Laser & Optoelectronics Progress, Volume. 60, Issue 10, 1028005(2023)

Vehicle Location and Reidentification Using Multisource Point Clouds and Images

Wei Wang1、*, Lü Bin1, Yirui Yang1, Xinyu Hu1, Yuchun Huang2, Zhongtao Ye3,4, and Minghui Wang3
Author Affiliations
  • 1School of Mechanical Engineering, Hubei University of Technology, Wuhan 430068, Hubei, China
  • 2School of Remote Sensing and Information Engineering, Wuhan University, Wuhan 430072, Hubei, China
  • 3China Railway Bridge Science Research Institute, Ltd., Wuhan 430034, Hubei, China
  • 4State Key Laboratory for Health and Safety of Bridge Structures, Wuhan 430034, Hubei, China
  • show less
    Figures & Tables(16)
    Monitoring system for bridge's moving load based on WIM and multi-cameras
    Generation and application of vehicle RGB-D data. (a) Registration of 3D point cloud and RGB camera; (b) center of gravity estimation and virtual scene generation by using RGB-D data
    Point cloud and image acquisition devices over the WIM
    Calibration of point cloud and image acquisition devices
    Interpolation of point cloud
    Vehicle mockup
    Point cloud acquisition device
    Original partial point cloud. (a) Point cloud from top side; (b) point cloud from right side; (c) point cloud from left side; (d) point cloud from front side
    Full 3D point cloud of the model
    Influence of different number of sampling points on variation of Euclidean distance of normal vector
    Color image and point cloud grayscale image of vehicle model
    Colored cloud point grid of vehicle mockup
    Virtual scene and real image of vehicle
    Deviation diagram along Y coordinate
    • Table 1. Time consuming for point cloud registration of the proposed algorithm and SAC-IA

      View table

      Table 1. Time consuming for point cloud registration of the proposed algorithm and SAC-IA

      MethodTime consuming /s
      Fig.8(b)Fig.8(c)Fig.8(d)
      SAC-IA15.9716.8316.61
      Proposed method3.954.364.23
    • Table 2. Vehicle center in world coordinate at different scenes

      View table

      Table 2. Vehicle center in world coordinate at different scenes

      SceneVehicle mockup center in world coordinate /mm
      ActualEstimated from 2D bounding boxEstimated from colored point cloud
      1(14.87,-65.71,18.49)(14.91,-62.89)(14.88,-65.70,18.49)
      2(12.78,-34. 90,18.49)(12.82,-32.05)(12.78,-35.63,18.49)
      3(10.48,-28.95,18.49)(10.46,-26.21)(10.48,-30.73,18.49)
      4(10.33,-13.96,18.49)(10.27,-9.97)(10.34,-15.62,18.49)
      5(19.79,7.27,18.49)(19.81,12.79)(19.83,5.95,18.49)
      6(13.55,14.49,18.49)(13.39,21.18)(13.45,12.79,18.49)
      7(16.78,26.74,18.49)(16.70,33.84)(16.76,23.38,18.49)
      8(16.28,33.62,18.49)(16.11,45.38)(16.21,31.51,18.49)
    Tools

    Get Citation

    Copy Citation Text

    Wei Wang, Lü Bin, Yirui Yang, Xinyu Hu, Yuchun Huang, Zhongtao Ye, Minghui Wang. Vehicle Location and Reidentification Using Multisource Point Clouds and Images[J]. Laser & Optoelectronics Progress, 2023, 60(10): 1028005

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Category: Remote Sensing and Sensors

    Received: Jan. 28, 2022

    Accepted: Feb. 16, 2022

    Published Online: May. 17, 2023

    The Author Email: Wei Wang (wangw@hbut.edu.cn)

    DOI:10.3788/LOP220660

    Topics