Dálkový průzkum hydrologických procesů a říční dynamiky pomocí UAV

Lendzioch, Theodora

UAV REMOTE SENSING OF HYDROLOGICAL PROCESSES AND FLUVIAL DYNAMICS THEODORA LENDZIOCH Abstract Using drones and machine or deep learning algorithms (ML or DL) for environmental monitoring offers several advantages over traditional methods, including gathering high spatial resolution data quickly and non-invasively, acquiring real-time data, and covering large and remote areas. This dissertation focuses on snow cover, river granulometry, river sustainability, river bathymetry, and peatland dynamics based on approaches of drone-based imagery that are critical for understanding fluvial processes in mid-mountain regions and their implications for streamflow patterns and ecosystem health. Measuring Snow Depth (SD) and vegetation characteristics like Leaf Area Index (LAI) accurately is essential for effectively predicting snow cover and snowpack persistence across study sites (papers I and III). A further aspect of the fluvial process mediator involves the reproducibility of drone data. This allows for seamless coverage of riverbeds and the determination of ideal parameters for sediment surface cover detection. This can be done through photo-sieving or DL technique, which can analyze Particle Size Distributions (PSDs) of an entire river point bar from top-view UAV images (as described in papers II and V)....

guest :: login Digital Repository
		Search		Submit		Help		About