Structural Damage Detection Using AI Super Resolution and Visual Language Model

2025-08-23

AI Summary

To address the low efficiency, high cost, and safety risks associated with manual structural damage assessment after disasters, this paper proposes an end-to-end AI framework for damage evaluation using drone-captured imagery. Methodologically, it integrates a Video Restoration Transformer (VRT) to enhance the resolution of low-quality aerial video sequences and couples it with the large-parameter vision-language model Gemma-3B to enable automated damage detection, fine-grained classification (e.g., minor, moderate, severe), and natural-language risk description. The key contribution lies in the first joint modeling of video restoration and vision-language understanding, significantly improving interpretability and operational usability for non-expert users. Evaluated on real-world datasets from the Turkey earthquake and the Moore tornado (USA), the framework achieves 84.5% classification accuracy for damage severity and accelerates assessment throughput by multiple-fold, offering a robust technical solution for rapid response under resource-constrained conditions.

Technology Category

• Computer Vision: Remote Sensing / Geospatial AI
• Natural Language Processing: Safety and Robustness
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Natural disasters pose significant challenges to timely and accurate damage assessment due to their sudden onset and the extensive areas they affect. Traditional assessment methods are often labor-intensive, costly, and hazardous to personnel, making them impractical for rapid response, especially in resource-limited settings. This study proposes a novel, cost-effective framework that leverages aerial drone footage, an advanced AI-based video super-resolution model, Video Restoration Transformer (VRT), and Gemma3:27b, a 27 billion parameter Visual Language Model (VLM). This integrated system is designed to improve low-resolution disaster footage, identify structural damage, and classify buildings into four damage categories, ranging from no/slight damage to total destruction, along with associated risk levels. The methodology was validated using pre- and post-event drone imagery from the 2023 Turkey earthquakes (courtesy of The Guardian) and satellite data from the 2013 Moore Tornado (xBD dataset). The framework achieved a classification accuracy of 84.5%, demonstrating its ability to provide highly accurate results. Furthermore, the system's accessibility allows non-technical users to perform preliminary analyses, thereby improving the responsiveness and efficiency of disaster management efforts.

Problem

• Detecting structural damage from low-resolution aerial footage
• Classifying building damage levels using AI and visual models
• Improving disaster response with automated damage assessment

Innovation

• AI super-resolution enhances low-resolution disaster footage
• Visual language model classifies structural damage categories
• Integrated drone and satellite imagery for damage assessment