AUTOMATIC URBANISATION MONITORING FOR RISK ASSESSMENT BY REMOTE SENSING AND COPERNICUS DATA – A PRELIMINARY RESEARCH

• Published in: International Multidisciplinary Scientific GeoConference SGEM Vol. 4; no. 2; pp. 245 - 252
• Main Authors: Gašparovié, Iva, Gašparovié, Mateo, Radié, Filip, Uroš, Mario
• Format: Conference Proceeding
• Language: English
• Published: Sofia Surveying Geology & Mining Ecology Management (SGEM) 01.07.2024
• Subjects: Accuracy; Algorithms; Automation; Building construction; Buildings; Construction; Data acquisition; Disasters; Earth; Earth observations (from space); Earthquakes; Emergency preparedness; Environmental risk; Growth patterns; Image processing; Land use; Landsat; Machine learning; Monitoring; Remote sensing; Risk assessment; Satellite data; Satellite observation; Satellites; Seismic engineering; Seismic hazard; Seismic surveys; Social resilience; Spatial data; Structural stability; Topographic databases; Urban areas; Urban planning; Urban sprawl; Urbanization; Variables; Vegetation; Croatia
• ISSN: 1314-2704
• DOI: 10.5593/sgem2024v/4.2/s19.33

Rapid urbanization increases the vulnerability of cities to natural hazards, especially earthquakes, as unplanned growth can aggravate structural risks and strain infrastructure. This research is based on a project "Automatic urbanization monitoring for risk assessment by remote sensing and Copernicus data" which aims to improve the detection of urban growth patterns and identify areas with increased seismic vulnerability. This project's main goal is to develop and test a prototype for automatic urbanization monitoring for risk assessment aided by remote sensing and Copernicus data for fast and accurate data acquisition and to provide improved risk assessment for the study sites in Croatia. The development of an automatic system for urbanization monitoring for risk assessment will enable the acquisition of accurate and current spatial and attribute data of buildings, such as building construction year. Using advanced image processing and machine learning techniques, the system analyses Earth observation satellite data to map urban extent, assess changes in land use, and identify critical areas where rapid growth may affect structural stability. This preliminary research demonstrates an algorithm for building construction year detection from Earth observation (EO) data. This research utilized Sentinel-2 imagery to extract building construction years for Tresnjevka sjever in Zagreb, Croatia. The preliminary results are promising, demonstrating that Earth Observation (EO) data, specifically Sentinel-2, can effectively assess building construction year. This approach is adaptable to other locations worldwide and, as EO data, other satellite missions can be used like Landsat, PlanetScope, etc. The automated method offers valuable insights for urban planners and policymakers, supporting proactive disaster preparedness and enhancing urban resilience.