A calibration method for stereovision system based on solid circle target

Abstract

All kinds of calibration methods have many optimal variables and complicated computation. How to calibrate stereovision system rapidly and precisely is always a pressing problem in the field of visual measurement. Therefore, a calibration method for stereovision system based on solid circle target is proposed in the present paper. A solid circle calibration pattern is adopted to calibrate the stereovision system. The centers of solid circles on calibration pattern are extracted by Canny-Zernike combination algorithm and sorted based on a triangle mark on calibration pattern. According to the pinhole camera model, the intrinsic and extrinsic parameters of left and right cameras are firstly obtained. Then, the distortion model including four parameters is adopted, Levenberg-Marquardt algorithm is used to separately optimize the image coordinates of projection of 3D point on the imaging plane of left and right cameras, the intrinsic parameters including distortion coefficients and extrinsic parameters of stereovision system are acquired in this step. In the last step, the distance between two centers on diagonal direction of calibration pattern is regarded as the constraint to further optimize the extrinsic parameters of stereovision system. Furthermore, a calibration software for stereovision system based on solid circle target is compiled with MATLAB, and the comparison test of two calibration methods and two calibration patterns are performed in lab. Results show that the proposed method has a higher precision, can automatically proceed calibration and can be widely used in engineering.