geo-scraping

Scraping open source geospatial data resources for satellite imagery and vector labeling.

Long-term goal to create benchmark dataset consisting of high-resolution satellite imagery with hierarchical labels on the scale of ImageNet!

Documents

OSM can provide vector labels, but a global uniform image dataset does not exist at resolutions higher than 30m/px.

Need to create a pipeline to take a Lat/Lon bounding box and export all the contained classes as so:

Name		Name	Last commit message	Last commit date
Latest commit History 1 Commit
__pycache__		__pycache__
lib		lib
testing_nb		testing_nb
LibTest.ipynb		LibTest.ipynb
README.md		README.md
cli.py		cli.py
config.py		config.py
geoscraping.ipynb		geoscraping.ipynb
googlemaps.ipynb		googlemaps.ipynb
label_depth.ipynb		label_depth.ipynb
overpass_api.ipynb		overpass_api.ipynb
pipe_script.py		pipe_script.py
radiant_dev.ipynb		radiant_dev.ipynb
sample_config.py		sample_config.py
tool_demos.ipynb		tool_demos.ipynb
view_aus_set.ipynb		view_aus_set.ipynb
view_tusc_set.ipynb		view_tusc_set.ipynb