• Chinese Journal of Quantum Electronics
  • Vol. 41, Issue 3, 553 (2024)
YANG Miao1,*, ZHAN Ye2, FU Yuting3, and YANG Guang3
Author Affiliations
  • 1College of Electrical Engineering, Changchun Technical University of Automobile, Changchun 130013, China
  • 2College of Aviation Combat & Service, Aviation University of Air Force, Changchun , 130012, China
  • 3College of Instrumentation & Electrical Engineering, Jilin University, Changchun , 130061, China
  • show less
    DOI: 10.3969/j.issn.1007-5461.2024.03.017 Cite this Article
    Miao YANG, Ye ZHAN, Yuting FU, Guang YANG. Lithology analysis of rock with high repetition frequency laser⁃induced breakdown spectroscopy combined with convolutional neural network[J]. Chinese Journal of Quantum Electronics, 2024, 41(3): 553 Copy Citation Text show less
    Experiental Setup
    Fig. 1. Experiental Setup
    Experimental samples
    Fig. 2. Experimental samples
    Full spectrum of rock samples
    Fig. 3. Full spectrum of rock samples
    Spectral line calibration of rock sample (No.1/D03) (The intensity of the calibrated lines≥2000)
    Fig. 4. Spectral line calibration of rock sample (No.1/D03) (The intensity of the calibrated lines≥2000)
    Original spectra of rock sample (No.16/19DJ-5)
    Fig. 5. Original spectra of rock sample (No.16/19DJ-5)
    Spectra of rock sample (No.16/19DJ-5) after pretreatment
    Fig. 6. Spectra of rock sample (No.16/19DJ-5) after pretreatment
    Model 1D-CNN network parameter information
    Fig. 7. Model 1D-CNN network parameter information
    1D-CNN confusion matrix for rock 5 classification
    Fig. 8. 1D-CNN confusion matrix for rock 5 classification
    1D-CNN confusion matrix for rock 9 classification
    Fig. 9. 1D-CNN confusion matrix for rock 9 classification
    Model ResNet34 network parameter information
    Fig. 10. Model ResNet34 network parameter information
    ResNet34 confusion matrix for rock 5 classification
    Fig. 11. ResNet34 confusion matrix for rock 5 classification
    ResNet34 confusion matrix for rock 9 classification
    Fig. 12. ResNet34 confusion matrix for rock 9 classification
    Data preprocessing interface
    Fig. 13. Data preprocessing interface
    Test category interface
    Fig. 14. Test category interface
    NumberNameOriginLithologyCategory ACategory B
    1D03Great Khingancoarse sandstoneA5B1
    2D04Great Khinganmedium sandstoneA5B2
    3D06Great Khinganmedium-coarse sandstoneA5B3
    4D08Great Khingancoarse sandstoneA5B1
    5D10Great Khinganmedium-coarse sandstoneA5B3
    6P1-5Great Khingancoarse sandstoneA5B1
    7P1-7Great KhingansandstoneA5B4
    8P1-11Great Khingan

    conglomerate-medium-

    coarse sandstone

    A4B5
    9P1-13-1Great KhinganandesiteA3B6
    10P1-13-2Great KhinganandesiteA3B6
    11P1-13-3Great KhinganandesiteA3B6
    12P1-13-4Great KhinganandesiteA3B6
    13P1-13-5Great KhinganandesiteA3B6
    1419DJ-1Shuangyang Cityconglomerate sandstoneA1B7
    1519DJ-4Shuangyang CitysandstoneA5B8
    1619DJ-5Shuangyang Citymiscellaneous sandstoneA2B9
    Table 1. Information of rock samples
    True valuePredicted value
    10
    1TPFN
    0FPTN
    Table 2. Confusion matrix of dichotomous problem
    LabelTotal numberCorrect numberAccuracy/%Overall accuracy/%
    A1848297.698.86
    A2636298.4
    A37373100
    A47070100
    A5605998.3
    Table 3. 1D⁃CNN result for rock 5 classification
    LabelTotal numberCorrect numberAccuracy/%Overall accuracy/%
    B154275093.02
    B26969100
    B3756485.3
    B46969100
    B5767396.1
    B66464100
    B77676100
    B8787798.7
    B9696797.1
    Table 4. 1D⁃CNN result for rock 9 classification
    LabelTotal numberCorrect numberAccuracy/%Overall accuracy/%
    A1848410099.43
    A2636298.4
    A37373100
    A47070100
    A5605998.3
    Table 5. ResNet34 result for rock 5 classification
    LabelTotal numberCorrect numberAccuracy/%Overall accuracy/%
    B1544685.297.14
    B26969100
    B3756992
    B46969100
    B5767497.4
    B66464100
    B7767598.7
    B87878100
    B9696898.6
    Table 6. ResNet34 result for rock 9 classification
    Categorization1D-CNNResNet34
    Classification 598.86%99.43%
    Classification 993.02%97.14%
    Table 7. Comparative analysis of experimental results
    Miao YANG, Ye ZHAN, Yuting FU, Guang YANG. Lithology analysis of rock with high repetition frequency laser⁃induced breakdown spectroscopy combined with convolutional neural network[J]. Chinese Journal of Quantum Electronics, 2024, 41(3): 553
    Download Citation