融合OSM和车载激光扫描数据的建筑物三维快速提取
3D Buildings Fast Detection by Fusion of Open Street Map Data and MLS Point Clouds
DOI: 10.12677/GST.2018.64044, PDF,   
作者: 石蒙蒙*, 杨必胜, 刘缘, 宋爽, 董震：武汉大学，测绘遥感信息工程国家重点实验室，湖北 武汉
关键词: Open Street Map；车载激光点云；建筑物提取；数据融合；ICP配准；Open Street Map； MLS； Building Extraction； Data Fusion； ICP Registration
摘要: 建筑物作为一种重要的地理空间要素，在城市规划与建设、居住环境、交通管理、房地产以及灾害损失评估等领域占有极为重要的地位。车载激光扫描(Mobile Laser Scanning, MLS)数据是建筑物三维模型重建的主要数据源之一，本文针对基于MLS数据建筑物的提取耗时过长和错分漏提率高等问题，提出了一种融合Open Street Map (OSM)和MLS点云数据的三维建筑物快速提取方法。该方法包含数据预处理、改进的ICP精配准、OSM与MLS数据的信息融合及辅助提取四个步骤，实验结果表明该方法可以有效提升效率，处理时间可缩短50%以上；同时该方法还可有效避免误提取问题，对漏提取问题也有所改善，因此提升了建筑物提取的精度。两种数据的融合，实现了双方的优势互补，为建立位置精准、关系正确、几何细节和属性信息丰富，方便快速更新的城市三维模型，提供了数据支撑。
Abstract: As a geospatial element, building plays an extremely important role in fields of urban planning and construction, residential environment, traffic management, real estate industry, and hazard assessment. Mobile Laser Scanning (MLS) data is one of the main data sources for 3D-models reconstruction of buildings. This paper proposes a fast method of 3D-building extraction from MLS point clouds, by fused with Open Street Map (OSM) data, to solve the problems of time-consuming and building mistaken extraction. The method includes four steps: data preprocessing, improved ICP fine registration, information fusion of OSM and MLS, and assisted building extraction. The experimental results show that this method can effectively improve the efficiency, the processing time is shortened by more than 50%, and it is effective to avoid the problem of building mistaken and missing, therefore improves the precision and accuracy of building extraction. The fusion of OSM and MLS data can provide data support for 3D urban models with accurate location, correct relations, geometric details, and rich semantic information.
文章引用：石蒙蒙, 杨必胜, 刘缘, 宋爽, 董震. 融合OSM和车载激光扫描数据的建筑物三维快速提取[J]. 测绘科学技术, 2018, 6(4): 363-373. https://doi.org/10.12677/GST.2018.64044

参考文献

[1] 关丽, 丁燕杰, 张辉, 等. 面向数字城市建设的三维建模关键技术研究与应用[J]. 测绘通报, 2017(2): 90-94.
[2] Yang, B., Wei, Z., et al. (2013) Semiautomated Building Facade Footprint Extraction from Mobile LiDAR Point Clouds. IEEE Geoscience and Remote Sensing Letters, 10, 766-770. [Google Scholar] [CrossRef] 
[3] Wei, Z., Yang, B., Li, Q., et al. (2012) Automated Extraction of Building Footprints from Mobile LIDAR Point Clouds. Journal of Remote Sensing, 16, 286-296.
[4] 龚健雅, 崔婷婷, 单杰, 等. 利用车载移动测量数据的建筑物立面建模方法[J]. 武汉大学学报: 信息科学版, 2015, 40(9): 1137-1143.
[5] Varady, T. and Benko, P. (2000) Reverse Engineering B-Rep Models from Multiple Point Clouds. IEEE Geometric Modeling and Processing, Hong Kong, 10-12 April 2000, 3-12. [Google Scholar] [CrossRef] 
[6] 张瑞菊, 郑少开, 尹建英. 建筑物立面点云语义分割方法研究[J]. 城市勘测, 2017(5): 1672-8262.
[7] 杨必胜, 魏征, 李清泉, 毛庆洲. 面向车载激光扫描点云快速分类的点云特征图像生成方法[J]. 测绘学报, 2010, 39(5): 540-545.
[8] 刘亚文, 庞世燕, 左志奇. 蚁群算法的建筑立面点云数据提取[J]. 武汉大学学报: 信息科学版, 2012(11): 1307-1310.
[9] 彭晨, 余柏蒗, 吴宾, 等. 基于移动激光扫描点云特征图像和SVM的建筑物立面半自动提取方法[J]. 地球信息科学学报, 2016(7): 878-885.
[10] 倪欢, 张继贤, 段敏燕, 等. 面向对象的车载激光扫描点云建筑物立面提取[J]. 测绘科学, 2014(4): 92-96.
[11] 杨必胜, 董震, 魏征, 等. 从车载激光扫描数据中提取复杂建筑物立面的方法[J]. 测绘学报, 2013(3): 411-417.
[12] Yang, B., Zhen, D., et al. (2015) Hierarchical Extraction of Urban Objects from Mobile Laser Scanning Data. ISPRS Journal of Photogrammetry and Remote Sensing, 99, 45-57. [Google Scholar] [CrossRef] 
[13] 张云菲. 多源道路网与兴趣点的一致性整合方法[D]: [博士学位论文]. 武汉: 武汉大学, 2015.
[14] Davidovic, N., et al. (2016) Tagging in Volunteered Geographic Information: An Analysis of Tagging Practices for Cities and Urban Regions in Open Street Map. ISPRS International Journal of Geo-Information, 5, 232. [Google Scholar] [CrossRef] 
[15] Girres, J.-F., et al. (2010) Quality Assessment of the French Open Street Map Dataset. Transactions in GIS, 14, 435-459. [Google Scholar] [CrossRef] 
[16] Goodchild, M.F. and Li, L. (2012) Assuring the Quality of Volunteered Geographic Information. Spatial Statistics, 1, 110-120. [Google Scholar] [CrossRef] 
[17] Moré, J.J. (1978) The Levenberg-Marquardt Algorithm: Implementation and Theory. In: Watson, G.A., Ed., Numerical Analysis. Lecture Notes in Mathematics, vol 630, Springer, Berlin, Heidelberg. [Google Scholar] [CrossRef] 
[18] Hernández, J. and Matcotegui, B. (2009) Filtering of Artifacts and Pavement Segmentation from Mobile Lidar Data. International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 38, 329-333.
[19] Roynard, X., Deschaud, J.E. and Goulette, F. (2017) Paris-Lille-3D: A Large and High-Quality Ground Truth Urban Point Cloud Dataset for Automatic Segmentation and Classification. arXiv preprint arXiv, 1712.00032.

为你推荐