增强现实中的摄像机径向畸变校正

Correction Method for Camera Radial Distortion for Augmented Reality

增强现实系统中,有效地进行摄像机镜头畸变校正对提高虚拟环境的精确性具有重要意义.首先提出一种基于成像几何的畸变校正方法,采用带有一阶径向畸变的摄像机模型,对镜头径向畸变进行校正,再根据校正后的图像计算摄像机投影矩阵.实验表明,基于成像几何的畸变校正算法具有较高的畸变校正精度,640×480的图像中,最大畸变量达90多个像素.与利用理想针孔摄像机模型得到的增强现实环境相比,畸变校正后得到的叠加结果更为精确.

In augmented reality systems, it is important to correct the image distortions in order to build precise virtual environmental model. Using the pin-hole camera model with first-order radial lens distortion, an image distortion correction method is proposed for radial lens distortion based on imaging geometry. The projective matrix is calculated using the revised image for overlaying the virtual environmental model on the image. The max distortion calculated with image of 640 ×480 by this method is beyond 90 pixels. Comparison the corrected image with the real image is given to demonstrate that this algorithm is effective. The experimental results show that the virtual model calculated by this algorithm is more precise compared to that calculated by the ideal pin-hole camera model.