Fig9 results of extracting geometric information from the point 
cloud 
Example ID 
Points(k) 
Triangles(k) 
Vertices 
Edges 
Faces 
1 
12 
23 
14 
21 
9 
2 
13 
26 
8 
12 
6 
3 
8 
21 
10 
15 
7 
4 
13 
27 
16 
24 
10 
5 
21 
41 
8 
10 
3 
Tabl Related information of the examples 
In FiglO, there are two complex building from the LiDAR data 
of Lu Jia Zui district in Shanghai, China. Due to the complex 
and curved surface of the building, the result can not give the 
complete geometric features and other information. 
The two main reasons for this problem are as follow: 1) the 
object has a great quantity of complex, small, curved surfaces. 
So it is difficult to get the ideal triangle mesh [9] . 2) the method 
in this paper is not aimed at this condition. 
liia 
FiglO results of extracting geometric information from the 
buildings with complex surface 
4.CONCLUSIONS 
In this article, Triangles classification and Surface extraction, 
Surface contiguous relations restoration, and Polyhedron 
Vertices solution and Polyhedron model Reconstruction 
constitute the main research of extracting geometric features 
and building the model of the object. And from result of 
examples, those used to prove these algorithms, four 
conclusions can be summarized as follow: 
1. To the polyhedron and the object, that has the features of 
polyhedron, the method raised in the paper can effectively and 
completely extract the geometric features and topological 
information among them. 
2. To the object, that has a great quantity of complex, small, 
curved surfaces, this method can not effectively and completely 
extract the geometric features and topological information 
among them. 
3. The quality of triangle mesh of the surface has a great impact 
on the final result. 
4. This method only uses the geometric features (vertices, 
edges and surfaces) and the topological information among 
them to model the object, while occupy less memory space, and 
it can store more comprehensive structural information. 
ACKNOWLEDGEMENTS 
The paper is substantially supported by National High 
Technology Research and Development Program of China (863 
Program) “Complicated Features’ Extraction and Analysis from 
Central Districts in Metropolis”, ChangJiang Scholars 
Program ,Ministry of Education, PRC and Tongji University 
“985”Project Sub-system “Integrated Monitoring for City 
Spatial Information”. The author would like to thank Chang 
Jiang Scholars Dr. Ron., Li and Dr. Hongxing Sun for their 
substantial help. 
REFERENCES 
C Oblonsek, N Guid, 1998.A Fast Surface-Based Procedure for 
Object Reconstruction From 3D Scattered Points. Computer 
Vision and Image Understanding,69\ 185-195 
Christopher Gold, HugoLedoux, MarcinDzieszko. 2005.A Data 
Structure for the Construction and Navigation of 3D Voronoi 
and Delaunay Cell Complexes. In Proceedings 13th 
International Conference in Central Europe on Computer 
Graphics, Visualization and Computer Vision, ,21-22. 
Erlend Arege, Morten Dahlen, Aslak Tveito,1995. 
Approximation of Scattered Data Using Smooth Grid Function. 
Journal of Computational and Applied Mathematics, 
59:191-205 
H-C Kim, S-M Hur, S-H Lee,2002.Segmentation of the 
Measured Point Data in Reverse Engineering. International 
Journal of Advanced Manufacturing Technology, Vol20: 
571-580 
Hyungjun Park, Kwangsoo Kim. 1995.^4« Adaptive Method for 
Smooth Surface Approximation to Scattered 3D Points. 
Computer-A ided Design,Vol27:929~939 
Koivunen V, Pietikainen M,1991. Combined Edge and Region- 
Based Method for Range Image Segmentation. Proceedings of 
the International Society for Optical Engineering, 1381 : 
501-512 
Otsu, N., 1979. A threshold selection method from gray-level 
histograms. IEEE Trans. Systems, Man, and Cybernetics, 9(1), 
pp. 62-66. 
Shunyi Zheng, Zongqian Zhan, Zuxun Zhang,2004, 9. A 
Flexible and Automatic 3D Reconstruction Method, ISPRS 
Congress, Istanbul, Turkey,. 
Yan ping, Jiang Wanshou,2006. Reconstruction of Buildings 
with Multi Right Angles and Multilayers from LIDAR, 2006, 
31(6):492-495.