A novel practical approach of camera calibration for binocular stereo vision system

The paper introduces a simple and efficient approach to calibrate camera in the parallel binocular stereo vision system, which uses numerical analysis based on least square error to obtain the relative difference of the two cameras´ principle point and extrinsic parameters. First, we use Zhang Plane-based Calibration Method to verify the intrinsic parameters of cameras; second, we use least square error function of the fitted plane to obtain the relative difference of the two cameras´ principle points; third, according to triangulation of best fitted plane to extract extrinsic parameters. The results of experiments show that the method is effective and flexible.