• High Power Laser and Particle Beams
  • Vol. 33, Issue 8, 081011 (2021)
Zongjia Shi1、2, Zhenjiao Xiang1, Yinglei Du1、*, Min Wan1, Jingliang Gu1, Guohui Li1, Rujian Xiang1, Jiang You1、2, Jing Wu1, and Honglai Xu1
Author Affiliations
  • 1Institute of Applied Electronics, CAEP, Mianyang 621900, China
  • 2Graduate School of China Academy of Engineering Physics, Beijing 100088, China
  • show less
    DOI: 10.11884/HPLPB202133.210040 Cite this Article
    Zongjia Shi, Zhenjiao Xiang, Yinglei Du, Min Wan, Jingliang Gu, Guohui Li, Rujian Xiang, Jiang You, Jing Wu, Honglai Xu. Wavefront reconstruction method based on far-field information and convolutional neural network[J]. High Power Laser and Particle Beams, 2021, 33(8): 081011 Copy Citation Text show less
    Wavefront reconstruction system based on CNN
    Fig. 1. Wavefront reconstruction system based on CNN
    Different residual elements
    Fig. 2. Different residual elements
    Flow chart of CNN wavefront reconstruction
    Fig. 3. Flow chart of CNN wavefront reconstruction
    Far-field and wavefront Zernike coefficients of a sample
    Fig. 4. Far-field and wavefront Zernike coefficients of a sample
    The L1 loss change process of the four ResNet models in training is shown in (a) the training set and (b) the verification set
    Fig. 5. The L1 loss change process of the four ResNet models in training is shown in (a) the training set and (b) the verification set
    Single frame image prediction time of ResNet
    Fig. 6. Single frame image prediction time of ResNet
    Wavefront reconstruction results of a sample in the test set
    Fig. 7. Wavefront reconstruction results of a sample in the test set
    Scatter plot of PV and RMS of original wavefront and wavefront residuals of test set samples
    Fig. 8. Scatter plot of PV and RMS of original wavefront and wavefront residuals of test set samples
    Ratio of PV and RMS of sample wavefront residuals to original wavefront of test set
    Fig. 9. Ratio of PV and RMS of sample wavefront residuals to original wavefront of test set
    R0far field image size/ pixel L1 error normalized coefficient RMSE PV of the test set samples’ original wavefront/μm RMS of the test set samples’ original wavefront/μm PV of reconstructed wavefront residuals/μm RMS of reconstructed wavefront residuals/μm residual PV to original wavefront ratio (90% of sample)/% residual RMS to original wavefront ratio (90% of sample)/%
    1140×1400.00400.00512.67±1.630.54±0.370.12±0.070.02±0.0165
    0.5200×2000.02040.02665.06±2.761.0±0.551.14±0.750.20±0.133027
    Table 1. Wavefront reconstruction results of turbulence with different intensities
    Zongjia Shi, Zhenjiao Xiang, Yinglei Du, Min Wan, Jingliang Gu, Guohui Li, Rujian Xiang, Jiang You, Jing Wu, Honglai Xu. Wavefront reconstruction method based on far-field information and convolutional neural network[J]. High Power Laser and Particle Beams, 2021, 33(8): 081011
    Download Citation