Using Python to Build a GIS Data Pipeline for Rural-Urban Classification
summary
Python pipeline downloads map tiles, stitches and clips them, thins road rasters and polygonizes to vectors using GDAL and image thinning with Pillow, then extracts intersections, end‑points and segment lengths with Shapely and Fiona. Road‑network features feed a supervised classifier that matches the prior satellite approach (satellite gave ~96–97% accuracy) while avoiding cloud artefacts and multi‑million‑dollar yearly imagery costs; demo detects intersections and segment lengths correctly and the pipeline runs end‑to‑end.
AI-written summary · about
chapters (7)
- 0:00Introduction and context
- 2:52Defining the research problem
- 9:06Limitations of satellite data
- 12:56Road networks as a proxy
- 16:23Geospatial data fundamentals
- 22:31The GIS data pipeline
- 29:12Demonstration and conclusion
related talks
47mpython at scale – creating a High Definition Map of the world's road network
36mClassification of Satellite Imagery
28mAutomate Your City Data With Python
20mOpen-source processing pipelines for Cyclone Gabrielle
34mGeoscience Big Data Computing with Python
13mYirgacheffe: a declarative approach to geospatial data
11mTerraKit: Standardising AI-Ready Geospatial Data...
05mRegional Geological Mapping and data collection with the use of Remote Sensing and