Author Affiliations
1School of Automation, Nanjing University of Information Science & Technology, Nanjing 210044, Jiangsu , China2Jiangsu Provincial Collaborative Innovation Center of Atmospheric Environment and Equipment Technology, Nanjing 210044, Jiangsu , Chinashow less
Fig. 1. Principle of binary coding combined with error diffusion
Fig. 2. Schematic diagram of binary coding combined with error diffusion
Fig. 3. S-shaped path scanning
Fig. 4. Contrast of fringes. (a)(b) Local enlargement of images; (c)(d) four binary images; (e)(f) Gaussian filter image; (g)(h) grayscale value of Gaussian filter image
Fig. 5. Solution for standard sinusoidal phase. (a) Wrapped phase of the 400th row partial pixel; (b) unwrapped phase
Fig. 6. Simulation comparison of phase root mean square error with different periods
Fig. 7. Experimental comparison of phase root-mean-square error with different periods
Fig. 8. Experimental results of phase solving calibration plates. (a)-(c) Captured pictures; (d)-(f) unwrapped phase; (g)-(i) point cloud of calibration plate
Fig. 9. Results of sinusoidal fitting. (a) T=32 pixel; (b) T=96 pixel
Fig. 10. Measurement results of precision ball. (a) Point cloud of precision ball; (b) reconstruction results; (c) fitted sphere data; (d) fitting error distribution
Fig. 11. Reconstruction results of statue. (a) Unwrapped phase; (b) full renderings; (c) local enlarged view
Fringe period | Four-step phase shift method | | Traditional error diffusion combined with binary coding method | | Optimized error diffusion combined with binary coding method |
---|
RMSE | SSE | | RMSE | SSE | | RMSE | SSE |
---|
T=32 pixel | 1.6263 | 666.47 | | 1.0293 | 267.01 | | 0.8745 | 192.75 | T=96 pixel | 1.6501 | 718.86 | | 1.1816 | 368.56 | | 1.0465 | 289.12 |
|
Table 1. Sinusoidal fitting error analysis
Method | Diameter | Average distance | Standard deviation |
---|
Four-step phase shift | 50.7805 | -0.008973 | 0.092646 | Traditional error diffusion combined with binary coding | 50.8188 | 0.008043 | 0.094340 | Optimized error diffusion combined with binary coding | 50.8178 | 0.006544 | 0.093207 |
|
Table 2. fitting sphere accuracy analysis