Building damage feature analyses based on post-earthquake airborne LiDAR data
-
摘要: 本文利用2010年海地MW7.0地震震后获取的机载激光雷达(LiDAR)三维点云数据， 通过人机交互的方式选取受损程度不同的典型建筑物点云数据， 比较分析倒塌建筑物与完好建筑物点云数据的高度、 坡度和法向量等分布特征， 提出了用建筑物点云高度均值偏离度、 屋顶面坡度值以及法向量与天顶方向夹角等因子判定建筑物破坏程度. 试验分析结果表明， 高度均值偏离度因子对单个建筑物的破坏部分识别效果较好， 屋顶面坡度值因子可以识别建筑物破坏部分的边缘， 点云法向量与天顶方向夹角因子能够较好地识别大范围区域内的建筑物破坏情况， 因此上述判定因子均能在一定情况下表征建筑物的破坏情况.
-
关键词:
- 机载激光雷达(LiDAR)点云 /
- 海地MW7.0地震 /
- 建筑物破坏特征 /
- 判定因子
Abstract: Building damage detection can be more accuracy because that the airborne LiDAR system can acquire height of buildings and other high resolution information, therefore airborne LiDAR data will be an important data source in post-earthquake disaster evaluation in the future. This paper chooses the typical building point cloud data on different damage condition from the airborne LiDAR point cloud data acquired after the MW7.0 earthquake in Haiti in 2010, and compares the distribution of the features such as height, slope and normal vector of damaged and non-damaged buildings. And then we establish the building damage determination factors, such as mean height deviation, slope value of building roof, and the intersection angle between normal vector and zenith direction. The results show that all factors can be used to recognize building damage in different condition, that is to say, mean height deviation can be used to detect the damage of single building, the slope value can be used to detect the damage part border of building, the intersection angle is a better factor that can be used to detect building damage in large areas. -
-
![]()
图 1 p点坡度值计算示意图， 黑点为点云数据
Figure 1. The calculation schematic diagram of slope value for the point p. The black spots are the point cloud data
![]()
图 2 高度较低(a)和高度较高(b)区域的机载LiDAR点云数据
Figure 2. LiDAR point cloud data in lower altitude (a) and higher altitude (b)
![]()
图 3 部分倒塌建筑物(左)和完好建筑物(右)高度均值偏离度m的计算结果
(a) 光学影像; (b) 点云数据; (c) m值的空间分布; (d) m值与建筑物东坐标x的关系 (a) Optical image； (b) Point cloud data； (c) Spatial distribution of m； (d) Relationship between m and east coordinate x
Figure 3. Calculation results of mean height deviation m of part-damaged (left) and non-damaged buildings (right)
![]()
图 4 部分倒塌建筑物(a)和完好建筑物(b)的高度均值偏离度m分布直方图
Figure 4. Histogram distribution of mean height deviation m of part-damaged (a) and non-damaged buildings (b)
![]()
图 5 部分倒塌建筑物(左)和完好建筑物(右)坡度值s的计算结果
(a) s值的空间分布； (b) s值与建筑物东坐标x的关系 (a) Spatial distribution of s; (b) Relationship between s and east coordinate x
Figure 5. Calculation results of slope value s of part-damaged (left) and non-damaged buildings (right)
![]()
图 6 部分倒塌建筑物(a)和完好建筑物(b)的坡度值s分布直方图
Figure 6. Histogram distribution of slope value s of part-damaged (a) and non-damaged buildings (b)
![]()
图 7 部分倒塌建筑物(左)和完好建筑物(右)的法向量与天顶方向夹角θ的计算结果
(a) 法向量空间分布; (b) θ值的空间分布; (c) θ值与建筑物东坐标x的关系 (a) Spatial distribution of normal vector; (b) Spatial distribution of intersection angle θ; (c) Relationship between intersection angle θ and east coordinate x
Figure 7. Calculation results of intersection angle θ between normal vector and zenith direction of part-damaged (left) and non-damaged buildings (right)
![]()
图 8 部分倒塌建筑物(a)和完好建筑物(b)的法向量与天顶方向夹角θ分布直方图
Figure 8. Histogram distribution of intersection angle θ between normal vector and zenith diriection of part-damaged (a) and non-damaged buildings (b)
![]()
图 9 研究区域内各建筑物震害因子计算结果
(a) 光学影像； (b) 高度均值偏离度m； (c) 坡度值s； (d) 法向量与天顶方向夹角θ 红色为识别出的倒塌点， 蓝色为识别出的未倒塌点 (a)Optical image; (b)Mean height deviation m;(c)Slope value s;(d)Intersection angle θ between normal vector and zenith direction.Red and blue represent the detected damaged and un-damaged points,respectively
Figure 9. Calculation results of all factors for determing building damage in studied area
-
Hoppe H, DeRose T, Duchamp T, McDonald J, Stuetzle W. 1992. Surface reconstruction from unorganized points[C]// Proceedings of the 19th Annual Conference on Computer Graphics and Interactive Techniques. New York, USA: ACM: 71-78.
Khoshelham K, Elberink S O. 2012. Role of dimensionality reduction in segment-based classification of damaged building roofs in airborne laser scanning data[C]//Proceedings of the 4th GEOBIA. Rio de Janeiro, Brazil: 372-377.
OpenTopography. 2010. Post-January 2010 Haiti earthquake LiDAR data now available via OpenTopography[EB/OL]. [2015-08-17]. http://opentopography.org/news/post-january-2010-haiti-earthquake-lidar-data-now-available-opentopography.
Schweier C, Markus M. 2004. Assessment of the search and rescue diand for individual buildings[C]//Proceedings of the 13th World Conference on Earthquake Engineering. Vancouver, Canada: Mira Digital Publishing: 3092.
Vu T T, Matsuoka M, Yamazaki F. 2004. LIDAR-based change detection of buildings in dense urban areas[C]//Proceedings of the 2004 IEEE International Geoscience and Riote Sensing Symposium. Anchorage, AK: IEEE: 3413-3416.
下载:
计量
- 文章访问数: 683
- HTML全文浏览量: 268
- PDF下载量: 21
