Robust Building Damage Detection in Cross-Disaster Settings Using Domain Adaptation

2026-03-15

AI Summary

This study addresses the significant performance degradation in building damage assessment under cross-disaster scenarios due to domain shift, which undermines the reliability of human-AI collaborative emergency decision-making. The authors propose a two-stage ensemble approach that systematically validates, for the first time, the critical role of supervised domain adaptation (SDA) in cross-disaster building damage detection—demonstrating that its absence leads to complete model failure. Furthermore, they enhance RGB inputs using unsharp masking to improve robustness in classifying four damage severity levels. Evaluated on the unseen test set of Ida-BD, the method achieves a Macro-F1 score of 0.5552, substantially outperforming baseline approaches and offering a reliable solution for cross-domain disaster damage assessment.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Natural Language Processing: Safety and Robustness

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Rapid structural damage assessment from remote sensing imagery is essential for timely disaster response. Within human-machine systems (HMS) for disaster management, automated damage detection provides decision-makers with actionable situational awareness. However, models trained on multi-disaster benchmarks often underperform in unseen geographic regions due to domain shift - a distributional mismatch between training and deployment data that undermines human trust in automated assessments. We explore a two-stage ensemble approach using supervised domain adaptation (SDA) for building damage classification across four severity classes. The pipeline adapts the xView2 first-place method to the Ida-BD dataset using SDA and systematically investigates the effect of individual augmentation components on classification performance. Comprehensive ablation experiments on the unseen Ida-BD test split demonstrate that SDA is indispensable: removing it causes damage detection to fail entirely. Our pipeline achieves the most robust performance using SDA with unsharp-enhanced RGB input, attaining a Macro-F1 of 0.5552. These results underscore the critical role of domain adaptation in building trustworthy automated damage assessment modules for HMS-integrated disaster response.

Problem

• domain shift
• building damage detection
• cross-disaster
• remote sensing
• distributional mismatch

Innovation

• supervised domain adaptation
• cross-disaster damage detection
• building damage assessment
• domain shift
• remote sensing imagery