CIPA 2003 XIX th International Symposium, 30 September - 04 October, 2003, Antalya, Turkey
Figure 8 shows how there is no obvious change in the Stupinigi
test site that would require any updating of the cartography, if
some agricultural boundaries are excluded
5. THEMATIC MAPS DERIVED FROM
PANSHARPENED IMAGES
An added value of high resolution satellite images is due to the
availability of different spectral bands, whose exploitation can
be useful for map updating and cultural site management.
Remote sensing classification techniques, or simple visual
image interpretation, can be used to derive auxiliary information
both for the updating (allowing a better discriminations between
the objects) and for thematic map generation devoted to cultural
site management. Information about the territorial/urban context
can be derived from such data. The lower geometric resolution
of the multispectral bands suggests a classification approach
that is based on pan-sharpened bands.
The pan-sharpening procedure generates a multispectral image,
starting from a low resolution multispectral image (XS) and a
high resolution panchromatic. In this case, due to the lack of
correspondence between the spectral ranges of the panchromatic
and the multispectral bands which results in radiometric
anomalous pan-sharpened bands, a self-developed procedure
was adopted. This procedure maintains the radiometric
differences between the original multipsectral bands and the
pan-sharpened ones (due to the previously mentioned problem)
within an acceptable range, paying the price of a light
smoothing of the data. The group is still working on this aspect.
6. CONCLUSIONS
This study leads us to conclude that the map updating problem
based on high resolution satellite images must be considered
from two different points of view:
• the geometric/positioning one;
• the information content one.
It has been shown that Spot 5 images have an information
content that is suitable for a 1:10000 scale map; unfortunately it
has been demonstrated through positioning tests that the
reachable accuracy it is not sufficient for that scale.
Improvement can be obatined:
• using a rigorous sensor model approach (we are
currently working on this);
• collecting the GCPs on larger scale maps or through
GPS survey (GPS measurement campaign is currently
being projected)
Quickbird orthoprojected images instead present an accuracy
that is suitable for a 1:10000 scale map. In this case buildings
and relief displacements have to be considered: this problem
can be dealt with by using a Dense DEM (DEM + height
building information) while orthoprejecting. A DDEM could be
generated trhough aerial laser scanner acquisition or dedicated
softwares for numerical cartography processing (if height layer
is available).
According to technical rules adopted for the production of the
used 1:10000 regional reference cartography, updated entities
should be characterized by a quality code that specifier the
updating method and source (orthoprojected satellite images).
The solution of the correct orientation of along track stereopairs
(i.e. Eros Al) could encourage map updating users towards map
production.
Finally it has been shown that high resolution satellite images
represent an economic and reliable solution to monitor cultural
heritage sites according to the scale map requider from the
customer.
REFERENCES
Boccardo, P., Borgogno Mondino, E., Giulio Tonolo, F., 2003,
Urban areas classification tests using high resolution pan-
sharpened satellite images, Urban 2003 proceedings, Berlin
Lingua, A., Borgogno Mondino, E., 2002, High Resolution
Satellite Images Orthoprojection Using Dense DEM, Spie 2002
proceedings, Creete
Boccardo, P., Borgogno Mondino, E., 2001, Metric Quality
Evaluation of Satellite High Resolution Images in Urban Areas,
Urban 2001 proceedings, Rome
Tao, C., Hu, Y., 2001, A comprensive study of the Rational
Function Model for Photogrammetric processing,
Photogrammetric engineering & Remote Sensing, December
2001
AA. VV., Capitolato d'oneri — Carta tecnica regionale
numerica alia scala 1:10000, Regione Piemonte, Servizio
Cartografico, 1994
ACKNOWLEDGEMENTS
The remotely sensed images that were used in this study were
acquired as part of a 2 year project that is cofmanced by
MURST (2001) entitled “The study of remotely sensed images
for environmental analyses”
