LIU Meng, LIU Chang-jie, WU Xiao-han
（State Key Laboratory of Precision Measuring Technology and Instruments, Tianjin University, Tianjin 300072, China）
Abstract： High-speed vehicle dynamic envelope curve is defined as the maximum limit outline affected by a variety of adverse factors while the train is running. Considering the difficulties in the current measurement system such as complicated calibration process, cumbersome aided-instruments, strict limitation of working distance, this paper carries out an optical method in which two high-speed cameras with variable-zoom lenses are adopted as binocular stereo sensors of measurement system and a high-accuracy 3D target with fast reconstruction is designed. The intrinsic parameters of the sensors and the relative positions between coordinate systems are solved by the method of colinearity constrained optimization algorithm. The calibration process is easy to operate and the device is also of portability. Most importantly, the severe working distance limitation between sensors and measured body is solved, enhancing the adaptability of measurement system to environment. Experimental results show that when the sensors are in the range of 8-16 m away from the measured body, system accuracy can reach up to ±0.5 mm, which meets the requirements to measure the dynamic envelope curve of high-speed vehicle.
Key words： zoom lens; 3D target; colinearity constrain optimization algorithm; binocular stereo vision; dynamic envelope curve
CLD number： U238 Document code： A
Article ID： 1674-8042（2016）04-0398-07 doi： 10.3969/j.issn.1674-8042-2016-04-015
References
［1］HAN Mei, HAN Bo-ling, LI Hong-yan, et al. Calculation method of the distance between railway out-of-gauge goods and structure gauge. China Railway Science, 2011, 32(1): 122-126.
［2］TENG Wan-xiu, CHENG Ya-jun. Comparative study on computation methods for dynamic car profiles. Urban Rapid Rail Transit, 2009, 22(4): 40-45.
［3］GUO Yin, LIU Chang-jie, et al. Measurement system for dynamic envelope curve of high-speed train. Chinese Journal of Lasers, 2013, 40(2): 0208002.
［4］SUN Jun-hua, WU Zi-yan, LIU Qian-zhe, et al. Field calibration of stereo vision sensor with large FOV. Optics and Precision Engineering, 2009, 17(3): 633-640.
［5］GUO Yin. Research on dynamic envelope curve measurement of high-speed train.Tianjin: Tianjin University, 2013.
［6］Abdel-Aziz Y I. Photogrammetric potential of non-metric cameras. Illinois: University of Illinois at Urbana-Champaign, 1974.
［7］Zhang H J. Research on fast field calibration in measurement of high-speed vehicle dynamic envelope curve. Tianjin: Tianjin University, 2015.
［8］Weng J, Cohen P, Herniou M. Camera calibration with distortion models and accuracy evaluation. IEEE Transactions on Pattern Analysis & Machine Intelligence, 1992, 14(10): 965-980.
［9］WEI Guo-qing, MA Song-de. Implicit and explicit camera calibration: theory and experiments. IEEE Transactions on Pattern Analysis & Machine Intelligence, 1994, 16(5): 469-480.
高速列车动态包络线测量系统现场校准方法的优化
刘蒙， 刘常杰， 吴晓寒
（天津大学 精密测试技术及仪器国家重点实验室， 天津 300072）
摘要:高速列车动态包络线是指车体在运行过程中受各种不利因素影响所导致的最大极限轮廓。 针对当前动态包络线测量系统存在标定过程繁琐、 辅助设备不便携带、 系统受场地因素制约严重等缺点， 本文提出采用两个配置变焦镜头的高速相机作为测量系统的双目视觉传感器， 设计并制作可快速重构的高精度三维靶标， 通过共线性约束优化算法完成视觉传感器内参数和各空间位置关系参数的标定工作。 该标定方法操作简单， 标定设备便于携带， 同时克服了双目相机与被测车体间严格的距离约束， 增强了测量系统对复杂场地的适应性。 实验表明， 当双目相机距被测车体8-16 m时， 系统标定精度可达±0.5 mm， 满足高速列车动态包络线的测量要求。
关键词:变焦镜头； 三维靶标； 共线性约束优化算法； 双目视觉； 动态包络线
引用格式：LIU Meng, LIU Chang-jie, WU Xiao-han. An optimal on-site calibration method in measurement system of dynamic envelope curve of high-speed vehicle . Journal of Measurement Science and Instrumentation, 2016, 7(4)： 398-404. ［doi： 10.3969／j.issn.1674-8042.2016-04-015］
[full text view]
