K-means clustering of building displacements from PS InSAR data to discover patterns across geological formations and classify terrain types.
This repository contains the analysis of building displacements derived from Persistent Scatterer Interferometric Synthetic Aperture Radar (PS InSAR) data. The project aims to discover displacement patterns across geological formations and infer terrain behavior (e.g., soft soils vs. rock masses) using unsupervised machine learning techniques.
- Analyze PS InSAR-derived building displacement data in two Italian cities: Rieti and Ferrara.
- Hypothesis Testing for deformation rates across different geological types.
- Investigate the relationship between displacement patterns and underlying geological formations.
- Apply K-means clustering to identify patterns in building displacements and to recognize terrain behavior based on displacement characteristics.
Rieti, located in central Italy, offers a diverse geological setting with both soft soil areas and rock masses. This case study allows us to examine how building displacements vary across different geological contexts within a single urban area.
Ferrara, situated in the Po Valley of northern Italy, presents a contrasting geological environment predominantly characterized by soft soils. This case study provides insights into building displacement patterns in a more homogeneous geological setting.
This project is licensed under the MIT License - see the LICENSE.md file for details.
Giandomenico Mastrantoni - giandomenico.mastrantoni@uniroma1.it
Project Link: https://github.com/gmastrantoni/Geo-Clustering-Building-Displacements