Infrared and Laser Engineering, Volume. 50, Issue 11, 20210125(2021)

Real time 3D vision screening method of ore under complex lighting conditions

Xiuhua Zhang1,2,3, Hanyu Hong1,2、*, Yangyang Xu1,2,3, and Tianxu Zhang4
Author Affiliations
  • 1Hubei Key Laboratory of Optical Information and Pattern Recognition, Wuhan 430205, China
  • 2Hubei Engineering Technology Research Center of Video Image and HD Projection, Wuhan 430205, China
  • 3School of Mechanical & Electrical Engineering, Wuhan Institute of Technology, Wuhan 430205, China
  • 4Institute for Pattern Recognition and Artificial Intelligence, Huazhong University of Science and Technology, Wuhan 430074, China
  • show less
    Figures & Tables(18)
    Ore transportation image on site
    Spectral intensity of solar radiation
    Schematic diagram of infrared structured light ore image acquisition system
    Ore images of visible and infrared structured light under different illumination conditions. (a) Green line laser (no direct sunlight); (b) Infrared laser (no direct sunlight); (c) Green line laser (strong light); (d) Infrared laser (strong light)
    Ideal gray distribution of line structured light cross section. (a) Ideal Gaussian distribution; (b) Flattened Gaussian distribution
    Gray distribution of visible and infrared structure cross section under different light environment. (a) Green line laser (no direct sunlight); (b) Infrared laser (no direct sunlight); (c) Green line laser (strong light); (d) Infrared laser (strong light)
    Schematic diagram of sliding differential template. (a) Backward; (b) Forward
    Forward cumulative difference
    Measurement of gauge block. (a) Schematic diagram of gauge block projection line; (b) Schematic diagram of height formation
    Extraction result of different method. (a) Gauss method; (b) Steger method; (c) Proposed method
    Results of local center line deviation. (a) Gauss method; (b) Steger method; (c) Proposed method
    Endpoint centerline bifurcation by Steger method
    Enlarged result in the yellow box in Fig.12. (a) Box A; (b) Box B; (c) Box C; (d) Box D
    3D point cloud display and screening results of ore
    • Table 1. Coordinates of the center point with different values of m (Unit: pixel)

      View table
      View in Article

      Table 1. Coordinates of the center point with different values of m (Unit: pixel)

      Column 761m=3 m=5 m=7 m=11
      Center point coordinates296.6066296.6066296.6308296.6308
    • Table 2. Height measurement results of the gauge block with proposed method (Unit: mm)

      View table
      View in Article

      Table 2. Height measurement results of the gauge block with proposed method (Unit: mm)

      Gauge blockStandard block sizeProposed methodMeasurement errors
      H1100.02399.979−0.044
      H270.03169.997−0.034
      H339.99840.029+0.031
    • Table 3. Comparison of off-line structured light centerline extraction results

      View table
      View in Article

      Table 3. Comparison of off-line structured light centerline extraction results

      Method/resultR2Off-line time/ms
      Gauss fitting0.956020.8
      Steger0.952628.6
      Proposed0.96584.9
    • Table 4. Comprehensive comparison of online ore screening methods

      View table
      View in Article

      Table 4. Comprehensive comparison of online ore screening methods

      Screening methodInfrared structured light 3D visionArtificial vision2D image processing[2-3]
      Light environmentComplex lightingDistinguish and clearlyModerate light
      Screening efficiency1-2 m/sInefficiencyAbout 0.4 m/s
      AccuracyHighLowInstable
      Automation degree100%ArtificialSemi-automatic
      SecuritySecurityUnsafeSecurity
    Tools

    Get Citation

    Copy Citation Text

    Xiuhua Zhang, Hanyu Hong, Yangyang Xu, Tianxu Zhang. Real time 3D vision screening method of ore under complex lighting conditions[J]. Infrared and Laser Engineering, 2021, 50(11): 20210125

    Download Citation

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

    Category: Image processing

    Received: Mar. 2, 2021

    Accepted: --

    Published Online: Dec. 7, 2021

    The Author Email: Hanyu Hong (hhyhong@163.com)

    DOI:10.3788/IRLA20210125

    Topics