机的标定．中国农业大学学报 1996，1(6)：99～103
[20] 王丰元、宇仁德、杜守旭．道路交通事故现场摄影测量原理与模型试验．中国公路学报2001年第1期
[21] 侯宇．圆和椭圆边缘检测的快速方法．中国计量学院学报2000年第2期
[22] 王成儒、胡正平、练秋生．一种高效的混合圆／椭圆检测方法．贵州工业大学学报2002年第4期