Preamble¶
Tools: QGIS, VS Code, Python (conda), Jupyter Notebook
Data: Sentinel-2 sample images (provided during the session)
Goal: understand what makes Earth Observation (EO) imagery different from standard computer vision images and why CV is useful for large-scale environmental monitoring.
Learning Objectives¶
By the end of this session, students will be able to:
Explain what satellite imagery is and how it differs from everyday images
Describe the main characteristics of EO data (resolutions, spectral information, temporal dimension)
Understand the general Earth Observation processing pipeline (from sensor to decision)
Identify typical computer vision tasks applied to EO data
Recognize key challenges and limitations when applying AI to satellite imagery
💡 Key takeaway
In Earth Observation, data quality, physical meaning of pixels, and validation strategies are as important as the choice of the model.
Course Material¶
The lecture slides used during this session are available here👇:
📚 Course slides: 📥 Introduction to Computer Vision & Earth Observation