Toward Reliable Post-Disaster Assessment: Advancing Building Damage Detection Using You Only Look Once Convolutional Neural Network and Satellite Imagery

Natural disasters continuously threaten populations worldwide, with hydrometeorological events standing out due to their unpredictability, rapid onset, and significant destructive capacity. However, developing countries often face severe budgetary constraints and rely heavily on international suppor...

Full description

Saved in:
Bibliographic Details
Main Authors: César Luis Moreno González, Germán A. Montoya, Carlos Lozano Garzón
Format: Article
Language:English
Published: MDPI AG 2025-03-01
Series:Mathematics
Subjects:
Online Access:https://www.mdpi.com/2227-7390/13/7/1041
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850213245104685056
author César Luis Moreno González
Germán A. Montoya
Carlos Lozano Garzón
author_facet César Luis Moreno González
Germán A. Montoya
Carlos Lozano Garzón
author_sort César Luis Moreno González
collection DOAJ
description Natural disasters continuously threaten populations worldwide, with hydrometeorological events standing out due to their unpredictability, rapid onset, and significant destructive capacity. However, developing countries often face severe budgetary constraints and rely heavily on international support, limiting their ability to implement optimal disaster response strategies. This study addresses these challenges by developing and implementing YOLOv8-based deep learning models trained on high-resolution satellite imagery from the Maxar GeoEye-1 satellite. Unlike prior studies, we introduce a manually labeled dataset, consisting of 1400 undamaged and 1200 damaged buildings, derived from pre- and post-Hurricane Maria imagery. This dataset has been publicly released, providing a benchmark for future disaster assessment research. Additionally, we conduct a systematic evaluation of optimization strategies, comparing SGD with momentum, RMSProp, Adam, AdaMax, NAdam, and AdamW. Our results demonstrate that SGD with momentum outperforms Adam-based optimizers in training stability, convergence speed, and reliability across higher confidence thresholds, leading to more robust and consistent disaster damage predictions. To enhance usability, we propose deploying the trained model via a REST API, enabling real-time damage assessment with minimal computational resources, making it a low-cost, scalable tool for government agencies and humanitarian organizations. These findings contribute to machine learning-based disaster response, offering an efficient, cost-effective framework for large-scale damage assessment and reinforcing the importance of model selection, hyperparameter tuning, and optimization functions in critical real-world applications.
format Article
id doaj-art-7e497641a37643b397e002e43aebe632
institution OA Journals
issn 2227-7390
language English
publishDate 2025-03-01
publisher MDPI AG
record_format Article
series Mathematics
spelling doaj-art-7e497641a37643b397e002e43aebe6322025-08-20T02:09:11ZengMDPI AGMathematics2227-73902025-03-01137104110.3390/math13071041Toward Reliable Post-Disaster Assessment: Advancing Building Damage Detection Using You Only Look Once Convolutional Neural Network and Satellite ImageryCésar Luis Moreno González0Germán A. Montoya1Carlos Lozano Garzón2Systems and Computing Engineering Department, Universidad de los Andes, Bogotá 111711, ColombiaSystems and Computing Engineering Department, Universidad de los Andes, Bogotá 111711, ColombiaSystems and Computing Engineering Department, Universidad de los Andes, Bogotá 111711, ColombiaNatural disasters continuously threaten populations worldwide, with hydrometeorological events standing out due to their unpredictability, rapid onset, and significant destructive capacity. However, developing countries often face severe budgetary constraints and rely heavily on international support, limiting their ability to implement optimal disaster response strategies. This study addresses these challenges by developing and implementing YOLOv8-based deep learning models trained on high-resolution satellite imagery from the Maxar GeoEye-1 satellite. Unlike prior studies, we introduce a manually labeled dataset, consisting of 1400 undamaged and 1200 damaged buildings, derived from pre- and post-Hurricane Maria imagery. This dataset has been publicly released, providing a benchmark for future disaster assessment research. Additionally, we conduct a systematic evaluation of optimization strategies, comparing SGD with momentum, RMSProp, Adam, AdaMax, NAdam, and AdamW. Our results demonstrate that SGD with momentum outperforms Adam-based optimizers in training stability, convergence speed, and reliability across higher confidence thresholds, leading to more robust and consistent disaster damage predictions. To enhance usability, we propose deploying the trained model via a REST API, enabling real-time damage assessment with minimal computational resources, making it a low-cost, scalable tool for government agencies and humanitarian organizations. These findings contribute to machine learning-based disaster response, offering an efficient, cost-effective framework for large-scale damage assessment and reinforcing the importance of model selection, hyperparameter tuning, and optimization functions in critical real-world applications.https://www.mdpi.com/2227-7390/13/7/1041machine learningdeep learningcomputer visiondetection modelsnatural disastershydrometeorological disasters
spellingShingle César Luis Moreno González
Germán A. Montoya
Carlos Lozano Garzón
Toward Reliable Post-Disaster Assessment: Advancing Building Damage Detection Using You Only Look Once Convolutional Neural Network and Satellite Imagery
Mathematics
machine learning
deep learning
computer vision
detection models
natural disasters
hydrometeorological disasters
title Toward Reliable Post-Disaster Assessment: Advancing Building Damage Detection Using You Only Look Once Convolutional Neural Network and Satellite Imagery
title_full Toward Reliable Post-Disaster Assessment: Advancing Building Damage Detection Using You Only Look Once Convolutional Neural Network and Satellite Imagery
title_fullStr Toward Reliable Post-Disaster Assessment: Advancing Building Damage Detection Using You Only Look Once Convolutional Neural Network and Satellite Imagery
title_full_unstemmed Toward Reliable Post-Disaster Assessment: Advancing Building Damage Detection Using You Only Look Once Convolutional Neural Network and Satellite Imagery
title_short Toward Reliable Post-Disaster Assessment: Advancing Building Damage Detection Using You Only Look Once Convolutional Neural Network and Satellite Imagery
title_sort toward reliable post disaster assessment advancing building damage detection using you only look once convolutional neural network and satellite imagery
topic machine learning
deep learning
computer vision
detection models
natural disasters
hydrometeorological disasters
url https://www.mdpi.com/2227-7390/13/7/1041
work_keys_str_mv AT cesarluismorenogonzalez towardreliablepostdisasterassessmentadvancingbuildingdamagedetectionusingyouonlylookonceconvolutionalneuralnetworkandsatelliteimagery
AT germanamontoya towardreliablepostdisasterassessmentadvancingbuildingdamagedetectionusingyouonlylookonceconvolutionalneuralnetworkandsatelliteimagery
AT carloslozanogarzon towardreliablepostdisasterassessmentadvancingbuildingdamagedetectionusingyouonlylookonceconvolutionalneuralnetworkandsatelliteimagery