TOWARDS ACCURATE INSTANCE SEGMENTATION IN LARGE-SCALE LIDAR POINT CLOUDS

Grain size analysis is the key to understanding the sediment dynamics of river systems and is an important indicator for mitigating flood risk and preserving biodiversity in aquatic habitats. We propose GRAINet, a data-driven approach based on deep learning, to regress grain size distributions from georeferenced UAV images. This allows for a holistic analysis of entire gravel bars, resulting in robust grading curves and high-resolution maps of spatial grain size distribution at large scale.

In spring 2018 in the alpine Dischma valley (Switzerland), we tested different industrial photogrammetric platforms for snow depth mapping. These platforms were high-resolution satellites, an airplane, unmanned aerial systems and a terrestrial system. Therefore, this study gives a general overview of the accuracy and precision of the different photogrammetric platforms available in space and on earth and their use for snow depth mapping.

Forest management shapes forest structure and in turn surface–atmosphere interactions. We used Fennoscandian forest maps and inventory data to develop a classification system for forest structure. The classification was integrated with the ESA Climate Change Initiative land cover map to achieve complete surface representation. The result is an improved product for modeling surface–atmosphere exchanges in regions with intensively managed forests.