The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part B3b. Beijing 2008 
subset the image into several small images and deals with them 
one by one. 
In the segmentation step, the select of parameter scale is very 
important. The weight of DSM layer is set to 0. It means that 
the DSM layer doesn’t involve in segmentation. That’s because 
the edge of buildings in DSM layer usually has low contrast 
with the background, and also because the low resolution of 
DSM. If with DSM in segmentation layer, the edges of the 
buildings in the extraction become rather rough. 
Because we have adopted the object-oriented method, we deal 
with the objects; the edge of extraction is not as smooth as 
pixel-based method. But this can be overcome through post 
processing, like using rectangle restriction to make the edge 
regular. Also, building extraction by DSM and high-resolution 
imagery is robust, but its popularity is not so good as the 
model-based building extraction. 
ACKNOWLEDGEMENT 
This work was developed within the National Key Basic 
Research and Development Program under grant 
NO.2006CB701303. 
REFERENCE 
Definiens, UserGuide, 2007 
Fu, K. S., Mui, J.K. 1981. A survey on image segmentation. 
Pattern Recognition, 13, 3-16. 
Forstner, W.1999. 3D-City Models: Automatic and 
Semiautomatic Acquisition Methods, Photogrammetric Week, 
Wichmann Verlag, Heidelberg, pp291-303. 
Gerke, M., Straub, B. M., Koch, A. Automatic Detection of 
Buildings and Trees from Aerial Imagery Using Different 
Levels of Abstraction. http://cmsv021.rrzn.uni- 
hannover.de/uploads/tx tkpublikationen/gerke_straub_koch_dg 
pf.pdf (accessed 10 Sep. 2007) 
Huang, H.P. 2003.Scale Issues in object-oriented Image 
Analysis, Ph.D dissertation. 
Huber, M., Schickler, W., Hinz, S., Baumgartner. A. Fusion of 
LIDAR Data and Aerial Imagery for Automatic Reconstruction 
of Building Surfaces, 
http://www.sanbom.com/Pdfs/Article_Fusion_of_Lidar_Schickl 
er.pdf(accessed 10 Sep. 2007) 
Jiang, W.S. 2004. Multiple Aerial Image Matching and 
Automatic Building Detection, Ph.D dissertation 
Lin,C., Nevada, R. 1998. Building detection and description 
from a single intensity image. Computer Vision and Image 
Understanding: CVIU 72(2), pp. 101-121. 
Mathisa, O., Xavier, D., Josiane, Z. 2002. Building Extraction 
from Digital Elecation Model. INRIA Report. 
Müller, S., Zaum, D. W. 2005. Robust Building Detection in 
Aerial Images, IAPRS, Vol.36, part3/W24. Vienna,Austria, 
August29-30 
Nevada, R., Lin, C., Huertas, A. 1997. A System for Building 
Detection from Aerial Images, Monocular Building Detection 
Ascona Workshop ,Ascona Switzewrland, Apr. 1997. 
Pasko, M., Gruber, M., 1996. Fusion of 2D GIS Data and Aerial 
Images for 3D Building Reconstruction. IAPRS, Vol. 31, Part 
B3, pp. 257-260. 
Roux, M., McKeown, D. 1994. Feature Matching for Building 
Extraction from Multiple Views. Proc. IEEE Computer Vision 
and Pattern Recognition, Seattle, WA, pp. 46-53. 
Zhang, Y.J. An Overview of Image and Video Segmentation, 
2006,http://image.ee.tsinghua.edu.cn/Papers/E123=AIVS- 
06b.pdf(accessed 15 Sep. 2007)