Teacher: Maria Brovelli (maria.brovelli@polimi.it)
Tutors: Daniele Oxoli (daniele.oxoli@polimi.it); Alberto Vavassori (alberto.vavassori@polimi.it)
Keyword: Sentinel-2, Local Climate Zones, spectral unmixing
Technologies: Python, Jupyter Notebook, QGIS
Project description: The first objective of the project is to develop a Python pipeline for multi-temporal Local Climate Zone (LCZ) mapping using Sentinel-2 multispectral imagery and ancillary urban canopy parameters. This involves the collection of training and testing samples, LCZ classification through machine-learning methods, and accuracy assessment. The second objective is to use Sentinel-2 data to derive multi-temporal surface material fractions (e.g., asphalt, concrete, and vegetation) through spectral unmixing techniques. The project output will be used for Land Surface Temperature modelling and simulation analysis.