• Laser & Optoelectronics Progress
  • Vol. 57, Issue 24, 241020 (2020)
Zihao Yu1, Jin Liu1、*, Haima Yang2、*, Pengcheng Zhang1, and Yi Chen2
Author Affiliations
  • 1College of Electronic and Electrical Engineering, Shanghai University of Engineering Science, Shanghai 201620, China
  • 2College of Optical-Electrical and Computer Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China
  • show less
    DOI: 10.3788/LOP57.241020 Cite this Article Set citation alerts
    Zihao Yu, Jin Liu, Haima Yang, Pengcheng Zhang, Yi Chen. Three-Dimensional Surface Reconstruction Based on Edge Detection and Reliability Sorting Algorithm[J]. Laser & Optoelectronics Progress, 2020, 57(24): 241020 Copy Citation Text show less
    Schematic diagram of the structured light 3D reconstruction system
    Fig. 1. Schematic diagram of the structured light 3D reconstruction system
    Schematic diagram of the triangulation method
    Fig. 2. Schematic diagram of the triangulation method
    Reliability value between adjacent pixels. (a) Reliability value of a single pixel; (b) reliability value of the expanded path
    Fig. 3. Reliability value between adjacent pixels. (a) Reliability value of a single pixel; (b) reliability value of the expanded path
    Process of reliability sequencing. (a) First time; (b) second time; (c) third time; (d) fourth time
    Fig. 4. Process of reliability sequencing. (a) First time; (b) second time; (c) third time; (d) fourth time
    Edge detection results of the three operators. (a) Laplacian operator; (b) LoG operator; (c) Canny operator
    Fig. 5. Edge detection results of the three operators. (a) Laplacian operator; (b) LoG operator; (c) Canny operator
    Process of edge detection processing. (a) Ear modulation map with missing information; (b) image obtained after NMS processing; (c) image obtained after edge processing
    Fig. 6. Process of edge detection processing. (a) Ear modulation map with missing information; (b) image obtained after NMS processing; (c) image obtained after edge processing
    Flow chart of phase unwrapping
    Fig. 7. Flow chart of phase unwrapping
    Error analysis of phase unwrapping. (a) 3D description of the phase function; (b) phase unwrapping errors of different algorithms
    Fig. 8. Error analysis of phase unwrapping. (a) 3D description of the phase function; (b) phase unwrapping errors of different algorithms
    Phase unwrapping results of our algorithm in different environments. (a) Ideal situation; (b) random noise; (c) information loss; (d) random noise and information loss
    Fig. 9. Phase unwrapping results of our algorithm in different environments. (a) Ideal situation; (b) random noise; (c) information loss; (d) random noise and information loss
    Physical map of the built system
    Fig. 10. Physical map of the built system
    Image and ear model modulation after the raster is expanded. (a) Unfolded image of grating phase; (b) modulated grating image of ear model
    Fig. 11. Image and ear model modulation after the raster is expanded. (a) Unfolded image of grating phase; (b) modulated grating image of ear model
    Results obtained by the Goldstein branch cutting method. (a) 2D image of the ear model; (b) ear model after phase expansion; (c) true height of the ear model
    Fig. 12. Results obtained by the Goldstein branch cutting method. (a) 2D image of the ear model; (b) ear model after phase expansion; (c) true height of the ear model
    Wrapping phase and phase unfolding diagrams of the two algorithms. (a)-(b) Least square method; (c)-(d) our algorithm
    Fig. 13. Wrapping phase and phase unfolding diagrams of the two algorithms. (a)-(b) Least square method; (c)-(d) our algorithm
    Phase unwrapping results of the two algorithms. (a) Least squares algorithm (2D); (b) our algorithm (2D); (c) least squares algorithm (3D); (d) our algorithm (3D)
    Fig. 14. Phase unwrapping results of the two algorithms. (a) Least squares algorithm (2D); (b) our algorithm (2D); (c) least squares algorithm (3D); (d) our algorithm (3D)
    Cochlear model obtained by 3D reconstruction
    Fig. 15. Cochlear model obtained by 3D reconstruction
    Simulation environmentIdeal situationRandom noiseInformation lostRandom noise and information lost
    Simulation time0.9086.0780.6853.217
    Table 1. Phase unfolding time of our algorithm in different environments unit: s
    Zihao Yu, Jin Liu, Haima Yang, Pengcheng Zhang, Yi Chen. Three-Dimensional Surface Reconstruction Based on Edge Detection and Reliability Sorting Algorithm[J]. Laser & Optoelectronics Progress, 2020, 57(24): 241020
    Download Citation