当前位置: X-MOL 学术J. Mod. Optic. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Geometric calibration method based on Euler transformation for a large field of view polarimetric imager
Journal of Modern Optics ( IF 1.2 ) Pub Date : 2020-11-27 , DOI: 10.1080/09500340.2020.1871521
Chan Huang 1, 2, 3 , Yuyang Chang 1, 2, 3 , Lin Han 1, 3 , Su Wu 1, 2 , Shuang Li 1, 3 , Donggen Luo 1, 3 , Liang Sun 1, 3 , Jin Hong 1, 3
Affiliation  

ABSTRACT A geometric calibration method based on a two-dimensional turntable and Euler transformation with simple operation, high precision and wide application is proposed in this paper. Instead of precisely adjusting the relative position of the sensor and the turntable, the method introduces rotation matrix and translation vector to effectively eliminate system error and alignment error. The geometric calibration verification experiments of the directional polarimetric camera were carried out in the laboratory with the original method without Euler transformation and the proposed method. The experimental results show the calibration residual of the proposed method is less than 0.2 pixel while the maximum residual of the original method reaches about 0.35 pixel, and the root mean squared error and mean reprojection error are also improved compared with the original method. These facts prove the availability and superiority of the proposed method.

中文翻译:

基于欧拉变​​换的大视场偏振成像仪几何标定方法

摘要 提出了一种操作简单、精度高、应用广泛的基于二维转台和欧拉变换的几何标定方法。该方法不是精确调整传感器与转台的相对位置,而是引入旋转矩阵和平移向量,有效消除系统误差和对准误差。定向偏振相机的几何标定验证实验在实验室进行,采用原无欧拉变换的方法和提出的方法。实验结果表明,该方法的校准残差小于0.2像素,而原方法的最大残差达到约0.35像素,均方根误差和平均重投影误差也较原方法有所改善。这些事实证明了所提出方法的可用性和优越性。
更新日期:2020-11-27
down
wechat
bug