Automatic mapping of national surface water with OpenStreetMap and Sentinel-2 MSI data using deep learning

Large-scale mapping activities can benefit from the vastly increasing availability of earth observation (EO) data, especially when combined with volunteered geographical information (VGI) using machine learning (ML). High-resolution maps of inland surface water bodies are important for water supply and natural disaster mitigation as well as for monitoring, managing, and preserving landscapes and ecosystems.

In a recent published paper, we propose an automatic surface water mapping workflow by training a deep residual neural network (ResNet) based on OpenStreetMap (OSM) data and Sentinel-2 multispectral data, where the Simple Non-Iterative Clustering (SNIC) superpixel algorithm was employed for generating object-based training samples. As a case study, we produced an open surface water layer for Germany using a national ResNet model at a 10 m spatial resolution, which was then harmonized with OSM data for final surface water products. Moreover, we evaluated the mapping accuracy of our open water products via conducting expert validation campaigns, and comparing to existing water products, namely the WasserBLIcK and Global Surface Water Layer (GSWL). Using 4,600 validation samples in Germany, the proposed model (ResNet+SNIC) achieved an overall accuracy of 86.32% and competitive detection rates over the WasserBLIcK (87.47%) and GSWL (98.61%). This study provides comprehensive insights into how to best explore the synergy of VGI and ML of EO data in a large-scale surface water mapping task.

Fig: The surface water maps of four selected study areas. Left: The distribution of validation pixels and the selected study areas; Right: Predicted surface water maps using SNIC+ResNet with 10 band MSI and Random+ResNet with 6 band MSI together with the Sentinel-2 False color images.

In this recent paper published in International Journal of Applied Earth Observation and Geoinformation, we aim to emphasize the potential synergy of VGI and ML of EO data with the insights shared and lessons learned towards facilitating future large-scale and up-to-date mapping applications.

Li, H.; Zech, J.; Ludwig, C.; Fendrich, S.; Shapiro, A.; Schultz, M.; Zipf, A.(2021) Automatic mapping of national surface water with OpenStreetMap and Sentinel-2 MSI data using deep learning. International Journal of Applied Earth Observation and Geoinformation. https://doi.org/10.1016/j.jag.2021.102571

Online service and tool developed by GIScience/HeiGIT:

Previous related work at GIScience/HeiGIT (selection):