Landcover monitoring with emphasis on vegetation under the climatic change pressure using multitemporal and multisource remote sensing data fusion
News
Research news can be found in our GIScience News Blog.
Motivation
Climate change develops a pressure on natural ecosystems. Ecosystem continuous monitoring is, thus, the key for understanding their components and interrelations. Moreover, it indicates the intensity of climate and environmental changes and success rate of protective steps towards sustainable development and conservation of biodiversity. Rapid developments in remote sensing (RS) in the last decades have enabled the Earth observation in vast range of spatial, spectral and time scales. The era of “big data” and progress in artificial intelligence bring new possibilities and challenges in extracting information from RS data.
Fig. 1: Fusion of image and 3D point cloud Earth observation data for vegetation monitoring (cf. https://doi.org/10.1109/MGRS.2018.2890023)
Objective
This collaboration project is funded in the framework of 4EU+ Flagship 4: Biodiversity and Sustainable Developement with Markéta Potůčková (Department of Applied Geoinformatics and Cartography, Charles University Prague) as PI of the project and Heidelberg University, University of Copenhagen and University of Warsaw as project partners.
Joint research activities are developed towards i) exploitation and development of methodologies for multisource (multi-and hyperspectral, LiDAR) and multitemporal data fusion on different platforms (spaceborne, airborne, unmanned aircraft systems), ii) using multisource and multitemporal data for studying dynamics of selected ecosystems sensitive to climate and environmental changes (tundra, temperate and flood-plain forests) on different spatial scales (regional and plot-based), iii) retrieval of quantitative vegetation parameters from multisource and multitemporal data.
The 3DGeo research group supports this project within the 4EU+ Programme with methods and infrastructure to capture and process 3D/4D geospatial point clouds and methods for machine/deep learning for 3D/4D geodata analysis. Additionally, methods for data fusion of 3D/4D point clouds with other remote sensing data streams and methods to perform virtual laser scanning of vegetation in simulation environments are provided.
