基于虚拟点模型的机载LiDAR系统自动检校方法

Automated Airborne LiDAR System Calibration Using Virtual Tie Point Model

提出一种全自动的LiDAR系统误差检校方法。针对传统检校方法大多依赖手工提取几何特征,自动化程度低且对于实际数据的适应性差等不足,尝试建立一种基于特征点对应的新型检校方法,以发挥点特征数量丰富易于提取的优势,提高检校算法的自动性和实用性。首先提出虚拟连接点模型,解决了离散点云间的同名点对应问题;其次利用LiDAR数据的强度信息,设计了同名点自动提取流程;最后采用真实数据,并与基于共面约束的检校方法进行对比试验,证明了本文方法的可行性和有效性。

An automated method of systematic bias self-calibration is presented for airborne LiDAR traditional calibration approaches relied on manual extraction of geometric features which made them lack adaptability for practical data. Anovel method is established based on conjugate point features extracted from overlapping LiDAR strips to take the advantage of point features in abundance and accessibility. Firstly, the virtual tie point model is proposed to solve the problem of point corresponding in discrete point cloud. Then, to employ intensify information of LiDAR data, an automatic workflow for corresponding point extraction is given. Finally, practical LiDARdata is used to test our method. Compared with coplanar constraint calibration method, the result illustrates the feasibility and effectiveness o; the proposed method.