测绘通报 ›› 2022, Vol. 0 ›› Issue (6): 143-147.doi: 10.13474/j.cnki.11-2246.2022.0187.

• 技术交流 • 上一篇    下一篇

多源数据在复杂场景建筑物三维重建的应用

谢元, 董梦, 马新建   

  1. 北京市勘察设计研究院有限公司, 北京 100038
  • 收稿日期:2021-08-30 发布日期:2022-06-30
  • 作者简介:谢元(1987-),女,工程师,主要研究方向为多源数据在复杂场景建筑物三维重建的应用。E-mail:476870758@qq.com

Application of multi-source data in 3D reconstruction of buildings in complex scenes

XIE Yuan, DONG Meng, MA Xinjian   

  1. Beijing Engineering Consultant Co., Ltd., Beijing 100038, China
  • Received:2021-08-30 Published:2022-06-30

摘要: 随着数字化城市建设的推进,更多复杂大型的建筑需要完成三维建模工作。对于异形且复杂的建筑而言,采用传统的数据采集方式较为烦琐费力。本文利用地面激光扫描仪获取某大型建筑内、外部点云数据,同时结合近景摄影测量技术获得建筑物的多角度影像及正射影像数据,经过点云配准、去噪、几何建模及纹理映射等处理,最终完成大型建筑的三维场景重建,并对此过程中涉及的关键技术进行了研究。

关键词: 三维激光扫描, 三维建模, 近景摄影测量, 点云配准, 去噪滤波, 纹理

Abstract: With the development of digital city construction, more and more complex large buildings need to complete 3D modeling. For the irregular and complex buildings, the traditional total station data acquisition method is more cumbersome and laborious. In this paper, the ground laser scanner is used to obtain the point cloud data inside and outside a large building, and the close range photogrammetry is used to obtain the orthophoto map of the top of the building. The site photos are taken at the small components, and the point cloud registration, denoising, image processing geometric modeling and texture mapping process are carried.Finally the 3D reconstruction of large buildings is completed, and the key technology is researched in the process.

Key words: 3D laser scanning, 3D modeling, close-range photogrammetry, point cloud registration, denoising filter, texture

中图分类号: