欢迎访问机床与液压官方网站!

咨询热线:020-32385312 32385313 RSS EMAIL-ALERT
基于双边滤波的三维曲面重构技术在钢轨检测中的应用
作者:
作者单位:

作者简介:

通讯作者:

中图分类号:

基金项目:

国家自然科学基金(51665014;52065021;51865009;51975210);江西省重点研发计划科技项目(20192BAB206022)


Application of 3D Surface Reconstruction Technology Based on Bilateral Filtering in Rail Inspection
Author:
Affiliation:

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    钢轨廓形的测量对指导钢轨打磨、提高列车运行安全、延长钢轨寿命具有重要的意义。采用三维激光扫描重建技术,获取钢轨廓形点云数据;分别采用迭代最近邻算法、双边滤波法以及曲率精简法对钢轨点云数据进行拼接、降噪和精简处理;对比在不同高斯滤波参数下,该技术对钢轨点云数据的滤波效果。结合UM动力学软件和有限元分析软件对三维重构后的钢轨模型进行仿真对比分析。结果表明:采用该三维重构技术可以快速地对钢轨的型面进行重构;高斯滤波参数σc为20时,对钢轨点云数据的滤波效果较好,能够使钢轨轨顶保持良好的细节特征。

    Abstract:

    The measurement for rail profile is of great significance to guide rail grinding, improve train operation safety and prolong rail life. The 3D laser scanning reconstruction technology was used to obtain the rail profile point cloud data; iterative closest points algorithm, bilateral filtering method and curvature reduction method were used to splice, reduce noise and simplify the rail point cloud data respectively; the filtering effect of this technology on the rail point cloud data under different Gaussian filtering parameters was compared. Combined with UM-dynamics software and finite element analysis software, the rail model after 3D reconstruction was simulated and analyzed. The results show that by using the 3D reconstruction technology, the rail profile can be reconstructed quickly; when the Gaussian filter parameter σc is 20, the filtering effect of this technology on the rail point cloud data is better, and the detail characteristics of the rail top can be kept well.

    参考文献
    相似文献
    引证文献
引用本文

林凤涛,胡伟豪,杨洋,李志和,肖乾.基于双边滤波的三维曲面重构技术在钢轨检测中的应用[J].机床与液压,2021,49(6):6-11.
LIN Fengtao, HU Weihao, YANG Yang, LI Zhihe, XIAO Qian. Application of 3D Surface Reconstruction Technology Based on Bilateral Filtering in Rail Inspection[J]. Machine Tool & Hydraulics,2021,49(6):6-11

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:
  • 最后修改日期:
  • 录用日期:
  • 在线发布日期: 2022-03-24
  • 出版日期: 2021-03-28