集成空天地多源异构多视遥感数据的特大城市超高层建筑物三维重建研究

项目名称： 集成空天地多源异构多视遥感数据的特大城市超高层建筑物三维重建研究

项目编号： No.41201425

项目类型： 青年科学基金项目

立项/批准年度： 2013

项目学科： 地理学

项目作者： 乔刚

作者单位： 同济大学

项目金额： 25万元

中文摘要： 随着城市化进程的加快，建筑物三维重建成为测绘与遥感领域的研究热点。已有研究集中在中低层建筑物，较少关注具有更大需求的特大城市超高层建筑物，且多基于航空影像/LiDAR数据，不能同时获取建筑物顶部和立面精确细节，难以满足三维重建的细节性与完整性需求。针对以上问题，以特大城市超高层建筑物三维重建为目标，本项目将充分发挥卫星影像高分辨率、机载LiDAR高程点云和地面MMS序列影像多视角的优势，突破空天地多源异构多视遥感数据集成融合与超高层建筑物三维重建的理论与关键技术瓶颈，重点研究面向建筑物的多源数据基准统一与高精度定位，机载LiDAR辅助的立体高分影像建筑物顶面轮廓细节自适应提取，基于知识推理的MMS序列影像建筑物立面信息提取，超高层建筑物三维模型的高质量成型策略，形成一套集成空天地遥感数据、面向精确性-细节性-完整性的超高层建筑物三维重建方法，为新一代的城市精细三维建模提供理论和技术支撑。

中文关键词： 建筑物；移动测量系统；高分辨率影像；信息提取；三维重建

英文摘要： 3D building reconstruction has been a research hot spot in survey engineering and remote sensing community with the rapid development of urbanization. Existing research mainly focus on the low-rise and middle-rise buildings rather than super high-rise buildings in megalopolis with more demand, and based on aerial imagery/LiDAR data, thus precise rooftop and facade information could not be obtained at the same time. To solve this issue, this research will take advantages of the High Resolution Satellite Imagery (HRSI), LiDAR, and Mobile Mapping Sequential imagery (MMS) for the integration of the spaceborne-airborne-groundborne, multisource, heterogenous, and multiview remote sensing data, and for the 3D reconstruction of super high-rise buildings. The research consists of 4 parts, spatial datum unification and high accuracy positioning of multisource data for building reconstruction, adaptive extraction of building rooftop outline by integration of LiDAR and HRSI, building facade extraction from MMS images based on knowledge reasoning, and a strategy for 3D modelling of high-rise buildings with high quality. A new method will be produced, by integrating spaceborne-airborne-groundborne data, and aiming at new 3D reconstruction for super high-rise buildings with precision, details, and completeness.

英文关键词： building；mobile mapping system；high resolution imagery；information extraction；3D reconstruction