基于直线投影特征的镜头畸变校正方法

Lens distortion correction method based on linear projection features

摘要:

针对摄像机镜头畸变校正方法的简便和快速问题，设计了基于直线投影特征的校正方法。介绍了镜头主要畸变产生原因和畸变模型；给出直线三点在理想投影下的关系，确定了适应度函数，利用遗传算法得到了畸变参数组最优解。基于matlab软件编写校正程序，并进行了实验验证。实验表明，利用畸变参数组的最优解能够实现图像畸变校正，效果较好。该标定方法只需场景内有直线存在即可实现对摄像机镜头畸变参数校正，方法所需实验条件简单，程序简便，便于现场快速校正。

Abstract:

In order to solve the simple and fast problem of camera lens distortion correction, a method based on linear projection feature was designed. The main reason and distortion model of lens are introduced. The relation between three points of a straight line under ideal projection is given, and the fitness function is determined. Matlab software based on the preparation of calibration procedures, and experimental verification. The experiment shows that the image distortion correction can be achieved by using the optimal solution of the distortion parameter set. The calibration method in this paper can realize the correction of camera lens distortion parameters only if there is a straight line in the scene. The experimental conditions are simple, the program is simple, and the field correction is convenient.