tion 
tion 
hing 
) be 
are 
raw 
that 
are 
lese 
of 
onal 
ject 
the 
phic 
oli- 
and 
nbol 
the 
sted 
sult 
hich 
tion 
phic 
vely 
ass 
the 
er- 
the 
is 
are 
preserving the local relations. The combination 
gets neseccary when the distance between 
two buildings is smaller than the given 
threshold and according to the scale the 
original gap should no longer be presented 
in the derived DCM. 
Two variants are implemented in the software 
module for the combination of buildings: 
a relatively small building is moved towards 
a larger building and two objects are joined 
if both buildings are to be kept placed 
according to the ground plan because of their 
considerable spatial extension. 
As a result from these computation steps 
you get the buildings for the final DCM (see 
figure 6). 
T. RECOGNITION OF CARTOGRAPHICAL CONFLICTS 
The preceding descriptions explained the 
standalone processing of the single feature 
classes  ^traffic roads”: “and buildings’. 
Furthermore graphical conflicts in presentation 
of the generalized and symbolized objects have 
to be recognized and subsequently removed. 
The local interactions among different feature 
classes have to be considered in the way that 
objects of low priority must be adjusted 
to all other neighbouring objects of higher 
hierarchy. Displacement and homogenization, 
two important steps in generalization come 
to fruition here. 
The symbolization especially of the roads 
brings about a widened representation for 
the DCM and the displacement of buildings 
gets absolutely necessary. Objects from the 
feature class “buildings” of lower priority 
have to be moved or aligned to the traffic 
road symbols. The 1IfK’s generalization 
software offers a special component for 
automatic recognition of graphical conflicts 
that have to be modified. The following 
removal of conflicts has to be carried out 
interactively by the GIS-user in charge. The 
resulting DCM can be presented as soft- or 
hardcopy after the final graphic layout. 
8. CONCLUSIONS 
The aim of developments at the Institute of 
Cartography (IfK) at Hannover University is to 
provide robust computer-assisted software 
tools for generalization and to force the 
integration of these modules into complex 
GlS-mainframes. The request for appropriate 
visualization tools in GIS is expected to grow 
extremely in future as a versatile and flexible 
use of spatial GlS-data is desirable. 
The described generalization modules for 
generalization are nowadays still available 
with the IfK's generalization software 
CHANGE for large scale applications of 
commercial GlS-users. The IfK’s solutions 
for displacement exist in several prototype 
667 
programs and are an important part of the 
actual research in the field of computer- 
assisted generalization at the IfK. Furthermore 
modern computer technologies as for example 
the use of exclusively object-oriented or 
rule-based methods and procedures have 
to be established in computer-assisted 
generalization in future (Grünreich, Powitz, 
Schmidt, 1992). 
9. References 
Brassel, K.; Weibel, R. 1988: A review and 
conceptual framework of automated map 
generalization.- International Journal of 
Geographical Information Systems Vol. 2, No. 3. 
Buttenfield, B.; McMaster R.B. 1991: Map 
Generalization.- Longman, London, GB. 
Goodchild, M.; Maguire D.J.; Rhind, D.W. 1991: 
Geographical Information Systems.- Longman, 
London, GB. 
Grünreich, D. 1991: Introduction to Session 
"Computer-Assisted Cartographic  Genera- 
lization".- GlS-Konferenz, Brno, CSFR. 
Grünreich, D.; Powitz, B.M.; Schmidt, C. 1992: 
Research and Development in  Computer- 
Assisted Generalization of Topographic 
Information at the Institute of Cartography. 
Proceedings, EGIS 1992, München, Germany. 
Jüger, E. 1987: Computer-Assisted Symbo- 
lization by Raster Data Processing.- NaKaVerm 
No.46, Series ll, Frankfurt a.M. 1987, Germany. 
Lichtner, W. 1979b: Computer- Assisted 
Processes of Cartographic Generalization in 
Topographic Maps.- Geo-Processing 1, 1979. 
Meyer, U. 1987: Computer-Assisted Genera- 
lization of Buildings for Digital Landscape 
Models by Classification Methods.- NaKaVerm 
No.46, Series ll, Frankfurt a.M. 1987, Germany. 
Muller, J.- C. 1989b: Theoretical Considerations 
for Automated Map  Generalization.- ITC 
Journal 3/4, Netherlands. 
Powitz, B.M. 1990: Automationsgestützte Gene- 
ralisierung - Voraussetzungen, Strategien, 
Losungen. Kartographische Nachrichten 3/90, 
pp97 ff, Bonn, Germany. 
Powitz, B.M. 1991: Reflections on Practical 
Computer-Assisted Generalization. GIS- 
Konferenz 1991, Brno, CSFR. 
Powitz, B.M.; Schmidt, C. 1991: Aspects of 
Computer-Assisted Generalization for Large 
Scale Maps. ICA-Conference, Bournemouth, GB. 
Staufenbiel, W. 1973: Zur Automation der 
Generalisierung topographischer Karten mit 
besonderer Berücksichtigung groBmafBstübiger 
Gebdudedarstellungen.- WissArbUH Nr. 51, 
Hannover, Germany.