Natural disasters continue to cause substantial economic losses and societal disruptions worldwide, underscoring the urgent need for accurate and timely disaster impact assessments. Traditional approaches to post-disaster damage evaluation, particularly those relying on manual inspections or limited data sources, often fail to capture the complex and localized impacts on the built environment. These limitations hinder effective disaster response and long-term recovery planning. Recent advances in artificial intelligence and point cloud data acquisition technologies have opened new opportunities for detailed analysis of disaster-affected areas. However, many existing systems still lack the spatial resolution and automation necessary for precise damage quantification, especially in densely built environments. To address these challenges, this study proposes an automated framework that leverages LiDAR (Light Detection and Ranging) point cloud data from post-disaster built environments to perform detailed and quantitative damage assessments. The framework consists of three main stages: (1) pre-processing and planar patch extraction, (2) segmentation refinement, and (3) damage assessment and visualization. The proposed framework, validated on LiDAR datasets of tornado-affected built environments, achieved over 95% F1-score across all datasets for building region identification and successfully detected multiple structural damage types. This approach contributes to enhancing the efficiency of post-disaster assessments and offers a scalable foundation for future resilient built environment monitoring.
Kim et al. (Sun,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: