Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2009;9(6):4525-42.
doi: 10.3390/s90604525. Epub 2009 Jun 9.

Building facade reconstruction by fusing terrestrial laser points and images

Shi Pu  1 , George Vosselman
Affiliations

Building facade reconstruction by fusing terrestrial laser points and images

Shi Pu et al. Sensors (Basel). 2009.

Abstract

Laser data and optical data have a complementary nature for three dimensional feature extraction. Efficient integration of the two data sources will lead to a more reliable and automated extraction of three dimensional features. This paper presents a semiautomatic building facade reconstruction approach, which efficiently combines information from terrestrial laser point clouds and close range images. A building facade's general structure is discovered and established using the planar features from laser data. Then strong lines in images are extracted using Canny extractor and Hough transformation, and compared with current model edges for necessary improvement. Finally, textures with optimal visibility are selected and applied according to accurate image orientations. Solutions to several challenge problems throughout the collaborated reconstruction, such as referencing between laser points and multiple images and automated texturing, are described. The limitations and remaining works of this approach are also discussed.

Keywords: building reconstruction; data fusion; terrestrial laser scanning; texturing.

PubMed Disclaimer

Figures

Figure 1.
Figure 1.
Building reconstruction process.
Figure 2.
Figure 2.
A reconstructed building facade model, shown together with segmented laser points (color of point indicates segment; color of line indicates feature type).
Figure 3.
Figure 3.
Registration between laser points and images.
Figure 4.
Figure 4.
Selecting tie points for spatial resection.
Figure 5.
Figure 5.
Homography estimation using extracted SURF in two images (red lines: linked SURF pairs; black quadrangle: homographic transformed viewing plane of the upper image).
Figure 6.
Figure 6.
Locating occurrences of the same laser points in different image perspectives.
Figure 7.
Figure 7.
Extracting significant lines from an image (black: Canny edges; red: Hough lines; blue: projections of model edges; green: candidate matches; purple: best matches).
Figure 8.
Figure 8.
Refining the wall outlines of a building facade.
Figure 9.
Figure 9.
Refining the window outlines of a building facade.
Figure 10.
Figure 10.
Choosing the optimal texture images for different model parts.
Figure 11.
Figure 11.
An occlusion and its central perspective projection on a wall.
Figure 12.
Figure 12.
Reconstructed facade model of three joined houses.
Figure 13.
Figure 13.
A reconstructed facade model before and after removing occluded texturing.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Dick A., Torr P., Ruffle S., Cipolla R. Combining single view recognition and multiple view stereo forarchitectural scenes. Eighth IEEE International Conference on Computer Vision, 2001. 2001;1 ICCV 2001. Proceedings.
    1. Schindler K., Bauer J. A model-based method for building reconstruction. First IEEE International Workshop on Higher-Level Knowledge in 3D Modeling and Motion Analysis, 2003. 2003:74–82. HLK 2003.
    1. Frueh C., Jain S., Zakhor A. Data processing algorithms for generating textured 3D building facade meshes from laser scans and camera images. Inter. J. Comput. Vision. 2005;61:159–184.
    1. Becker S., Haala N. Combined Feature Extraction for Façade Reconstruction. Proceedings of the ISPRS Workshop Laser Scanning 2007 and SilviLaser 2007; Espoo, Finland. 12-14 Sept. 2007; pp. 241–247.
    1. Pu S., Vosselman G. Knowledge based reconstruction of building models from terrestrial laser scanning data. ISPRS J. Photogramm. Remote Sens. 2009 In Press, Corrected Proof. - PubMed

LinkOut - more resources

  NODES
INTERN 4
Project 3
twitter 2