Automobile four-wheel positioning refers to the accurate measurement of a specific set of angles of the car wheel. The accuracy of the four-wheel positioning is directly related to the overall performance of the car. Camera calibration as the first step in four-wheel positioning measurement and three-dimensional modeling is of great significance for improving the accuracy of four-wheel positioning measurement based on machine vision. In order to improve the accuracy of four-wheel positioning measurement based on machine vision, this paper based on the imaging model of the camera, based on Zhang Zhengyou calibration method, combined with openCV to calibrate the camera, and calculate the actual internal parameters and distortion parameters of the camera, which lay the foundation for the next four-wheel positioning parameters measurement.