International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B4. Istanbul 2004
the points, also the operation codes (IV, DV, ...) together with
integer values indicating point id’s have to be coded.
Altogether, this is in the order of n which basically means that
all representations of an object can be transmitted for the price
of transmitting one.
Coding displacement is more demanding concerning the data
volume, as it requires the same number of coordinates, as the
points are only moved. However, the numbers are small, as the
movements of the points are typically very small compared to
the large coordinate values, that need double precision values.
Furthermore, as only changes are encoded, not the whole data
set has to be transferred in all scale-steps. Finally, also an
operation could be defined, that encodes the movement of an
object as a whole. During typification the objects are replaced
by new objects, i.e. completely new objects are created. Thus,
no incremental change from the old situation to the new one can
be done, which has the consequence that the full object
representation has to be created and hardly a reduction in
volume can be achieved.
6. CONTINUOUS GENERALIZATION
When a map representation is switched due to generalization,
this usually leads to a visible “popping” effect. Compared to
switching between different, fixed levels of detail, the use of
EGO's is already an improvement, since it gradually modifies
the polygon rather than just replacing it as a whole.
However, one can still improve on this. Intermediate states can
be defined which continuously change the object in response to
an EGO. For example, a "collapse extrusion" EGO (see Figure
5b) would be interpreted as “move extrusion until in coincides
with the main part, then change the topology accordingly”. We
term this approach continuous generalization as it effectively
allows to morph the object continuously from its coarsest to its
finest representation. It is realized by decomposing the
movement into a number of intermediate steps that give the
impression of smooth changes. For more details sce [Brenner &
Sester, 2003].
7. SUMMARY
An approach was presented to decompose changes in object
geometry into a small set of simple operations. These
operations can express the creation of objects as well as
iterative refinement of their shapes. This coding scheme was
used to represent different generalization levels of objects
efficiently. For different generalization operations it could be
shown, how this representation was generated. A comparison
concerning the storage and coding demands with respect to
representing the full geometry was made and it was shown that
a reduction in the amount of data to be transmitted by
approximately the factor n can be achieved. Besides
incrementally presenting the iterative changes in the geometry.
it was also shown that the changes can be animated, leading to
nearly invisible changes between the different representations
when changing the scale.
8. REFERENCES
Bertolotto, M., and Egenhofer, M., [2001], Progressive
Transmission of Vector Map Data over the World Wide Web,
Geolnformatica - An International Journal on Advances of
Computer Science for Geographic Information Systems, Vol. 5
(4), Kluwer Academic Publishers, pp.345-373
Brenner, C. and Sester, M. [2003]: Continuous Generalization
for Small Mobile Displays, International Workshop on Next
Generation Geospatial Information, October 19-21, 2003,
Cambridge (Boston), Massachusetts, USA.
Cecconi, A., Weibel, R. & Barrault, M. [2002]. Improving
automated generalization for on-demand web mapping by
multiscale databases. 70" International Symposium on Spatial
Data Handling, Ottawa, Canada.
Douglas, D. H. & Peucker, T. K., [1973]. Algorithms for the
reduction of the number of points required to represent a
digitized line or its caricature. The Canadian Cartographer,\0,
pp. 112-22.
Hoppe, H. [1996]. Progressive Meshes. Proceedings of
SIGGRAPH 96 (New Orleans, LA, August 4-9, 1996). In
Computer Graphics Proceedings, Annual Conference Series,
1996, ACM SIGGRAPH, pp. 99 — 108.
Kreveld, M. van [2001], Smooth Generalization for Continuous
Zooming, Proceedings of the ICC, Beijing, China, 2001.
Oosterom, P. van [1995]: *On-the-Fly" Map Generalization of
an Area Partitioning. In: GIS and Generalization, Methodology
and Practice. Editors J.C. Müller, J.P. Lagrange and R. Weibel.
Taylor & Francis, London, pages 120-132, 1995.
Sarjakoski, T, L.T Sarjakoski, L. Lehto, M. Sester, A. Illert, F.
Nissen, R. Rystedt, and R. Ruotsalainen [2002]: Geospatial
Info-mobility Services - a Challenge for National Mapping
Agencies. Proceedings of the Joint International Symposium on
"GeoSpatial ^ Theory, Processing and Applications”
(ISPRS/Commission IV/SDH2002), Ottawa, Canada, July 8-12,
2002, 5p, CD-ROM.
Sester, M. [2000]. Generalization Based on Least Squares
Adjustment. In: International Archives of Photogrammetry and
Remote Sensing, Amsterdam, Netherlands, Vol. XXXIII, Part
B4, pp. 931-938.
Sester [2004]: Optimizing Approaches for Generalization and
Data Abstraction, accepted for publication in: International
Journal of Geographic Information Science.
Thiemann, F. [2002], Generalization of 3D building data,
IAPRS Vol. 34, Part 4, *GeoSpatial Theory, Processing and
Applications", Ottawa, Canada.
ACKNOWLEDGEMENT
This research is part of the GiMoDig project, funded by the
European | Union, IST 2000, 30090, and the
VolkswagenStiftung.
1298
“In:
KEY WC
ABSTRA
Within th
Co-Invest
products
togramme
Earth. Th:
tegrated ii
Center (D
fying illur
ation. The
optimizati
scribed as
extended :
data as we
results cor
Automatic
surface is
graphy Gr
gator Scier
Camera) is
Express Oti
logical, mi
graphic stu
stitute of P
ter (DLR),
groups — is
ence data p
velopment
ic and GIS
of Mars. TI
ates for a vi
Having bee
Baikonur s;
Express spa
after ejectic
manoeuvrin
it was trans
Orbiter scie
down to Ea
lowing few
different re:
down-linkec
