Fig. 1. Camera imaging model

Fig. 2. Simulation vision system based on absolute phase calibration method

Fig. 3. 22 simulated poses in the camera's view

Fig. 4. Part of the simulated images

Fig. 5. Simulated results. (a) Pixel error in u direction; (b) pixel error in v direction

Fig. 6. Absolute error at different noise levels. (a)(d)(g) Absolute error of checkerboard calibration method; (b)(e)(h) absolute error of circle calibration method; (c)(d)(g) absolute error of APT calibration method

Fig. 7. Absolute error at different fuzzy levels. (a)(d)(g) Absolute error of checkerboard calibration method; (b)(e)(h) absolute error of circle calibration method; (c)(d)(g) absolute error of APT calibration method

Fig. 8. APT experimental device

Fig. 9. Part calibration images and part feature points. (a)(c)(e) In-focus images; (b)(d)(f) out-of-focus images

Fig. 10. Target poses in the camera's view. (a) 16 poses under in-focus condition; (b) 14 poses under out-of-focus condition

Fig. 11. Re-projection errors of three calibration methods under in-focus condition. (a) Checkerboard calibration method; (b) circle calibration method; (c) APT calibration method

Fig. 12. Re-projection errors of three calibration methods when the target is far away from the effective working distance of the camera. (a) Checkerboard calibration method; (b) circle calibration method; (c) APT calibration method

Fig. 13. Column charts of world coordinate difference before and after optimization. (a)(d) Checkerboard calibration method under in-focus condition and out-of-focus condition; (b)(e) circle calibration method under in-focus condition and out-of-focus condition; (c)(f) APT calibration method under in-focus condition and out-of-focus condition