Quality Evaluation of Sizeable Surveying-Industry-Produced Terrestrial Laser Scanning Point Clouds That Facilitate Building Information Modeling—A Case Study of Seven Point Clouds

Terrestrial laser scanning can provide high-quality, detailed point clouds, with state-of-the-art research reporting the potential for sub-centimeter accuracy. However, state-of-the-art research may not represent real-world practices reliably. This study aims to deliver a different perspective through collaboration with the surveying industry, where time constraints and productivity requirements limit the effort which can go to ensuring point cloud quality. Seven sizeable buildings’ point clouds (490 to 1392 scanning stations) are evaluated qualitatively and quantitatively. Quantitative evaluations based on independent total station control surveys indicate that sub-centimeter accuracy is achievable for smaller point cloud portions (e.g., a single building story) but caution against such optimism for sizable point clouds of large, multi-story buildings. The control surveys reveal common registration errors around the 5 cm range, resulting from complex surface geometries, as in stairways. Potentially hidden from visual inspection, such systematic errors can cause misalignments between point cloud portions in the compound point cloud structure, which could be detrimental to further applications of the point clouds. The study also evaluates point cloud georeferencing, affirming the resection method’s capability of providing high consistency and an accuracy of a few centimeters. Following the study’s findings, practical recommendations for terrestrial laser scanning surveys and data processing are formulated.

Dieses Werk wurde unter der Creative-Commons-Lizenz Namensnennung 4.0 International (CC-BY 4.0) veröffentlicht und darf unter den Lizenzbedinungen vervielfältigt, verbreitet, öffentlich zugänglich gemacht, sowie abgewandelt und bearbeitet werden. Dabei muss der Urheber bzw. Rechteinhaber genannt und die Lizenzbedingungen eingehalten werden.