引用本文: 王建, 姚吉利, 赵雪莹, 赵猛, 杨承昆, 李彩林. 点线混合控制基元的点云地理化参数联合解算[J]. 武汉大学学报 ( 信息科学版), 2020, 45(5): 760-767.
WANG Jian, YAO Jili, ZHAO Xueying, ZHAO Meng, YANG Chengkun, LI Cailin. Joint Solution of Point Cloud's Georeferencing Parameters Based on Points and Line Mixed Primitives[J]. Geomatics and Information Science of Wuhan University, 2020, 45(5): 760-767.
 Citation: WANG Jian, YAO Jili, ZHAO Xueying, ZHAO Meng, YANG Chengkun, LI Cailin. Joint Solution of Point Cloud's Georeferencing Parameters Based on Points and Line Mixed Primitives[J]. Geomatics and Information Science of Wuhan University, 2020, 45(5): 760-767.

## Joint Solution of Point Cloud's Georeferencing Parameters Based on Points and Line Mixed Primitives

• 摘要: 当今数字时代，将不同扫描站点云的空间基准统一到地理坐标系是多源数据融合的基础。目前，以最少的工作量获取高精度的地理化点云是点云地理化中重点关注的问题。在控制基元最少的情况下，尚无基于混合控制基元的地理化参数联合直接解法。利用点基元与点云数据中广泛存在的线基元相结合，提出基于2个控制点和1条控制线的混合控制基元点云地理化参数联合解法，借鉴点的三维坐标转换公式，根据反对称矩阵和罗德里格矩阵的性质，推导2点1线混合控制基元的地理化方程，从而解算出旋转矩阵及地理化参数，实现点云粗地理化。实验表明，该方法的平面点位中误差为11.20 mm，空间点位中误差为16.66 mm，与传统控制点地理化方法的精度相当，同时地理化工作量大大减少，可为精确的地理化平差计算提供可靠的初值。

Abstract: In the current digital age, it is the basis for multi-source data fusion that unifying the spatial reference of different scanning point clouds to the geographic coordinate system. A key issue in georeferencing is to obtain high-precision georeferenced point clouds with least efforts. In the case of minimum control primitives, there are not joint solutions of georeferencing parameters based on mixed control primitives. Line primitives are widely present in point clouds, so a joint solution of georeferencing parameters based on two control points and one control line is presented. According to the translation model of 3D point, the properties of antisymmetric matrix and Rodrigues matrix, the translation model of two control points and one control line is deduced. Furthermore, the rotation matrix and georeferencing parameters of single station point cloud georeferencing is calculated, and then the coarse georeferencing is achieved. Experiments show that our proposed method has high efficiency and less efforts. The root mean square error (RMSE) of plane position is 11.20 mm, and the RMSE of place position is16.66 mm. Our proposed method can be used as the way to obtain the initial value of georeferencing parameters in precise georeferencing.

/

• 分享
• 用微信扫码二维码

分享至好友和朋友圈