US 12,243,271 B2
Calibration board and calibration method and system
Lihua Wu, Shenzhen (CN); Xianzhuo Liu, Shenzhen (CN); Minjian Pang, Shenzhen (CN); Zhiming Huang, Shenzhen (CN); Xiaomeng Wang, Shenzhen (CN); Jie Zeng, Shenzhen (CN); and Hongfei Yang, Shenzhen (CN)
Assigned to Orbbec Inc., Shenzhen (CN)
Filed by ORBBEC INC., Guangdong (CN)
Filed on May 26, 2022, as Appl. No. 17/825,404.
Application 17/825,404 is a continuation of application No. PCT/CN2020/090925, filed on May 18, 2020.
Claims priority of application No. 202010210792.0 (CN), filed on Mar. 24, 2020.
Prior Publication US 2022/0284630 A1, Sep. 8, 2022
Int. Cl. G06T 7/80 (2017.01); G06T 7/62 (2017.01); H04N 17/00 (2006.01)
CPC G06T 7/80 (2017.01) [G06T 7/62 (2017.01); H04N 17/002 (2013.01); G06T 2207/30244 (2013.01)] 15 Claims
OG exemplary drawing
 
1. A calibration method, comprising:
obtaining a calibration image of a calibration board captured by a camera, wherein the calibration board comprises checkerboard cells arranged on a surface of the calibration board, and at least one dot is arranged in each of the checkerboard cells, and the dots of the checkerboard cells comprise at least a first feature dot and a second feature dot, and a diameter of the first feature dot is greater than a diameter of the second feature dot;
determining a position of the calibration board in the calibration image, and detecting dots on the calibration board;
gridding the detected dots based on the checkerboard cells to obtain a first two-dimensional matrix;
determining types of the detected dots, binarizing the detected dots based on the first two-dimensional matrix according to the types of the detected dots, to obtain a second two-dimensional matrix;
matching the second two-dimensional matrix with a calibration board number template matrix, to obtain dot-centroid coordinate number information of the second two-dimensional matrix; and
outputting, according to the obtained dot-centroid coordinate number information, corner-point number information corresponding to the checkerboard cells in which the detected dots are located, and calculating calibration data, according to the dot-centroid coordinate number information, the corner-point number information, and a corresponding camera model, by using a calibration algorithm.