Automatic Detection and Sorting Algorithm for Checkerboard Corner Points

摘要

In view of the detection of checkerboard corners in camera calibration, the advantages and disadvantages of automatic detection and manual detection are analyzed. In combination with Hessian Matrix theory and Harris algorithm, dynamic self-adaptive determination is performed by calculating the image shape operator and using a circular symmetric template. Corner points and culling pseudo-corner points were used to automatically identify poles using the principle of Minimum Convex Hull, and sort corners and sub-pixel positions. Experiments show that the algorithm has a wide range of applications, high precision, and good robustness.