• Optics and Precision Engineering
  • Vol. 32, Issue 2, 208 (2024)
Tongmeng WANG1, Fen GAO1,*, and Bing LI2
Author Affiliations
  • 1College of Optoelectronic Engineering, Xi′an Technological University, Xi'an70032, China
  • 2State Key Laboratory for Manufacturing Systems Engineering, Xi'an Jiaotong University, Xi'an710049, China
  • show less
    DOI: 10.37188/OPE.20243202.0208 Cite this Article
    Tongmeng WANG, Fen GAO, Bing LI. Phase unwrapping technology about point diffraction interference fringe based on atrous spatial convolutional networks[J]. Optics and Precision Engineering, 2024, 32(2): 208 Copy Citation Text show less
    Neural network structure of ASPP-CNN
    Fig. 1. Neural network structure of ASPP-CNN
    Structure of ASPP
    Fig. 2. Structure of ASPP
    Training and testing process of neural networks
    Fig. 3. Training and testing process of neural networks
    Loss function curve
    Fig. 4. Loss function curve
    Unwrapping results of sphere
    Fig. 5. Unwrapping results of sphere
    Unwrapping results of random surface data
    Fig. 6. Unwrapping results of random surface data
    Optimized unwrapping results and errors of spherical and random surfaces
    Fig. 7. Optimized unwrapping results and errors of spherical and random surfaces
    Wrapped phase data with simulated environmental noise and unwrapping result
    Fig. 8. Wrapped phase data with simulated environmental noise and unwrapping result
    Point diffraction interferometry system
    Fig. 9. Point diffraction interferometry system
    Seven-step phase-shift interferograms collected by the point diffraction interferometry system
    Fig. 10. Seven-step phase-shift interferograms collected by the point diffraction interferometry system
    Wrapped phase and unwrapped phase by ASPP-CNN method and 2D surface shape
    Fig. 11. Wrapped phase and unwrapped phase by ASPP-CNN method and 2D surface shape
    Surface shape, PV and RMS values obtained by ESDI interferogram analysis software
    Fig. 12. Surface shape, PV and RMS values obtained by ESDI interferogram analysis software
    Unwrapping results, 3D surface fitting results, and errors under different methods
    Fig. 13. Unwrapping results, 3D surface fitting results, and errors under different methods
    图像种类函数
    仿真球面

    参考球面:(X1-R0)2+y2+z2=R02

    被测球面:(X2-a)2+y-b2+(z-c)2=R12

    随机矩阵

    自由曲面

    W2=(ymax-ymin)*(φ-xmin)/(xmax-xmin)+ymin

    泽尼克拟合

    自由曲面

    W3ρ,θ=n=0km=-nnCnmZnm
    Table 1. Unctions and parameters of simulated datasets
    项目ASPP-CNNFFT法U-NETDeepLabv3+专业图像处理软件
    PV/λ0.303 90.819 80.320 80.312 00.291 8
    PV差值/λ0.012 10.5280.0290.020 2-
    RMS/λ0.057 30.132 50.059 40.057 90.053 1
    RMS差值/λ0.004 20.079 40.006 30.004 8-
    Times/s0.03510.560.0320.193.38
    RMSE/rad0.022 211.638 51.2320.269 1-
    Table 2. PV, RMS values, and processing time of different methods
    Tongmeng WANG, Fen GAO, Bing LI. Phase unwrapping technology about point diffraction interference fringe based on atrous spatial convolutional networks[J]. Optics and Precision Engineering, 2024, 32(2): 208
    Download Citation