Integrating Multi-Source Geospatial Data and Evaluating Model Architecture for Deep Learning-Based Flood Inundation Mapping
Remote Sensing · 02.11.2026 · 18 (21) · 3769
DOI: 10.3390/rs18213769 ↗Kurz gefasst
Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 02.11.2026. Verfasst von Abu Ubaida und 5 weiteren Autor:innen. Die Arbeit nennt 53 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Deep-learning research on flood inundation mapping is driven largely by architectural innovation and single-run reporting. We ask whether the segmentation architecture or the input representation contributes more to accuracy in flood-event surface-water mapping. We introduce GeoHybridNet, a hybrid CNN–Transformer network, and compare it on the hand-labeled Sen1Floods11 benchmark with a U-Net, an Attention U-Net, a parameter-matched wide U-Net, and SegFormer under matched data and training. The four convolutional and hybrid models differ by at most 1.4 IoU points, and none are significantly different under event-level tests; a compact SegFormer trained from scratch is about 3.5 points lower. In factorial designs crossing three architectures with up to five input representations (five seeds each), input representation explains about 95% of the variance in water IoU and architecture about 1%. The dominant input factor is radiometric normalization: raw bands standardized per chip reach only 60% IoU and adding NDWI restores 76%, whereas the same bands standardized with training-set statistics reach 77.5% without any index. The pattern holds when permanent water is excluded. On three held-out flood events, accuracy drops by 24–26 IoU points for every model. On this benchmark, input preparation and honest evaluation matter more than architecture.
Zitieren
APA
Ubaida, A., Haider, K. Z., Naqvi, S. A. A., Rehman, A., Aadil, F., & Khalid, S. (2026). Integrating Multi-Source Geospatial Data and Evaluating Model Architecture for Deep Learning-Based Flood Inundation Mapping. Remote Sensing, 18(21), 3769. https://doi.org/10.3390/rs18213769
Vancouver
Ubaida A, Haider KZ, Naqvi SAA, Rehman A, Aadil F, Khalid S. Integrating Multi-Source Geospatial Data and Evaluating Model Architecture for Deep Learning-Based Flood Inundation Mapping. Remote Sensing. 2026;18(21):3769. doi:10.3390/rs18213769
BibTeX
@article{Ubaida2026,
author = {Ubaida, Abu and Haider, Khurram Zeeshan and Naqvi, Syed Ali Asad and Rehman, Adnanul and Aadil, Farhan and Khalid, Shoaib},
title = {Integrating Multi-Source Geospatial Data and Evaluating Model Architecture for Deep Learning-Based Flood Inundation Mapping},
journal = {Remote Sensing},
year = {2026},
volume = {18},
number = {21},
pages = {3769},
doi = {10.3390/rs18213769}
}
EUVIMED