Abstract:

Late years, innovative close-range remote sensing technology such as Unmanned Aerial Vehicle (UAV) photogrammetry and Terrestrial Laser Scanning (TLS) are widely applied in the field of geoscience due to their efficiency in collecting data about surface morphology. Their main advantage stands on the fact that conventional methods are mainly collecting point measurements such as compass measurements of bedding and fracture orientation solely from accessible areas. The current research aims to demonstrate the applicability of UAVs in managing landslide and rockfall hazard in mountainous environments during emergency situations using object-based approach. Specifically, a detailed UAV survey took place in a test site namely as Proussos, one of the most visited and famous Monasteries in the territory of Evritania prefecture, in central Greece. An unstable steep slope across the sole road network results in continuous failures and road cuts after heavy rainfall events. Structure from Motion (SfM) photogrammetry is used to provide detailed 3D point clouds describing the surface morphology of landslide objects. The latter resulted from an object-based classification approach of the photogrammetric point cloud products into homogeneous and spatially connected elements. In specific, a knowledge-based ruleset has been developed in accordance with the local morphometric parameters. Orthomosaic and DSM were segmented in meaningful objects based on a number of geometrical and contextual properties and classified as a landslide object (scarp, depletion zone, accumulation zone). The resulted models were used to detect and characterize 3D landslide features and provide a hazard assessment in respect to the road network. Moreover, a detailed assessment of the identified failure mechanism has been provided. The proposed study presents the effectiveness and efficiency of UAV platforms to acquire accurate photogrammetric datasets from high-mountain environments and complex surface topographies and provide a holistic object-based framework to characterize the failure site based on semantic classification of the landslide objects.

https://www.int-arch-photogramm-remote-sens-spatial-inf-sci.net/XLII-2-W13/425/2019/