In this small project, we leverage open data to extract feature maps for Switzerland. We rely on OpenStreetMap data to extract topographical features at a 100x100 meters resolution. This includes building footprints, roads, railways, water bodies, and forests. The output is stored as GeoTIFF, each feature as a separate channel, using LV95 as coordinate system (c.f. Swiss coordinate system).
Roads (in red), buildings (in green), and railways (in blue) of Switzerland. The image is post-processed for better visualization, see ./notebooks/extract_png.ipynb. Please refer to the Releases section regarding the license.
If you use Conda, an environment.yml is provided:
conda env create --file environment.yml
Alternatively, you can install requirements.txt using Pip. The Cairo library must be installed separately, typically using a package manager; here is an example on Ubuntu:
sudo apt-get install libcairo2
pip install -r requirements.txt
In order to download an extract of OpenStreetMap data for Switzerland, we rely on Geofabrik GmbH dumps, which are updated daily. To proceed, download switzerland-latest-free.shp.zip to ./data/:
curl -o ./data/switzerland-latest-free.shp.zip https://download.geofabrik.de/europe/switzerland-latest-free.shp.zip
The recommended way to process this input file is to open and run ./notebooks/generate_geotiff.ipynb using Jupyter, which provides an interactive session:
jupyter notebook
Alternatively, you can use papermill to run it from the terminal:
papermill --cwd ./notebooks/ ./notebooks/generate_geotiff.ipynb ./notebooks/generate_geotiff.out.ipynb
While Geofabrik uploads dumps daily, we process a new version every 6 months, as the output is unlikely to change significantly. The provided switzerland.tif is released under the Open Data Commons Open Database License (ODbL), as per OpenStreetMap guidelines.
The code used to process the data is released under the MIT license.