Iso varied. 
ass than if 
patch but 
set of 100 
s. Image 
described 
Machine 
tches into 
fier using 
sigmoidal 
| a simple 
e patches, 
| image of 
building. 
onfidence 
reliability 
' decision 
produced 
ns, 1998), 
f that was 
dertaken. 
Although 
may still 
) building 
i and the 
classified 
y 8 were 
> original 
ce of the 
data. This 
er work is 
ng set to 
> can be 
te set of 
iS case, a 
it of data 
Formation 
cation is 
6 and 32 
x 16 and 
1 wavelet 
ions, the 
was no 
At lower 
as lost to 
S Of tlie 
n. As the 
generates 
ttempt to 
ISPRS Commission III, Vol.34, Part 3A ,,Photogrammetric Computer Vision", Graz, 2002 
  
6. CONCLUSION 
Machine learning methods have been used successfully in 
several image processing and machine vision domains but there 
has been little research into their potential for photogrammetric 
applications. While these techniques often cannot satisfy the 
metric requirements of photogrammetry, they can provide 
useful starting points and heuristic filters in the area of 
automated object extraction. 
The Support Vector Machine is well suited to this application, 
as it does not suffer from the problem of local minima and 
produces a statistically robust decision surface. The SVM 
recasts the problem into high dimensional feature space, where 
problems that are not linearly separable in lower-dimensional 
feature space may become separable. 
An important aspect of machine learning in vision applications 
is to extract a representative set of characteristics from the 
image. The multi-resolution approach of wavelets does this 
quite nicely and is well supported by psycho-physical evidence 
suggesting that mammalian vision systems operate in a similar 
manner. 
Although some refinement and further testing are required, the 
machine learning approach outlined in this paper could be used 
to identify image patches that are likely to contain a building. 
As such, it would act as a heuristic filter, providing only image 
patches that had a high probability of containing a building to 
the functions that perform the building extraction processes. 
7. REFERENCES 
Agouris, P., Gyftakis, S. & Stefanidis, A., 1998. Using A Fuzzy 
Supervisor for Object Extraction within an Integrated 
Geospatial Environment. In: International Archives of 
Photogrammetry and Remote Sensing, Columbus, Ohio, Vol. 
XXXII, Part III/1, pp. 191-195. 
Boser, B. E., Guyon, I. M. & Vapnik, V. N., 1992. A training 
algorithm for optimal margin classifiers. In: The 5th Annual 
ACM Workshop on Computational Learning Theory, 
Pittsburgh, ACM Press. 
Canny, J. F., 1986. A Computational Approach to Edge 
Detection. IEEE Transactions on Pattern Analysis and Machine 
Intelligence 8(6): pp. 679-686. 
Cristianini, N. & Shawe-Taylor, J., 2000. An Introduction to 
Support Vector Machines and other kernel-based learning 
methods. Cambridge, UK, Cambridge University Press. 
Field, D., 1994. What is the Goal of Sensory Coding? Neural 
Computation 6(4), pp. 559-601. 
Grossberg, S., 1988. Nonlinear Neural Networks. Principles, 
Mechanisms, And Architectures. Neural Networks 1: pp. 17-61. 
Gruen, A. & Dan, H., 1997. TOBAGO- a topology builder for 
the automated generation of building models. In : Automatic 
Extraction of Man-Made Objects from Aerial and Space 
Images. Eds. A. Gruen, Baltsavias, E.P. & Henricsson, O. 
Basel, Switzerland, Birkhauser,: 393. 
Henricsson, O.,1996. Analysis of image structures using color 
attributes and similarity relations. Unpublished PhD Thesis, 
Department of Geodesy and Photogrammetry. Zurich, Swish 
Federal Institute of Technology: pp. 124. 
Israel, S. and Kasabov, N., 1997. Statistical, connectionist and 
fuzzy inference techniques for image classification. Journal of 
Electronic Imaging, 6(3): 1-11. 
Joachims, T., 1998. Making Large-Scale SVM Learning 
Practical. In: Advances in Kernel Methods - Support Vector 
Learning. Eds. B. Scholkopf, Burges, C.J. & Smola, AJ. 
Cambridge, USA, MIT Press. 
Lee, D. & Schenk, T., 1998. An Adaptive Approach for 
Extracting Texture Information and Segmentation. In: 
International Archives of Photogrammetry and Remote 
Sensing, Columbus, Ohio, Vol. XXXII, Part III/1. 
Li, R., Wang, W. & Tseng, H.-Z., 1998. Object Recognition and 
Measurement from Mobile Mapping Image Sequences using 
Hopfield Neural Networks: Part 1. ASPRS Annual Conference, 
Tampa, Florida, USA, American Society of Photogrammetry 
and Remote Sensing. 
Loung, G. and Tan, Z., 1992. Stereo matching using artificial 
neural networks. International Archives of Photogrammetry and 
Remote Sensing, XXIX(B3/III): 417-422. 
Mallat, S. G. 1989., A Theory for Multiresolution Signal 
Decomposition: The Wavelet Representation. IEEE 
Transactions on Pattern Analysis and Machine Intelligence 
II(7): pp. 674-693. 
Marr, D., 1982. Vision : A computational investigation into the 
human representation and processing of visual information. 
New York, W.H. Freeman and Company. 
Michel, A., Oriot, H. & Goretta, O., 1998. Extraction of 
Rectangular Roofs on Stereoscopic Images - An Interactive 
Approach. In: International Archives of Photogrammetry and 
Remote Sensing, Columbus, Ohio, Vol. XXXII, Part III/1. 
Nevatia, R., Huertas, A. & Kim, Z., 1998. The MURI Project 
for Rapid Feature Extraction in Urban Areas. In: International 
Archives of Photogrammetry and Remote Sensing, Columbus, 
Ohio, Vol. XXXII, Part III/1. 
Oren, M., Papageorgiou, C., Sinha, P., Osuna, E. & Poggio, T., 
1997. Pedestrian Detection Using Wavelet Templates. In: 
Proceedings of Computer Vision and Pattern Recognition, 
Puerto Rico. 
Osuna, E., Freund, R. & Girosi, F., 1997. Training Support 
Vector Machines: An Application to Face Detection. In: 
Proceedings IEEE Conference on Computer Vision and Pattern 
Recognition, San Juan, pp. 130-136. 
Papageorgiou, C. P., Evgeniou, T. & Poggio, T., 1998. A 
Trainable Pedestrian Detection System. In: Intelligent Vehicles, 
Stuttgart, Germany. 
Papageorgiou, C. P., Oren, M. & Poggio, T., 1998. A General 
Framework for Object Detection. International Conference on 
Computer Vision, Bombay, India. 
Poggio, T. & Shelton, C. R.,1999. Machine Learning, Machine 
Vision, and the Brain. AZ Magazine 20: pp. 37-55. 
Rabbani, M. & Joshi, R., 2002. An overview of the JPEG 2000 
still image compression standard. Signal Processing: Image 
Communication 17: pp. 3-48. 
Sarkar, S. & Boyer, K., 1993. Perceptual Organisation in 
Computer Vision: A Review and a Proposal for a Classificatory 
Structure. JEEE Transactions on Systems, Man and Cybernetics 
23(2): pp. 382-399.