Fig. 1. Camera pinhole projection model

Fig. 2. Ideal marginal model. (a) Original ideal edge model; (b) rotated ideal edge model

Fig. 3. Edge point simulation map

Fig. 4. The process of edge contour sampling and point taking

Fig. 5. Schematic diagram of sorting simulated feature points

Fig. 6. Schematic diagrams of feature point sorting before and after. (a) Unsorted circular calibration plate; (b) sorted circular calibration plate

Fig. 7. Central projection of perspective model and real camera model

Fig. 8. The process of feature point correction and compensation

Fig. 9. Circle feature edge extraction. (a) Original image; (b) Canny operator; (c) proposed algorithm

Fig. 10. Simulated circular image. (a) Simulated projection image; (b) Gaussian white noise local image with σ=0.2

Fig. 11. Positioning error when adding white noise with different standard deviations

Fig. 12. Calibration board pose

Fig. 13. The image of the checkerboard and the center calibration plate in the case of large distortion. (a) Checkerboard image under distortion; (b) effect picture of circle center extraction

Fig. 14. Reprojection error scatterplot. (a) Zhang's method; (b) traditional circle fitting method; (c) proposed algorithm