This repository contains the code for detecting road upgrades using publicly available satellite imagery (Sentinel-2, MODIS) to detect road upgrades. Primarily using the Google Earth Engine Python API. The contained code was developed as part of Jeffrey Clark and Tamina Matti's master's thesis at Stockholm School of Economics. The models were trained on a dataset of roads provided by researchers at the World Bank. While the exact coordinates to these roads are not shared, all other data (including images and gifs) is available. For copies of files with data (multiple band value changes for over a million of points) you can replicate our data collection or contact us. We hope to provide a detailed guide on how to run all code. Edits are ongoing.
Link to thesis: COMING SOON
GIFs of detected road upgrades
