椭圆拟合的圆环模板摄像机标定  

Camera calibration based on the circular pattern and ellipse fitting

作　　者：李龙[1] 何明一[1] 李娜[2] 

机构地区：[1]西北工业大学电子信息学院,陕西西安710129 [2]西安电子科技大学通信工程学院,陕西西安710071

出　　处：《西安电子科技大学学报》2010年第6期1148-1154,共7页Journal of Xidian University

基　　金：国家自然科学基金资助项目(60736007)

摘　　要：结合平面模板和一维标定物摄像机标定方法的优点,提出了一种新的基于椭圆拟合的圆环模板摄像机标定方法.该方法设计了一种新的圆环模板,只需要3副模板图像即可标定摄像机;通过引入椭圆拟合精确定位圆环模板特征点,利用非线性优化进一步提高了摄像机的标定精度.仿真实验结果表明,该方法精确度高、鲁棒性强,通过在真实图像上的实验,验证了该算法的性能.A novel camera calibration method hased on the circular pattern and ellipse fitting is proposed, which combines the advantages of camera calibration based on the plantar pattern and one dimensional object. The proposed method designs a new circular pattern, which needs three images of the pattern to calibrate the camera. Ellipse fitting is introduced to locate the feature points in the circular pattern precisely and nonlinear optimization is utilized, and thus the calibration precision has been improved significantly. Experimental results for synthetic data show that the proposed strategy is both accurate and robust. Experimental results for real images prove the performance of the proposed algorithm.

关 键 词：摄像机标定 圆环模板 椭圆拟合 非线性优化 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]