Previous Article in Journal
Discrete Element Simulation Parameter Calibration of Wheat Straw Feed Using Response Surface Methodology and Particle Swarm Optimization–Backpropagation Hybrid Algorithm
Previous Article in Special Issue
ShadeNet: Innovating Shade House Detection via High-Resolution Remote Sensing and Semantic Segmentation
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
This is an early access version, the complete PDF, HTML, and XML versions will be available soon.
Article

Deep Learning-Based Detection and Assessment of Road Damage Caused by Disaster with Satellite Imagery

1
Department of Urban Planning and Landscape Architecture, Dong-A University, Busan 49315, Republic of Korea
2
Department of Electronic Engineering, Dong-A University, Busan 49315, Republic of Korea
3
Department of Urban Planning and Engineering, Dong-A University, Busan 49315, Republic of Korea
*
Author to whom correspondence should be addressed.
These authors contributed equally to this work.
Appl. Sci. 2025, 15(14), 7669; https://doi.org/10.3390/app15147669 (registering DOI)
Submission received: 9 June 2025 / Revised: 30 June 2025 / Accepted: 7 July 2025 / Published: 8 July 2025
(This article belongs to the Special Issue Remote Sensing Image Processing and Application, 2nd Edition)

Abstract

Natural disasters can cause severe damage to critical infrastructure such as road networks, significantly delaying rescue and recovery efforts. Conventional road damage assessments rely heavily on manual inspection, which is labor-intensive, time-consuming, and infeasible in large-scale disaster-affected areas. This study aims to propose a deep learning-based framework to automatically detect and quantitatively assess road damage using high-resolution pre- and post-disaster satellite imagery. To achieve this, the study systematically compares three distinct change detection approaches: single-timeframe overlay, difference-based segmentation, and Siamese feature fusion. Experimental results, validated over multiple runs, show the difference-based model achieved the highest overall F1-score (0.594 ± 0.025), surpassing the overlay and Siamese models by approximately 127.6% and 27.5%, respectively. However, a key finding of this study is that even this best-performing model is constrained by a low detection recall (0.445 ± 0.051) for the ’damaged road’ class. This reveals that severe class imbalance is a fundamental hurdle in this domain for which standard training strategies are insufficient. This study establishes a crucial benchmark for the field, highlighting that future research must focus on methods that directly address class imbalance to improve detection recall. Despite its quantified limitations, the proposed framework enables the visualization of damage density maps, supporting emergency response strategies such as prioritizing road restoration and accessibility planning in disaster-stricken areas.
Keywords: road damage detection and assessment; satellite imagery; deep learning; semantic segmentation; infrastructure resilience road damage detection and assessment; satellite imagery; deep learning; semantic segmentation; infrastructure resilience

Share and Cite

MDPI and ACS Style

Cha, J.; Lee, S.; Kim, H.-K. Deep Learning-Based Detection and Assessment of Road Damage Caused by Disaster with Satellite Imagery. Appl. Sci. 2025, 15, 7669. https://doi.org/10.3390/app15147669

AMA Style

Cha J, Lee S, Kim H-K. Deep Learning-Based Detection and Assessment of Road Damage Caused by Disaster with Satellite Imagery. Applied Sciences. 2025; 15(14):7669. https://doi.org/10.3390/app15147669

Chicago/Turabian Style

Cha, Jungeun, Seunghyeok Lee, and Hoe-Kyoung Kim. 2025. "Deep Learning-Based Detection and Assessment of Road Damage Caused by Disaster with Satellite Imagery" Applied Sciences 15, no. 14: 7669. https://doi.org/10.3390/app15147669

APA Style

Cha, J., Lee, S., & Kim, H.-K. (2025). Deep Learning-Based Detection and Assessment of Road Damage Caused by Disaster with Satellite Imagery. Applied Sciences, 15(14), 7669. https://doi.org/10.3390/app15147669

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop