Journal of Applied Optics, Volume. 44, Issue 3, 677(2023)

Image classification of optical element surface defects based on convolutional neural network

Jinyao HOU1... Weiguo LIU1,*, Shun ZHOU1, Aihua GAO1, Shaobo GE1 and Xiangguo XIAO2 |Show fewer author(s)
Author Affiliations
  • 1College of Photoelectric Engineering, Xi'an Technological University, Xi'an 710021, China
  • 2Xi'an Institute of Applied Optics, Xi'an 710065, China
  • show less
    Figures & Tables(11)
    Surface defect images
    MER model of surface defects
    Classification flow chart of surface defect images of optical elements
    Establishment of dataset for surface defect images
    AlexNet network model
    Asymmetric convolution process
    Proposed network structure
    • Table 1. Feature extraction of surface defects

      View table
      View in Article

      Table 1. Feature extraction of surface defects

      表面缺陷长轴/pixel短轴/pixel面积/pixel2外接矩形面积/pixel2长宽比矩形度P
      1114114
      2114114
      11318332.67
      12228412
      13321061.51.67
      168648501.330.96
      179856721.130.78
      1811.267.736987.071.460.79
      1910.299.8478101.201.050.77
      2032.6531.7588104.281.030.84
      2687.7428.8713762 532.953.10.54
    • Table 2. Comparison of network model parameters before and after improvement

      View table
      View in Article

      Table 2. Comparison of network model parameters before and after improvement

      模型模型总参数
      Alexnet58 271 811
      Alexnet+AC54 405 027
    • Table 3. Identification results of proposed network%

      View table
      View in Article

      Table 3. Identification results of proposed network%

      表面缺陷ScratchDigNoisy
      准确率10099.7597.4
    • Table 4. Model comparison experiments

      View table
      View in Article

      Table 4. Model comparison experiments

      Model灵敏度/%特异性/%精确性/%运行时间/s
      VGG1669.3991.2096.4081.1
      InceptionV371.7492.2997.1279.35
      ResNet5073.8794.1797.3278.56
      AlexNet77.2092.3497.2018.54
      Ours89.9796.7399.0518.46
    Tools

    Get Citation

    Copy Citation Text

    Jinyao HOU, Weiguo LIU, Shun ZHOU, Aihua GAO, Shaobo GE, Xiangguo XIAO. Image classification of optical element surface defects based on convolutional neural network[J]. Journal of Applied Optics, 2023, 44(3): 677

    Download Citation

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

    Category: Research Articles

    Received: May. 30, 2022

    Accepted: --

    Published Online: Jun. 19, 2023

    The Author Email: LIU Weiguo (wgliu@163.com)

    DOI:10.5768/JAO202344.0305003

    Topics