Mapping and characterising buildings for flood exposure analysis using open-source data and artificial intelligence

Kushanav Bhuyan (), Cees Westen, Jiong Wang and Sansar Raj Meena
Additional contact information
Kushanav Bhuyan: University of Padova
Cees Westen: University of Twente
Jiong Wang: University of Twente
Sansar Raj Meena: University of Padova

Natural Hazards: Journal of the International Society for the Prevention and Mitigation of Natural Hazards, 2023, vol. 119, issue 2, No 3, 805-835

Abstract: Abstract The mapping and characterisation of building footprints is a challenging task due to inaccessibility and incompleteness of the required data, thus hindering the estimation of loss caused by natural and anthropogenic hazards. Major advancements have been made in the collaborative mapping of buildings with platforms like OpenStreetMap, however, many parts of the world still lack this information or the information is outdated. We created a semi-automated workflow for the development of elements-at-risk (EaR) databases of buildings by detecting building footprints using deep learning and characterising the footprints with building occupancy information using building morphological metrics and open-source auxiliary data. The deep learning model was used to detect building EaR footprints in a city in Kerala (India) with an F1 score of over 76%. The footprints were classified into 13 building occupancy types along with information such as average number of floors, total floor space area, building density, and percentage of built-up area. We analysed the transferability of the approach to a different city in Kerala and obtained an almost similar F1 score of 74%. We also examined the exposure of the buildings and the associated occupancies to floods using the 2018 flood susceptibility map of the respective cities. We notice certain shortcomings in our research particularly, the need for a local expert and good quality auxiliary data to obtain reasonable building occupancy information, however, our research contributes to developing a rapid method for generating a building EaR database in data-scarce regions with attributes of occupancy types, thus supporting regional risk assessment, disaster risk mitigation, risk reduction initiatives, and policy developments.

Keywords: Deep learning; Building detection; Building morphology; Building characterisation; Open-source data; Exposure assessment (search for similar items in EconPapers)
Date: 2023
References: Add references at CitEc
Citations:

Downloads: (external link)
http://link.springer.com/10.1007/s11069-022-05612-4 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:nathaz:v:119:y:2023:i:2:d:10.1007_s11069-022-05612-4

Ordering information: This journal article can be ordered from
http://www.springer.com/economics/journal/11069

DOI: 10.1007/s11069-022-05612-4

Access Statistics for this article

Natural Hazards: Journal of the International Society for the Prevention and Mitigation of Natural Hazards is currently edited by Thomas Glade, Tad S. Murty and Vladimír Schenk

More articles in Natural Hazards: Journal of the International Society for the Prevention and Mitigation of Natural Hazards from Springer, International Society for the Prevention and Mitigation of Natural Hazards
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().