Deep learning semantic segmentation of airborne LiDAR point clouds for urban flood inundation mapping

dc.contributor.authorHafidha BOUYERBOU
dc.contributor.authorZineb ABDERRAHMANE
dc.date.accessioned2026-10-17T19:57:29Z
dc.date.issued2026-11
dc.description.abstractAbstract : This study presents a deep learning approach based on airborne LiDAR data for urban flood mapping. Applied to Hurricane Harvey in Texas, the method uses the PointNet++ architecture to directly segment 3D point clouds, distinguishing flooded and non-flooded areas from urban structures and vegetation. The results highlight the critical role of ground-truth label quality in improving flood detection accuracy and demonstrate the potential of LiDAR-based deep learning for rapid post-disaster flood assessment. Résumé : Cette étude présente une approche de deep learning fondée sur les données LiDAR aéroportées pour la cartographie des inondations en milieu urbain. Appliquée à l’événement de l’ouragan Harvey au Texas, elle exploite l’architecture PointNet++ pour segmenter directement les nuages de points 3D et distinguer les zones inondées des zones non inondées ainsi que les principales structures urbaines. Les résultats soulignent l’importance de la précision des données de référence dans la fiabilité de la détection et démontrent le potentiel de cette méthode pour l’évaluation rapide des inondations après catastrophe.
dc.identifier.citationHafidha Bouyerbou, Zineb Abderrahmane, Deep learning semantic segmentation of airborne LiDAR point clouds for urban flood inundation mapping: a case study of Hurricane Harvey, Harris County, Texas, J. Appl. Remote Sens. 20(4), 046504 (2026), doi: 10.1117/1.JRS.20.046504
dc.identifier.issn046504-18
dc.identifier.urihttps://dspace.ensb.dz/handle/123456789/525
dc.language.isoen
dc.publisherJournal of Applied Remote Sensing
dc.subjectFlood Mapping
dc.subjectUrban Flooding
dc.subjectDeep Learning
dc.subjectRemote Sensing
dc.subject3D Point Cloud Segmentation
dc.subjectPointNet++
dc.subjectAirborne LiDAR
dc.subjectHurricane Harvey
dc.subjectFlood Inundation Detection.
dc.titleDeep learning semantic segmentation of airborne LiDAR point clouds for urban flood inundation mapping
dc.title.alternativea case study of Hurricane Harvey, Harris County, Texas
dc.typeArticle

Files

Original bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
Dr. BOUYERBOU.pdf
Size:
5.52 MB
Format:
Adobe Portable Document Format

License bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
license.txt
Size:
1.71 KB
Format:
Item-specific license agreed to upon submission
Description:

Collections