Résumé : |
(Auteur) For several years now, terrestrial laser scanning has become an additional surveying technique in geodesy. Recent developments have improved several aspects of terrestrial laser scanners, e.g. the data acquisition rate, accuracy, and range. Since such instruments are relatively new and constructed by manufacturers who do not have advanced experience in surveying instruments, investigations are needed to assess the quality of the instrumental characteristics and the acquired data. In this way, manufacturers will understand the needs of geodesists and in turn enable geodesists to provide the necessary support in the development of improvements. This thesis has three objectives, the calibration and investigation of a terrestrial laser scanner, the post-processing of point clouds acquired by laser scanners, and applications of terrestrial laser scanning.
The first objective is a comprehensive calibration and investigation of a specific laser scanner, the Imager 5003 of Zoller+Frohlich GmbH (Germany). The investigation and calibration procedures shall give a general impulse for all users of terrestrial laser scanning regarding instrumental and non-instrumental errors, the assessment of the quality of distance and angle measurements, and the influencing parameters. Laser scanners are a black box instrument that produces a huge number of 3D points in the form of a point cloud in a short time. However, it is the surveyor, who has to assess the reliability and quality of the resulting data. Therefore, the potential and the limitations of laser scanner systems must be identified. This is particularly important when a distance measurement is influenced by several parameters that can bias the data. Since laser scanning is an active surveying method, mostly independent of lighting conditions, distance measurements do not require prisms. Thus, surveying of almost every object is conceivable.
The second objective involves post-processing of the point clouds. Terrestrial laser scanning consists not only of data acquisition, but also processing of the acquired 3D data, which include an intensity value of the reflected laser beam. The point clouds define the objects and the data contains nearly all the information about the objects due to the high sampling interval of laser scanners. To produce the final result, data processing needs to be completed and this can be quiet involving, e.g. registration, data filtering, noise reduction, triangulation, and modeling. The ratio between post-processing and data acquisition can be 10:1 or greater, which means ten (or more) days of post-processing follow one day of data acquisition. This aspect of post-processing applies for both static laser scanning and kinematic laser scanning. The only difference is that kinematic laser scanning requires an unique method of registration and geo-referencing.
The third objective examines the applications of terrestrial laser scanning. Laser scanning can be used in different fields of applications, e.g. industrial metrology, cultural heritage, reverse engineering, and engineering geodesy. Due to the increased requirements regarding accuracy engineering geodesy appears to be a challenging field. Therefore, three different applications are presented which verify the successful use of terrestrial laser scanning in engineering geodesy. The first application involves the field of urban water management. A road surface was scanned to derive catchment areas and water flow directions. The second application covers the field of engineering geology. A tunnel during and after excavation was scanned to characterize rock mass structures and to derive displacement maps of surfaces and object points. Since the first two applications are based on static laser scanning, which means the laser scanner did not change in position and orientation during scanning, the third application is a kinematic one, which means the laser scanner was in motion during scanning. Such kinematic applications are of great interest since the performance of laser scanning can be increased significantly. Tunnels and roads are especially appropriate for kinematic laser scanning. The potential of kinematic laser scanning is tested by moving the laser scanner along a track line. The quality is assessed by scanning reference points. |