Laser & Optoelectronics Progress, Volume. 58, Issue 16, 1630006(2021)

Wire Rope Defect-Detection Method Based on Otsu Segmentation and Edge Detection

Ruochen Dai1, Mingfu Zhao1, Bin Tang1、*, Liyong Dai1, Taojiang Wu1,2, and Shanghai Jiang1
Author Affiliations
  • 1Intelligent Fiber Sensing Technology of Chongqing Municipal Engineering Research Center of Institution of Higher Education, Chongqing Key Laboratory of Fiber Optic Sensor and Photodetector, Chongqing University of Technology, Chongqing 400054, China
  • 2Tielian Operation and Maintenance of Chongqing Municipal Engineering Research Center of Institution of Higher Education, Chongqing 402260, China
  • show less
    Figures & Tables(10)
    Wire rope detection system
    Damaged wire rope sample
    Hyperspectral images of wire rope with defects. (a) 410 nm band image; (b) 470 nm band image; (c) 566 nm band image; (d) 661 nm band image; (e) 737 nm band image; (f) 793 nm band image
    Spectral curves of the wire rope area with the defect and the background area. (a) Reflectivity of randomly selected ROIs; (b) two parts of ROI reflectivity after average processing
    Hyperspectral images and gray histograms. (a) 406 nm band image and its grayscale histogram; (b) 566 nm band image and its grayscale histogram; (c) 645 nm band image and its grayscale histogram
    Image threshold segmentation results at different wavelengths. (a) 406 nm band image; (b) 566 nm band image; (c) 645 nm band image
    Comparison of edge detection results based on different operators. (a) Roberts operator; (b) Sobel operator; (c) Prewitt operator; (d) Canny operator
    Hyperspectral images after Hough transform. (a) Based on Roberts operator; (b) based on Sobel operator; (c) based on Prewitt operator; (d) based on Canny operator
    • Table 1. Number of pixels at different positions of the wire rope

      View table

      Table 1. Number of pixels at different positions of the wire rope

      Position1234567
      Number of pixels Ni94.1391.0193.0292.1463.0191.0092.02
      Absolute deviation value2.111.011.000.1229.011.020
      Distance value2.091.000.990.1228.721.000
    • Table 2. Treatment effect of different methods

      View table

      Table 2. Treatment effect of different methods

      Treatment methodMSEPSNRSSIM
      Otsu segmentation1888.415.36990.9572
      Hough transform5316.810.87430.0024
      Otsu+Hough650.919.99570.9404
    Tools

    Get Citation

    Copy Citation Text

    Ruochen Dai, Mingfu Zhao, Bin Tang, Liyong Dai, Taojiang Wu, Shanghai Jiang. Wire Rope Defect-Detection Method Based on Otsu Segmentation and Edge Detection[J]. Laser & Optoelectronics Progress, 2021, 58(16): 1630006

    Download Citation

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

    Category: Spectroscopy

    Received: Sep. 8, 2020

    Accepted: Oct. 21, 2020

    Published Online: Aug. 16, 2021

    The Author Email: Tang Bin (tangbin@cqut.edu.cn)

    DOI:10.3788/LOP202158.1630006

    Topics