摘要
着重解决如何将抛物面镜与普通CCD摄像机配置成满足中心投影约束的全景摄像机及其参数标定问题.首先通过数值仿真实验给出全景摄像机的配置与拟合误差之间的关系图,从而确定最佳参数配置范围.接着,研究了利用空间平面的图像信息标定抛物面镜摄像机的算法.在摄像机内参数已知情况下,借助平面诱导的单应矩阵将标定问题转化为多项式特征值问题,从而获得待标定参数的解析解,避开复杂的数值计算,达到快速标定参数的目的.结果表明,该算法最少只需5对图像匹配点便可以实现问题求解.最后,通过数值仿真实验和真实图像实验验证算法的有效性和鲁棒性.
This paper mainly solves the problems of how to efficiently configure and calibrate a parabolic camera composed by one parabolic mirror and one traditional CCD camera, which satisfies the constraint of central projection. Firstly, the relation- ship between various configuration parameters and its corresponding system errors is tested and depicted using numerical simulation experiments, with which the range for best configuration parameters is obtained. Then, a new algorithm is suggested for calibrating the parabolic camera with the image information of a spatial plane. With the known intrinsic parameters, the calibration problem is formulated into a fourth order polynomial eigenvalue problem via the plane-induced Homographic matrix. An analytic solution is obtained to avoid the complicated numerical computation, and hence rapid calibration is achieved. The results Mso show that five or more pairs of corresponding image points are sufficient for solving this problem. Finally, numeri- cal simulation and real data experiments are performed to validate the accuracy and robustness of the proposed algorithm, and satisfactory results are obtained in those experiments.
出处
《系统科学与数学》
CSCD
北大核心
2015年第1期31-41,共11页
Journal of Systems Science and Mathematical Sciences
基金
国家自然科学基金(60802087)
国家电子商务信息处理国际联合研究中心项目(2013B01035)资助课题
关键词
抛物面镜摄像机
中心投影约束
单应矩阵
多项式特征值问题.
Parabolic camera, central projection constraint, homographic matrixpolynomial eigenvalue problem.