B03
Deep learning for satellite-based land use and land cover reconstruction
Prof. Dr. Ribana Roscher
University of Bonn | +49 228 73-60854 |
Summary
The goal of this project is the reconstruction of land use and land cover from optical satellite data using deep learning. To this end, we will develop spatio-temporal deep neural networks that consider the specific biogeographical characteristics of the regions of interest in order to ensure a high generalization capability across the study region. Furthermore, predictive uncertainties for the derived land use and land cover maps will be determined and the model’s capabilities in the context of multi-task learning will be studied. Lastly, we will also explore the potential of generated data to fill spatial and temporal data gaps.
Graphical Summary
Contribution to CRC
Land use and land cover maps generated in our project will be reused by other projects in DETECT for a variety of purposes: On the one hand, the A and B clusters will rely on these data to determine relevant regions for their own purposes, such as agricultural or water areas. The C and D clusters, on the other hand, will use the maps as inputs for terrestrial models or for subsequent analysis.
Approach
To train our model we will rely on openly available medium resolution, multispectral imagery from Sentinel-2 and the Landsat satellites. Observations from the LUCAS surveys will be used as reference data. As those data are distributed rather sparsely, special methods such as Neural Networks operating on images’ region adjacency graphs need be used. The above-described research ideas shall be implemented as amendments or extensions to this general model.
Main results in 2022
Conceptualization and start of implementation of the general model and approaches for improving its generalizability.
