Laser & Optoelectronics Progress, Volume. 59, Issue 14, 1415015(2022)

Connector Surface Crack Detection Method

Wenxi Yu and Guili Xu*
Author Affiliations
  • College of Automation Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, Jiangsu , China
  • show less
    Figures & Tables(16)
    Flow chart of the designed method
    Original image
    Result of image binarization processing. (a) Image to be processed; (b) result of binarization processing
    Boundary detection result. (a) Image of the area to be detected; (b) masked image of the area to be detected
    12 structural elements constructed in this paper
    Comparison of extraction effect between the proposed method and black hat operation. (a) Black hat operation; (b) proposed method
    Columnwise projection result
    Image segmentation result
    Schematic of location features
    Schematic of spindle direction
    Schematic of the equivalent ellipse of the cracked area
    Detection result of Blob analysis
    Some pseudo cracks and detection results. (a) Pseudo-crack original images; (b) test results
    Physical picture of crack detection system
    Detection result comparison of different methods
    • Table 1. Detection results

      View table

      Table 1. Detection results

      MethodTotal number of samplesCorrectly identified number of samplesNumber of false detectionsNumber of missed testsFalse detection rate /%Accuracy /%
      Method in Ref.[520861848236211.488.6
      Method in Ref.[102086197710725.294.8
      Proposed method208620255832.997.1
    Tools

    Get Citation

    Copy Citation Text

    Wenxi Yu, Guili Xu. Connector Surface Crack Detection Method[J]. Laser & Optoelectronics Progress, 2022, 59(14): 1415015

    Download Citation

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

    Category: Machine Vision

    Received: Apr. 24, 2022

    Accepted: May. 20, 2022

    Published Online: Jul. 1, 2022

    The Author Email: Guili Xu (guilixu@nuaa.edu.cn)

    DOI:10.3788/LOP202259.1415015

    Topics