Chinese Journal of Lasers, Volume. 47, Issue 11, 1104006(2020)

Dynamic Inspection of Wheel Profile Based on ROI-RSICP Algorithm

Yi Qian1, Zhong Haoyu1, Liu Long1, Liu Wenlong1,2, and Yi Bing1、*
Author Affiliations
  • 1School of Traffic and Transportation Engineering, Central South University, Changsha, Hunan 410075, China
  • 2School of Mechanical Engineering, Hanyang University, Seoul 0 4763, South Korea
  • show less
    Figures & Tables(15)
    Measured wheel profiles under different situations. (a) Intersection line of laser plane and wheel surface not crossing wheel axis; (b) wheel profile with distortion; (c) intersection line of laser plane and wheel surface crossing wheel axis; (d) wheel profile without distortion
    Wear evolution of wheel profile
    Point cloud segmentation network of PoinNet
    Flow chart of dynamic inspection of wheel profile
    Total loss versus training times
    mIoU versus training times
    Partial results of extracted ROI based on PointNet. (a) Standard; (b) slightly worn; (c) badly worn
    Accuraccy comparison among ICP, SICP and proposed methods. (a) Original model; (b) ICP; (c) SICP; (d) ROI-RSICP
    Registration results of wheel profile under different IoU values. (a) Slightly worn; (b) badly worn
    Difference between inspection result by proposed method and that by the 4th kind of inspector versus IoU
    Stability comparison of ICP, SICP and proposed methods under different included angles. (a) 30°; (b) 35°; (c) 40°; (d) 45°; (e) 50°; (f) 55°; (g) 60°
    Difference between tread wear calculated by each method and result measured by the 4th kind of inspector
    • Table 1. Tread wear detected by ICP, SICP, ROI-RSICP and artificial methods unit: mm

      View table

      Table 1. Tread wear detected by ICP, SICP, ROI-RSICP and artificial methods unit: mm

      MethodData typePosition 1Position 2Position 3Position 4Position 5Position 6
      ICPSlightly worn1.601.741.881.232.231.49
      SICP1.131.271.141.251.561.36
      ROI-RSICP1.761.941.991.752.091.84
      Artificial method1.801.901.901.802.001.70
      ICPbadly worn2.703.112.552.961.042.43
      RSICP0.991.011.812.212.141.78
      ROI-RSICP4.194.114.154.214.124.39
      Artificial method4.104.004.104.304.104.50
    • Table 2. Repeated detection results of badly worn wheel unit: mm

      View table

      Table 2. Repeated detection results of badly worn wheel unit: mm

      Number ofmeasurementsFlange heightFlange thicknessTread wear
      ROI-RSICPArtificial methodROI-RSICPArtificial methodROI-RSICPArtificial method
      131.1431.1030.1130.204.144.10
      231.1831.1030.0930.104.184.10
      331.1231.0030.1230.204.124.00
      431.1331.0030.0730.104.134.00
      531.1231.1030.1230.204.124.10
      631.2031.1030.1030.104.204.10
      Mean31.1531.0730.1030.154.154.07
      SD0.0330.0510.01940.0540.0330.051
    • Table 3. Dynamic inspection results of wheel geometric parameters unit: mm

      View table

      Table 3. Dynamic inspection results of wheel geometric parameters unit: mm

      ParameterMethodMeasurement resultMaximumerror
      123456
      Flange heightROI-RSICP32.1132.0832.1132.0732.0832.04
      Artificial method32.232.132.132.132.232.10.12
      Flange thicknessROI-RSICP31.0831.1231.1331.1231.1431.14
      Artificial method31.231.231.231.131.231.20.12
      Tread wearROI-RSICP5.115.085.115.075.085.04
      Artificial method5.15.15.15.05.15.00.07
    Tools

    Get Citation

    Copy Citation Text

    Yi Qian, Zhong Haoyu, Liu Long, Liu Wenlong, Yi Bing. Dynamic Inspection of Wheel Profile Based on ROI-RSICP Algorithm[J]. Chinese Journal of Lasers, 2020, 47(11): 1104006

    Download Citation

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

    Category: Measurement and metrology

    Received: Jun. 4, 2020

    Accepted: --

    Published Online: Oct. 23, 2020

    The Author Email: Bing Yi (bingyi@csu.edu.cn)

    DOI:10.3788/CJL202047.1104006

    Topics