Advancing Earth Observation Through Machine Learning: A TorchGeo Tutorial
This addresses the challenge of adapting standard computer vision workflows to Earth observation tasks for researchers and practitioners, but it is incremental as it focuses on tutorial dissemination rather than novel research.
The paper introduces a tutorial for TorchGeo, a PyTorch library designed to handle geospatial data in machine learning pipelines, demonstrating its use through code examples and an end-to-end case study on multispectral water segmentation from Sentinel-2 imagery, achieving practical results for geospatial analysis.
Earth observation machine learning pipelines differ fundamentally from standard computer vision workflows. Imagery is typically delivered as large, georeferenced scenes, labels may be raster masks or vector geometries in distinct coordinate reference systems, and both training and evaluation often require spatially aware sampling and splitting strategies. TorchGeo is a PyTorch-based domain library that provides datasets, samplers, transforms and pre-trained models with the goal of making it easy to use geospatial data in machine learning pipelines. In this paper, we introduce a tutorial that demonstrates 1.) the core TorchGeo abstractions through code examples, and 2.) an end-to-end case study on multispectral water segmentation from Sentinel-2 imagery using the Earth Surface Water dataset. This demonstrates how to train a semantic segmentation model using TorchGeo datasets, apply the model to a Sentinel-2 scene over Rio de Janeiro, Brazil, and save the resulting predictions as a GeoTIFF for further geospatial analysis. The tutorial code itself is distributed as two Python notebooks: https://torchgeo.readthedocs.io/en/stable/tutorials/torchgeo.html and https://torchgeo.readthedocs.io/en/stable/tutorials/earth_surface_water.html.