Research On Camera Calibration Method In Large Field Of View

High-precision calibration of camera parameters is the basis for high-precision videogrammetry.The traditional calibration method requires that control points or control lines be uniformly distributed in the camera field of view to obtain a more stable and accurate calibration result.However,under the condition of large field of view,we can't easily distribute the control points uniformly in the view of camera field.In this paper,in order to solve the problem of camera calibration under the condition of large field of view,we proposed two kinds of calibration method: the one that based on the differential GPS system and the other that based on collimator.For the two calibration methods,we have done a lot of validation experiments.Experiments show that re-projection residuals of the two methods are small,and the two calibration methods can be effectively applied to high-precision videogrammetry in large field conditions.For the camera calibration method based on differential GPS system,the rover part of the differential GPS system is loaded on the unmanned aerial vehicle.The UAV is controlled to move within the field of view of the camera which to be calibrated and we can obtain the three-dimensional coordinates of the control point and the corresponding image coordinates.Then the initial values of the calibration parameters are solved according to the six-points or the optical center + four-points methods,and finally the accurate calibration parameters are obtained through optimization.For the camera calibration method based on collimator,we use collimator to create a series of infinity targets on its reticle and then use theodolites to measure the angle-pointing information of these infinity targets.Combined with the image coordinates of the corresponding infinity targets,the internal parameters of the camera are calculated according to the collinear equation.Finally,we fixed the calibrated camera on the working position and use several control points for absolute orientation,and then we can solve the camera's external parameters.