net 
the 
sper 
1ded 
ition 
nts, 
lo 
heet 
the 
red. 
not 
5. CHANGE DETECTION 
Investigation is in progress to introduce certain 
degree of automation into the identification of 
regions of possible changes and support a manual 
or semi-automated feature extraction scheme 
using multi-temporal digital orthophotos. The 
change detection operation can be further benefit 
by using multi-temporal digital orthophotos to 
further automate the identification of regions of 
possible changes and support a manual or semi- 
automated feature extraction scheme. For 
example, after performing histogram equalization 
and histogram matching on the temporal images 
covering the same area, it is possible to detect 
changes by edge detection, image subtraction and 
noise removal (Peled, 1994). These image 
variations are then examined by the operator as 
potential areas of change. It should be pointed out 
that simple radiometric differences between the 
temporal orthophotos do not necessarily represent 
areas of change. Digital elevation models and 
digital surface models will have a role to play in 
the radiometric matching process of multi- 
temporal orthoimages by providing geometric 
constraints. The spatial and attribute knowledge of 
the "old" database information can also support the 
change detection operations by matching the 
database objects, (Servigne and Laurini, 1994), 
with the image objects,  semi-automatically 
extracted from digital orthophotos (Agouris et al., 
1994). 
6. CONCLUSIONS 
The use of digital orthophotos for monoscopic 
revision of 1:50 000 database set has proven the 
potential and merits of this approach. From the 
technical point of view this approach facilitates 
the change detection and insures uniform and 
improved accuracy throughout the map data set. 
From the operational point of view this approach 
performs revision efficiently and effectively, 
allows time-stamping of the revised data and 
facilitates the quality control process. Further 
research is required towards improving change 
detection and feature extraction and classification 
techniques from multi-temporal orthoimages. 
ACKNOWLEDGMENT 
The contributions of G. Auger, J. Fairlie, S. Lemay 
and R. Samson are gratefully acknowledged. 
REFERENCES 
Agouris P., D. Stallmann, H. Li (1994) “Semi- 
Automatic Monoplotting on a Digital 
99 
Photogrammetric Station”, Inter. Archives of 
Photogrammetry and Remote Sensing, Vol. 30, 
Part 2, Com. Il, pp. 146-153. 
Armenakis C., A.M. Regan, 
"Softcopy photogrammetric 
National Topographic Mapping”, 
49, No.4, pp.433-443. 
A. Dow (1995) 
applications for 
Geomatica, Vol. 
Peled, A. (1994) "Revision of Digital Maps and GIS 
Databases", Inter. Archives of Photogrammetry 
and Remote Sensing, Vol. 30, Part 4, Com. IV, pp. 
268-272. 
Servigne S., R. Laurini (1994) "Detecting Changes 
in Geographic Database using Aerial Photos", Inter. 
Archives of Photogrammetry and Remote Sensing, 
Vol. 30, Part 4, Com. IV, pp. 318-324. 
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B4. Vienna 1996