complete description of object. In this system, users can 
define various object types using object-oriented features. 
Because of the distributed architecture of the software system, 
it provides a multi-user measurement environment which 
facilitates data sharability and teamwork. After generating the 
database, data query and addressing can be expedited by using 
3D data visualization techniques. Figure 3 shows the area in 
Laval, Quebec, Canada, where the sample data set was 
collected in 1994. Figure 4 is a geo-referenced image and a 
view of the extracted 3D objects at the same location. 
  
am 
Figure 3: The test area 
7. CONCLUSIONS 
The purpose of developing this system is to enable fast, 
accurate and efficient object measurement from images from a 
moving vehicle. Instead of time-consuming measurements in 
the field, they can be performed using georeferenced stereo 
images in the comfort of the office. It makes the data 
recording procedure highly streamlined and automated. With 
strong referencing in the data processing, quality checking is 
very easy because objects are presented visually on-line. 
Assisted with automated image matching techniques, the 
interactive procedure requires much less effort, attention and 
professional skills. By integrating GIS databases in the system, 
the measurement data is effectively and consistently managed. 
Not only does the system create a comfortable surveying 
environment, it drastically reduces costs as well. 
8. ACKNOWLEDGMENTS 
The support of GEOFIT Inc., Laval, Quebec and the National 
Sciences and Engineering Research Council of Canada 
(NSERC) is gratefully acknowledged. 
9. REFERENCES 
El-Sheimy, N. and K.-P. Schwarz, 1993. Kinematic 
Positioning in Three Dimension Using CCD Technology, 
VNIS'93 conference, October 12-15, 1993. 
236 
  
OO 
  
Figure 4: Geo-referenced image and extracted 3D 
visualization 
Li, R., 1993. 3D GIS: A Simple Extension in the Third 
Dimension? Proceedings of ASPRS/ACSM convention’93, 
pp.218-227. 
Li, R., 1992. Building Octree Representations of 3D Objects 
in CAD/CAM by Digital Image Matching Techniques. Journal 
of Photogrammetric Engineering and Remote Sensing, Vol.58, 
No.12, pp.1685-1691. 
Li, R., M. A. Chapman and W. Zou, 1996. Optimal 
Acquisition of 3D Object Coordinates From Stereoscopic 
Image Sequences, Int. Arc. of ISPRS'96 Com. III, Vienna. 
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B2. Vienna 1996 
Li, R 
and J 
GeoS 
of Ca 
Nova 
Photc 
conve 
Schw 
Chap 
High 
the V