研究目的
To propose an airborne LiDAR based approach for assessing multi-level hurricane damage at the community scale by extracting building clusters, matching pre-event and post-event clusters, and computing multiple damage indicators.
研究成果
The proposed approach is capable of automatically extracting individual building clusters and conducting damage assessment at the individual building level. It performs well on severely damaged buildings and can be used for rapid community-level building damage assessment.
研究不足
The proposed roof-based approach might not be applicable to all building damage states, such as minor structure geometry deformation like holes on the wall or scour of the foundation. The approach's accuracy is influenced by the density of LiDAR data sets.
1：Experimental Design and Method Selection:
The study employs a density-based clustering algorithm for building extraction and a novel cluster matching algorithm for matching pre-event and post-event building clusters. Multiple geometric features are computed as damage indicators.
2：Sample Selection and Data Sources:
Pre-event and post-event airborne LiDAR datasets from three communities impacted by Hurricane Sandy are used.
3：List of Experimental Equipment and Materials:
Airborne LiDAR data, Terrasolid or LAStools for point cloud processing.
4：Experimental Procedures and Operational Workflow:
The approach involves data pre-processing, clustering of building point cloud data, matching building clusters, computing damage indicators, and a hierarchical damage detection process.
5：Data Analysis Methods:
The study uses geometric computing for damage indicators and a hierarchical determination process for damage assessment.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
