Search

The determination of a fully populated variance-covariance matrix (VCM) of a point cloud acquired with a terrestrial laser scanner (TLS) is not straightforward as physical correlations between TLS observations are not sufficiently known. Ignoring correlations leads to biased estimates, wrong modeling of surfaces, or the misinterpretation of geometric changes. For this reason, this study aims at progressing the empirical determination of correlations in TLS range observations by investigating short-scale correlations (mm to cm) of TLS rangefinders. Therefore, two methods are proposed to determine spatial correlations between TLS range observations. Both methods are evaluated for simulated laser scans, and then applied to empirical data of four different terrestrial laser scanners. The results demonstrate that short-scale correlations exist, and it is possible to quantify the relevant parameters to fill the VCM. It also shows that the correlation differs between the scanners and with the distance. Furthermore, we analyze the change of the correlation length with changing scanning configurations such as distance, incidence angle, and reflectivity with the result that this must be considered while applying the aforementioned methods to point clouds of large objects. All in all, this study trustfully quantifies physical short-scale correlations yielding from the rangefinder and suggests a guideline to apply this to point clouds of large objects. [ABSTRACT FROM AUTHOR]

Terrestrial laser scanner (TLS) measurements are unavoidably affected by systematic influences due to internal misalignments. The magnitude of the resulting errors can exceed the magnitude of random errors significantly deteriorating the quality of the obtained point clouds. Hence, the task of calibrating TLSs is important for applications with high demands regarding accuracy. In recent years, multiple in-situ self-calibration approaches were derived allowing the successful estimation of up-to-date calibration parameters. These approaches rely either on using manually placed targets or on using man-made geometric objects found in surroundings. Herein, we widen the existing toolbox with an alternative approach for panoramic TLSs, for the cases where such prerequisites cannot be met. We build upon the existing target-based two-face calibration method by substituting targets with precisely localized 2D keypoints, i.e. local features, detected in panoramic intensity images using the Förstner operator. To overcome the detriment of the perspective change on the feature localization accuracy, we estimate the majority of the relevant calibration parameters from a single station. The approach is verified on real data obtained with the Leica ScanStation P20. The obtained results were tested against the affirmed target-based two-face self-calibration. Analysis proved that the estimated calibration parameters are directly comparable both in the terms of parameter precision and correlation. In the end, we employ an effective evaluation procedure for testing the impact of the calibration results on the point cloud quality. [ABSTRACT FROM AUTHOR]

Laser scanning allows for objective area-based analysis of water dam structures to enable targeted interventions in case of displacements, requiring comparison of the same areas over different epochs. This comparison improves if identical areas, e.g. stones, from multiple epochs can be automatically derived in advance. We present an instance segmentation algorithm based on a 3D point cloud with reflected intensity information to identify individual stones in rubble masonry dams that can be used within deformation monitoring. The algorithm uses initial k-means classification followed by image-based processing steps, resulting in a segmented 3D point cloud with individual stones. It is evaluated against a manually labeled reference and analyzed on terrestrial laser scanning (TLS) and UAV-based data sets. Correctly identified stones are 89.92% for TLS and 90.82% for UAV data. Additionally, stone centroids and shapes were evaluated without significant deviation. A first outlook on deformation analysis was provided using ICP matching of stones from two simulated epochs. • Improvement of deformation monitoring with point clouds by detecting stone objects. • Novel approach for instance segmentation in rubble masonry. • Evaluation based on reference data sets from UAV-based laser scanning and TLS. • Demonstration of the potential in deformation analysis using ICP matching. [ABSTRACT FROM AUTHOR]