C9 – Automated land use mapping of urban dynamics

The continuous expansion of built-up areas remains one of the most significant land cover changes across Europe in recent decades. While urban sprawl is often associated with soil sealing, some artificial surfaces may also become underused or abandoned over time. This highlights the growing need for accurate and up-to-date land cover information.

In the context of EvoLand, an automated chain for detecting and characterizing urban changes was prototyped using optical Earth Observation data and machine learning techniques. The tool generates annual urban change maps at 5-meter resolution, thanks to the integration of deep learning-based super-resolution technique. This enables detailed and timely analysis of complex and heterogeneous urban dynamics.