基于最优匹配区域的结构光系统的投影仪标定

针对现有的投影仪标定算法精度不高的问题,根据投影仪的非线性模型,提出一种基于最优匹配区域的新投影仪标定算法。首先通过试验得到最优匹配区域,然后在最优匹配区域内计算角点的局部单应性矩阵,最后采用相机标定的方法对投影仪进行亚像素级标定。实验结果表明,该方法标定过程简便,投影仪标定误差最大为0.10617 pixel,算法的执行时间为0.1907 s,系统三维重构的点云密度达9.14个/mm2。该方法只需要普通的平面标定板,操作简单,可满足高精度的三维测量系统要求。

Projector calibration of structured light system based on optimal matching area

As the precision of existing projector calibration algorithms is low,a new calibration method based on the optimal matching area is proposed according to projector nonlinear model. Firstly,the optimal matching area is obtained through the experiments. Then,local homographies of corners in the optimal matching area are computed. Finally,the camera calibration method is used to calibrate the projector which reach the sub-pixel precision. The experiment results show that the proposed projector calibration algorithm is easy to operate. The calibration precision can achieve 0. 10617 pixel. The runtime costs 0.1907 s and point cloud density is 9.14 per mm2,which meets the measurement requirements of high precision.