钢轨轮廓全断面检测中的匹配方法研究
Research on matching algorithm in rail profile inspection

采用激光视觉测量技术对钢轨全断面廓形进行检测,在动态测量过程中由于车辆的随机振动,会影响钢轨轮廓数据的检测精度。因此,在高速动态条件下实现检测钢轨轮廓与标准廓形高精度自动匹配,是当前轨道廓形检测中面临的关键问题。在分析国内外钢轨廓形检测。廓形匹配现状的基础上,通过对目前采用最多的迭代最近点ICP(Iterative Close Point)算法进行简化,同时优化匹配过程中的对应点搜索策略,在保证匹配精度的前提下,解决了钢轨廓形检测中的匹配问题,达到实时性检测的需求。将该算法运用在地铁轨道检测设备中,验证了该方法的有效性。
Inspecting the whole section by laser vision measuring technology for rail profile, due to the random vibration of the vehicle in the process of dynamic measurement, affects the accuracy of rail profile data. Therefore, it is a key problem in the current orbital profile test to realize the automatic matching of the steel rail profile and the standard profile with high speed dynamic conditions. Detection based on the analysis of domestic and foreign rail profiles, profile matching on the basis of the status quo, by using Iterative closest Point to simplify the algorithm, and optimal matching corresponding points in the process of search strategy were performed under the premise that the precision of matching was ensured to solve the matching problem of rail profile detection and to achieve real-time detection requirements. Finally, the algorithm was applied to the subway track detection equipment, and the effectiveness of this method was verified