Generating Aerial Flood Prediction Imagery with Generative Adversarial Networks

Natasha Randall, Gernot Heisenberg, Juan Luis Ramirez Duval

2025

Abstract

Floods are one of the most dangerous, impactful natural disasters, and flood forecasting is a critical component of effective pre-flooding preparedness. In this paper a data-driven approach to flood forecasting is presented, which provides photorealistic predictions that are less computationally expensive to generate than traditional physically-based models. A ‘PairedAttention’ generative adversarial network (GAN) was developed, that combines attention and content mask subnetworks, and was trained on paired sets of pre- and post-flooding aerial satellite images aligned with topographical data. The PairedAttention GAN achieved 88% accuracy and an F1 score of 0.8 at flood predictions on three USA flood events, and an ablation study determined that the digital elevation model was the most significant factor to improving the GAN’s performance. Although the model is a successful proof-of-concept for the effectiveness of a data-driven GAN to generate photorealistic, accurate aerial flood prediction imagery, it nevertheless struggled with generalisation, indicating an important avenue for future research.

Paper Citation

in Harvard Style

Randall N., Heisenberg G. and Duval J. (2025). Generating Aerial Flood Prediction Imagery with Generative Adversarial Networks. In Proceedings of the 17th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management - Volume 1: KDIR; ISBN , SciTePress, pages 15-27. DOI: 10.5220/0013663400004000

in Bibtex Style

@conference{kdir25,
author={Natasha Randall and Gernot Heisenberg and Juan Duval},
title={Generating Aerial Flood Prediction Imagery with Generative Adversarial Networks},
booktitle={Proceedings of the 17th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management - Volume 1: KDIR},
year={2025},
pages={15-27},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013663400004000},
isbn={},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 17th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management - Volume 1: KDIR
TI - Generating Aerial Flood Prediction Imagery with Generative Adversarial Networks
SN -
AU - Randall N.
AU - Heisenberg G.
AU - Duval J.
PY - 2025
SP - 15
EP - 27
DO - 10.5220/0013663400004000
PB - SciTePress