LAUSR

Abstract The traditional camera calibration methods based on internal parameters and external parameters employ a uniform analytic function to meet the requirements of statistical optimization. It may fail in describing the whole measurement volume properly when the distortion is strong. A novel whole-field point to point calibration method is proposed, in which the mapping relationship between the pixel coordinates and world coordinates of each individual spatial point is described by look-up tables. To guarantee the high calculation accuracy of the world coordinates of each spatial point, a virtual telecentric optical projection technology is used to encode the phase of each spatial position by loading a carrier frequency to establish a virtual world coordinate system. Then the mapping tables between the eponymous pixel and the corresponding virtual world coordinates of each calibrated position with known height can be established employing polynomial fitting or interpolation methods. At last, the real world coordinates of each spatial point are calculated by removing the additional displacement caused by the carrier frequency. Employing our method, errors caused by system installation and lens distortion, etc. can be effectively ruled out. An exhaustive comparison between the proposed calibration method and the traditional 3D stereo target calibration (3DSTC) method based on pinhole model with optimization strategy is given. Our method yields more accurate calibration results than 3DSTC method because it is closer to the real physical model. Experiment results show the feasibility and validity of the proposed method. It is recommended to be used in monocular measurement system when higher accuracy is requested.