Journals Articles Conferences News About CLP
Submit a paper Email Alert
Search
Search
Articles
Journals
News
Advanced Search
Top Searches
2D MaterialsTransformation opticsQuantum PhotonicsSmall lasersSemiconductor UV PhotonicsSupersymmetric microring laser arrays

Laser & Optoelectronics Progress, Volume. 57, Issue 10, 101013(2020)

Classification of Carbon Fiber Reinforced Polymer Defects Based on One-Dimensional CNN

Xianglin Zhan** and Wanting Zhao*
Author Affiliations
  • College of Electronic Information and Automation, Civil Aviation University of China, Tianjin 300300, China
    • show less
    AbstractGet PDF(in Chinese)
    Figures&Tables (11)
    Equations (0)
    References (20)
    Cited By (3)
    Tools
    Paper Information
    Topics
    Figures & Tables(11)
    Overall process of automatic identification of CFRP defect types

    Fig. 1. Overall process of automatic identification of CFRP defect types

    Download full size
    Structure diagram of U-1DCNN

    Fig. 2. Structure diagram of U-1DCNN

    Download full size
    Structure diagram of parallel structure of multi-convolution blocks

    Fig. 3. Structure diagram of parallel structure of multi-convolution blocks

    Download full size
    Structure diagrams of the residual units. (a) No matching dimensions; (b) matching dimensions

    Fig. 4. Structure diagrams of the residual units. (a) No matching dimensions; (b) matching dimensions

    Download full size
    Schematic of one-dimensional convolution calculation

    Fig. 5. Schematic of one-dimensional convolution calculation

    Download full size
    Bayesian optimization algorithm

    Fig. 6. Bayesian optimization algorithm

    Download full size
    Visualization of the activation from U-1DCNN convolutional layers

    Fig. 7. Visualization of the activation from U-1DCNN convolutional layers

    Download full size
    Dimensionality reduction visualization of U-1DCNN features

    Fig. 8. Dimensionality reduction visualization of U-1DCNN features

    Download full size
    Comparison of classification accuracy of different methods

    Fig. 9. Comparison of classification accuracy of different methods

    Download full size

    • Table 1. Specifications of CFRP test blocks and dataset composition

      View table

      Table 1. Specifications of CFRP test blocks and dataset composition

      CFRP test blocksTest block 1Test block 2Test block 3
      Number of layers323232
      Thickness of test blocks /mm444
      Distribution of defectsUsing polytetrafluoroetylene(PTFE) films (thickness is0.25mm) to simulate delamination;a total of 30 defects, locatedat 2nd to 31st layersUsing PTFE films(thickness is 0.25mm)to simulate delamination;defect shapes:square and circleAdding hollow glassmicrospheres tosimulate gas cavity
      Defect type in thedatasetDelaminationNon-defectDelaminationNon-defectGas cavity
      Number of defectsin the dataset13001300390013002600

    • Table 2. Comparison of evaluation indicators of different methods

      View table

      Table 2. Comparison of evaluation indicators of different methods

      Defect typeEvaluationindicatorBP+WPTBP+SFCNN+STFTU-1DCNN
      DelaminationPrec /%80.8588.3096.70100.00
      R /%91.8399.33100.00100.00
      Gas cavityPrec /%97.7299.01100.0098.04
      R /%100.0099.6793.33100.00
      Non-defectPrec /%80.1499.55100.00100.00
      R /%56.5074.0099.8398.00
      Average valuePrec /%86.2495.6298.9099.36
      R /%82.7891.0097.7299.33
      F1 /%84.4793.2598.3199.34
      Error rate /%14.966.921.710.50
    Tools

    Get Citation

    Copy Citation Text

    Xianglin Zhan, Wanting Zhao. Classification of Carbon Fiber Reinforced Polymer Defects Based on One-Dimensional CNN[J]. Laser & Optoelectronics Progress, 2020, 57(10): 101013

    Download Citation

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

    Category: Image Processing

    Received: Sep. 16, 2019

    Accepted: Oct. 18, 2019

    Published Online: May. 8, 2020

    The Author Email: Xianglin Zhan (xlzhan@cauc.edu.cn), Wanting Zhao (zhaowtwt@163.com)

    DOI:10.3788/LOP57.101013

    Topics

    laser devices and laser physics
    Lasers and Laser Optics
    Laser physics
    laser manufacturing
    Instrumentation, Measurement and Metrology