...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Natural hazards and earth system sciences >UAV-based urban structural damage assessment using object-based image analysis and semantic reasoning
【24h】

UAV-based urban structural damage assessment using object-based image analysis and semantic reasoning

机译：基于对象图像分析和语义推理的基于无人机的城市结构破坏评估

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Structural damage assessment is critical after disasters but remains a challenge. Many studies have explored the potential of remote sensing data, but limitations of vertical data persist. Oblique imagery has been identified as more useful, though the multi-angle imagery also adds a new dimension of complexity. This paper addresses damage assessment based on multi-perspective, overlapping, very high resolution oblique images obtained with unmanned aerial vehicles (UAVs). 3-D point-cloud assessment for the entire building is combined with detailed object-based image analysis (OBIA) of fa double dagger ades and roofs. This research focuses not on automatic damage assessment, but on creating a methodology that supports the often ambiguous classification of intermediate damage levels, aiming at producing comprehensive per-building damage scores. We identify completely damaged structures in the 3-D point cloud, and for all other cases provide the OBIA-based damage indicators to be used as auxiliary information by damage analysts. The results demonstrate the usability of the 3-D point-cloud data to identify major damage features. Also the UAV-derived and OBIA-processed oblique images are shown to be a suitable basis for the identification of detailed damage features on fa double dagger ades and roofs. Finally, we also demonstrate the possibility of aggregating the multi-perspective damage information at building level.
机译：灾后结构性损坏评估至关重要，但仍然是一个挑战。许多研究已经探索了遥感数据的潜力，但是垂直数据的局限性仍然存在。尽管多角度图像也增加了复杂性的新维度，但倾斜图像被认为更有用。本文基于无人飞行器（UAV）获得的多角度，重叠，高分辨率的倾斜图像进行了损害评估。整个建筑物的3-D点云评估与双匕首和屋顶的详细的基于对象的图像分析（OBIA）相结合。这项研究的重点不在于自动损害评估，而是在于创建一种方法，该方法通常支持对中间损害水平进行不明确的分类，旨在得出综合的每栋建筑物损害评分。我们在3-D点云中识别出完全损坏的结构，并为所有其他情况提供基于OBIA的损坏指标，以供损坏分析人员用作辅助信息。结果证明了3-D点云数据可用于识别主要破坏特征。无人机衍生和OBIA处理的倾斜图像也被证明是识别双匕首和屋顶上详细损伤特征的合适基础。最后，我们还演示了在建筑物级别汇总多角度损坏信息的可能性。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号