A Correspondence Framework for ALS Strip Adjustments based on Variants of the ICP Algorithm
A Correspondence Framework for ALS Strip Adjustments based on Variants of the ICP Algorithm
复制标题
DOI：
10.1127/pfg/2015/0270
复制
发表时间：
2015-01-01
影响因子：
--
通讯作者：
Ressl, Camillo
中科院分区：
文献类型：
--
作者：
Glira, Philipp;Pfeifer, Norbert;Ressl, Camillo
为了最小化机载激光扫描（ALS）条带重叠区域内的差异，可以执行条带调整。除转换模型外，带式平差的质量还受此过程中使用的观测值的强烈影响。为了利用数据的全分辨率，应在原始点云而不是插值曲面或栅格的基础上建立对应关系，从而可以避免精度损失和系统插值效应。迭代最近点（ICP）算法是一种基于原始点云的曲面匹配方法。在这项研究中，几个ICP变量适合大量的数据进行了研究。我们介绍了一种新的方法来选择对应，这是基于一个点上的调整计算的影响。作为本研究的结果，提出了变体的组合，形成了针对大多数ALS数据优化的基线。调查的变体提供了一个对应的框架ALS带调整。特定变体的益处在具有挑战性的ALS场景的基础上得到证明。
In order to minimize discrepancies within the overlap area of airborne laser scanning (ALS) strips, strip adjustment can be performed. Apart from the transformation model, the quality of strip adjustment is strongly affected by the observations used in this process. In order to exploit the full resolution of the data, correspondences should be established on the basis of the original point cloud instead of interpolated surfaces or rasters, so that a loss in accuracy and systematic interpolation effects can be avoided. A surface matching method in which correspondences are based on the original point cloud is the iterative closest point (ICP) algorithm. In this study several ICP variants suitable for large amounts of data are investigated. We introduce a new method for the selection of correspondences which is based on the influence of a point on the adjustment calculations. As a result of this study, a combination of variants, forming a baseline optimized for most ALS data, is presented. The investigated variants provide a correspondence framework for ALS strip adjustment. The benefit of specific variants is demonstrated on the basis of a challenging ALS scene.