• Acta Photonica Sinica
  • Vol. 52, Issue 1, 0112002 (2023)
Jisheng XU1,2,3, Jiaojiao REN1,2,3,*, Dandan ZHANG1,2,3, Jian GU1,2,3..., Jiyang ZHANG3, Lijuan LI1,2,3 and Junwen XUE3|Show fewer author(s)
Author Affiliations
  • 1Key Laboratory of Photoelectric Measurement and Optical Information Transmission Technology of Ministry of Education,Changchun University of Science and Technology,Changchun 130022,China
  • 2College of Optoelectronic Engineering,Changchun University of Science and Technology,Changchun 130022,China
  • 3Zhongshan Institute of Changchun University of Science and Technology,Zhongshan 528400,China
  • show less
    DOI: 10.3788/gzxb20235201.0112002 Cite this Article
    Jisheng XU, Jiaojiao REN, Dandan ZHANG, Jian GU, Jiyang ZHANG, Lijuan LI, Junwen XUE. Terahertz Identification of Hot Melting Joint Defects in Polyethylene Pipe Based on Wavelet Scattering Network[J]. Acta Photonica Sinica, 2023, 52(1): 0112002 Copy Citation Text show less
    Actual hot melting welding process of polyethylene sheet
    Fig. 1. Actual hot melting welding process of polyethylene sheet
    Physical of THz-TDS system
    Fig. 2. Physical of THz-TDS system
    THz detection simulation diagram of polyethylene pipe
    Fig. 3. THz detection simulation diagram of polyethylene pipe
    THz detection waveform comparison
    Fig. 4. THz detection waveform comparison
    The sample has not undergone planar processing and corresponding THz kurtosis imaging after hot fusion welding
    Fig. 5. The sample has not undergone planar processing and corresponding THz kurtosis imaging after hot fusion welding
    The Incomplete fusion sample before welding and the corresponding THz detection kurtosis imaging diagram
    Fig. 6. The Incomplete fusion sample before welding and the corresponding THz detection kurtosis imaging diagram
    The embedded object of inclusion defect sample and the corresponding THz kurtosis imaging are required
    Fig. 7. The embedded object of inclusion defect sample and the corresponding THz kurtosis imaging are required
    Defect classification flow based on wavelet scattering network-convolution neural network
    Fig. 8. Defect classification flow based on wavelet scattering network-convolution neural network
    Wavelet scattering framework
    Fig. 9. Wavelet scattering framework
    CNN Network architecture
    Fig. 10. CNN Network architecture
    Accuracy and loss of each iteration of the first training model
    Fig. 11. Accuracy and loss of each iteration of the first training model
    Confusion matrix of test results using the first kind of training model
    Fig. 12. Confusion matrix of test results using the first kind of training model
    Accuracy and loss of each iteration of the second training model
    Fig. 13. Accuracy and loss of each iteration of the second training model
    Confusion matrix of test results using the second kind of training model
    Fig. 14. Confusion matrix of test results using the second kind of training model
    The comparison chart between the number of defect recognition and the actual number of two training models
    Fig. 15. The comparison chart between the number of defect recognition and the actual number of two training models
    Curve diagram of error between the number of defect recognition and the actual number of two training models
    Fig. 16. Curve diagram of error between the number of defect recognition and the actual number of two training models
    Sample typesFirst model defect type/Number of data setsSecond model defect type/Number of data sets
    Standard weldingBZHJ/3 000BZHJ/3 000
    Incomplete fusionWRH/3 000/
    Inclusion metalJZ/1 000JS/3 000
    Inclusion of coarse sandJZ/1 000CS/3 000
    Inclusion branchesJZ/1 000SZ/3 000
    Table 1. Dataset property table
    Jisheng XU, Jiaojiao REN, Dandan ZHANG, Jian GU, Jiyang ZHANG, Lijuan LI, Junwen XUE. Terahertz Identification of Hot Melting Joint Defects in Polyethylene Pipe Based on Wavelet Scattering Network[J]. Acta Photonica Sinica, 2023, 52(1): 0112002
    Download Citation