为提高摄像机标定精度，本文在分析Tsai两步标定法优缺点的基础上，提出了一种新的标定方法。该方法针对面阵CCD摄像机，建立新的综合畸变模型，并运用两步迭代法逐步逼近精确解，取消了Tsai两步法中对摄像机镜头只有径向畸变的限定，可用于多种复杂的镜头畸变情况。改进的方法只使用一平面标定板和单幅图象，操作简便，相比Tsai两步法中无法用平面标定板直接求解全部标定参数而言是一个突破。实验结果表明，改进方法速度快且精度较高，可广泛应用于机器视觉研究、工业三维测量等多个领域的摄像机标定。

To improve the accuracy of camera calibration, after analyzing the Tsai twostage method, the article gives a new camera calibration method based on the CCD area sensors array, and the new composite distortion model. It uses an iterative method without a radical distortion limit, and can be used in many kinds of camera lens distortion cases, and the improved method uses a coplanar set of points, monoview, and the experimental result shows that the algorithm is fast with high accuracy. This camera calibration method can be used in the machine vision research and the industry 3D measurement.