Deep Learning-Driven Disaster and Aerial Image Segmentation using U-Net and Multi-U-Net Architectures

DOI

10.2991/978-94-6463-718-2_35How to use a DOI?

Keywords

U-Net; MultiUNet; Flood Segmentation; Landslide Detection; Aerial Image Segmentation; Urban Infrastructure; Deep Learning; Disaster Response; Remote Sensing

Abstract

Flooding and landslides are significant destructive natural events to infrastructure and human life. Segmentation of affected disaster areas in satellite and aerial images has an important role in an effective response and mitigation of disasters. This paper suggests a deep learning approach, specifically U-Net and MultiUNet architectures, for the segmentation of images concerning floods, landslides, and aerial views of urban infrastructure. For disaster-affected segmentations such as floods and landslides, U-Net was employed. For the complexity of multi-class segmentation in aerial imagery containing features such as roads, buildings, and vegetation, a MultiUNet architecture was developed for this multi-class segmentation complexity. Our models are trained and tested on publicly available datasets of flood-affected regions and landslide-prone areas and on aerial images from urban regions. The results of the proposed approach present high accuracy in disaster regions and urban infrastructure components segmentation. Thus, such a methodology may be used as a post-disaster assessment and damage estimation tool for infrastructure on-site.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - K. Harish
AU  - Harish Abinav
AU  - Dinesh Kumar
PY  - 2025
DA  - 2025/05/23
TI  - Deep Learning-Driven Disaster and Aerial Image Segmentation using U-Net and Multi-U-Net Architectures
BT  - Proceedings of the International Conference on Sustainability Innovation in Computing and Engineering (ICSICE 2024)
PB  - Atlantis Press
SP  - 401
EP  - 409
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6463-718-2_35
DO  - 10.2991/978-94-6463-718-2_35
ID  - Harish2025
ER  -