Camera imaging model is a key factor to determine the accuracy of vision measurement. With the increasingly demand for accuracy in visual measurement, improvement of the accuracy of camera model and calibration are primary means to solve the accuracy problem. A new model of the rational function lens distortion model is introduced into the camera imaging model, which can better correct the errors caused by lens distortion, and thereby, it can improve the accuracy of the camera imaging model. Moreover, a two-step and step-iterative optimization method is proposed to precisely obtain the parameters of camera imaging model and to solve the problem that the parameters of the rational function lens distortion model cannot be solved directly. The experimental results demonstrate that the accuracy of the camera imaging model with the new distortion correction model is better than the conventional model and the new model can also effectively reduce the large errors. Therefore, the new camera imaging model and calibration method can improve the accuracy of visual measurement comprehensively.