gure 
0.3 
ions 
isa 
een 
gure 
cted 
nes. 
ully 
and 
on a 
and 
It 1s 
3D 
tem 
res. 
Bofang Zhao 
  
The test results show that its two major components: building detection and segment extraction, operate with 
satisfactory performance. At present, feature matching and 3D modelling components are under development. Future 
work will be focused on these areas. 
    
mos HE 
P AR " 
b TR 
Pl 
B ut 
  
(a) Original Image (b) Linear Features 
b 
   
(c) Detected building regions (d) Extracted building boundaries 
Figure 8. Processing Results 
REFERENCES 
  
| Baillard, C., Schmid, C., Zisserman, A. and Fitzgibbon, A., 1999. Automatic line matching and 3D reconstruction of 
buildings from multiple views. International Archives of Photogrammetry and Remote Sensing, Vol.32, Part 3-2W5, pp. 
69-80. 
Baltsavias, E, Mason, S. and Stallmann, D., 1995. Use of DTMs/DSMs and orthoimages to support building extraction. 
In Grün, A., Kübler, O. and Agouris, P. (eds), Automatic Extraction of Man-Made Objects from Aerial and Space 
Images, Birkhäuser Verlag, Basel, pp. 199-210. 
Berthod. M.. Gabet, L., Giraudon, G. and Lotti, J. L., 1995. High resolution stereo for the detection of buildings. . In 
Grün, A., Kübler, O. and Agouris, P. (eds), Automatic Extraction of Man-Made Objects from Aerial and Space Images, 
Birkhäuser Verlag, Basel, pp. 135-144. 
Bruun, A., Gülch, E., Lang, F., Fórstner, 1998. A hybrid concept for 3D building acquisition. ISPRS Jouranl of 
Photogrammetry and Remote Sending, Vol.53:119-129. 
Brunn. A. and Weidner, U., 1998. Hierarchical Bayesian nets for building extraction using dense digital surface models. 
ISPRS Jouranl of Photogrammetry and Remote Sending, V01.53:296-307. 
  
International Archives of Photogrammetry and Remote Sensing. Vol. XXXIII, Part B3. Amsterdam 2000. 1031