Vliv geometrie skenování na kvalitu lidarového bodového mračna pořízeného z UAV nosiče

Kolář, Michal

The use of UAV LiDAR allows for the acquisition of precise and detailed point clouds of the Earth's surface. Evaluating sources of uncertainty, such as direct georeferencing parameters (GNSS, IMU), atmospheric effects, data preprocessing, or scanning geometry, play a crucial role in determining the spatial accuracy of each individual point. With the vast amount of points collected and the variability of the scanned surface, it is important to consider what percentage of points accurately represent the Earth's surface, as well as understanding the noise fraction and how it varies with the scan geometry. The objective of this thesis is to identify noise in data captured from a rock glacier and to assess the impact of scanning geometry on the occurrence of noise. This study analyses a high-density ULS point cloud (1200 points/m2) acquired on the "Äußeres Hochebenkar" (AHK) stone glacier in the Ötztal Alps, Austria. The large volume of data (147 million points) poses a great challenge for its processing. To eliminate system errors, additional strip merging was first performed over a total of 29 flight strips. Subsequently, selected scan geometry attributes (scan distance, scan angle, incidence angle, slope, relative orientation of trajectory and slope exposure, etc.) were calculated. Five algorithms...

guest :: login Digital Repository
		Search		Submit		Help		About