Camera calibration method based on non-metric distortion correction

In order to keep consistency of lens distortion parameters calibration and simplify online camera calibration process in vision measurement, a distortion separated camera model is constructed and the linear method for camera calibration based on non-metric distortion correction is proposed. Based on the perspective projection invariance of collinear points, the distortion parameters of fixed-focus-lens are calibrated at one go by using massive collinear feature points and the camera internal parameters can be calibrated linearly by using feature points after distortion correction. The proposed method separates distortion parameters from the camera model and distortion parameters do not need to be incorporated into the camera model for multiple repeated calibration. Experimental results show that the distortion separated camera model is in good agreement with imaging regulation of fixed-focus-lens, and linear method for camera calibration based on non-metric distortion correction can obtain equivalent accuracy as traditional nonlinear calibration method and improve calibration speed near 3-fold. It advances machine vision one step to real engineering metrology applications by simplifying calibration process and improving vision measurement convenience.