bul 2004 
hly the same 
or analysis of 
ical situation, 
sed on digitized 
her than DSM. 
by taking for 
s results in à 
in better than 
iver valley the 
actual terrain 
  
  
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B4. Istanbul 2004 
height. In the wooded areas the DTM-derived grid points reflect 
the real terrain height. The rastermap with differences can 
always be used to establish the origin of each grid point in the 
hybrid DTM. 
3.4.Adding structure lines 
At places where slopes change abruptly (such as quarry edges, 
embankments and borders of water features) it is necessary to 
introduce additional linear elements. In the test these were taken 
from the data measured for the topographical map 1:10 000. 
Prior to their integration, their height was compared to the 
height suggested by the hybrid DTM. In figure 7 the results are 
given for the stereoplotted borders of water features for area 1. 
For the majority of the borders, their height is within 2m from 
the hybrid DTM. For the remaining borders, the hybrid DTM is 
several meters too high. This is mostly the case in places where 
no digitized contour lines were present and the DSM was too 
high due to vegetation. In these places it is necessary to 
incorporate the stereoplotted borders to correct the hybrid DTM. 
  
height difference (m 
cef 
-— ‘ ' + 1 1 n DEL A 
Ovo-1o0o045owoo^5mokobóNóooo 
  
v 
a 
© 
0 50 100 150, 
number of element vertices 
Figure 7. test area 1. Z-value derived from hybrid DTM minus z-value 
of stereoplotted borders of water features. 
[n the future, results from this kind of statistical analysis could 
be used to automate to a certain degree the incorporation of the 
stereoplotted elements as additional linear elements in the 
hybrid DTM. A fully automated procedure seems not very 
realistic and actually quite dangerous. At the moment it is still 
necessary to include the linear elements in a completely 
interactive way. 
4. CONCLUSIONS 
We have given an overview of the initial DEM situation at the 
NGI, the requirements for the development of a better DEM 
data layer and the constraints that have to be taken into account. 
We then proceeded with the general working strategy for the 
actual development of the DEM data layer, highlighting the 
need for recovery and upgrading of existing data, the need for 
flexibility, the need for a phased approach and an appropriate 
data structure. 
537 
A test was performed combining three available DEDSs using 
available software to construct a new DEDS which better 
reflects the terrain than any of the original DEDSs. Further 
testing should be performed to corroborate the positive results 
obtained and to heighten the degree of automation in the 
procedure. 
REFERENCES 
Ackermann, F. 1996. Report on the OEEPE Workshop on 
Application of Digital Photogrammetric Workstation “Some 
considerations about feature matching for the automatic 
generation of digital elevation models", Lausanne, Switzerland. 
http://phot.epfl.ch/workshop/wks96/art 3 4.html (accessed 23 
jan. 2003) 
  
Artuso, R., Bovet, S., Streilein, A., 2003. Report on ISPRS WG 
III/3 Workshop “Practical methods for the verification of 
countrywide terrain and surface models”, Dresden, Germany. 
http://www.isprs.org/commission3/wg3/workshop_laserscannin 
g/papers/Artuso_ALSDD2003.pdf (accessed 24 march 2004) 
  
  
Cory, M.J.,McGill, A., 1999, Report on the OEEPE Workshop 
on Automation in Digital Photogrammetric Production “DTM 
derivation at Ordnance Survey Ireland” , Marne La Vallée, 
France. http://phot.epfl.ch/workshop/wks99/5_2.html (accessed 
23 jan. 2003) 
  
Dupéret, A., 1999. Report on the OEEPE Workshop on 
Automation in Digital Photogrammetric Production "DTM 
edition in IGN France. An operational process to generate 
contour lines", Marne La Vallée, France. 
http://phot.epfl.ch/workshop/wks99/4 2.html (accessed 23 jan. 
2003) 
  
Haala, N., 1999. Combining multiple data sources for urban 
data acquisition. In: Photogrammetric Week '99. Wichmann, 
Heidelberg, pp. 329-339. 
McIntosh, K.,Krupnik, A., 2002. Integration of laser-derived 
DSMs and matched image edges for generating an accurate 
surface model. /SPRS Journal of Photogrammetry and Remote 
Sensing, Vol 56, pp.167-176. 
Mizukoshi, H., Aniya, M., 2002. Use of contour-based DEM's 
for deriving and mapping topographic attributes. 
Photogrammetric Engineering & Remote Sensing Vol.68, No.1, 
January 2002, pp..83-93. 
Reiss, P., 2002. Combination of Laser scanning, Digital 
Photogrammetry and Terrestrial Survey to generate high-quality 
DEMs. In: Proceedings Joint International Symposium on 
Geospatial Theory, Processing and Applications, July 8-12, 
2002, Ottawa, Canada. 
Ruiz, A., 2000. A countrywide TIN elevations database. In: 
International Archives of Photogrammetry and Remote Sensing, 
Vol XXXIII, Part B3, Amsterdam 2000, pp. 785-791. 
Vanommeslaeghe, J., 2001. Het opstellen van een digitaal 
hoogtemodel voor België. Internal report NGI Belgium