International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part BI. Istanbul 2004
4. Improvement of census cartography; study performed by
GeoTerralmage (South Africa) on Pretoria site (South
Africa)
The second SAVP domain is related to Natural Disaster
Management.
Earth Observation is more and more frequently used through
the International Charter “Space and Major Disasters”.
Improved SPOT 5 resolution is the key element improving
significantly satellite images capability to provide civil security
with useful information. One SAVP study is dedicated to this
topic :
5. Geological hazard identification and risk assessment for
humanitarian aid; study performed by UNOPS/UNOSAT
(Switzerland) on Matagalpa site (Nicaragua)
The third SAVP domain addresses Renewable Resources
(Agriculture and Forestry).
In such a large domain, Forestry is definitely the one that
highly benefits from 2.5 m resolution combined to multispectral
information. Indeed, previous SPOT images at 20 and 10 m
resolution were already used for mapping and monitoring forest
cuttings, but could not be used for precise forest management.
In Agriculture, benefits linked to resolution improvement are
obvious for crop monitoring in smaller parcels, and for very
specific topics such as control of European aids related to crop
surfaces, trees counting, or any linear units. Two SAVP studies
have been selected for this domain :
6. Boreal forest monitoring; study performed by Metria
Miljóanalys (Sweden) on Nássjó site (Sweden)
7. Agri-environmental measures control; study performed by
the JRC Ispra (Italy) on the Baden-Württemberg Zwie site
(Germany)
The last SAVP domain addresses — Three-Dimensional
Visualisation. This domain of higher interest both for civilian
and defence applications, and exploits the full range of SPOT 5
products : 2.5 m imagery and HRS DEM.
One SAVP study has been selected for addressing this new way
of visualising satellite images:
8. Three-dimension visualisation products for landing
approach simulation; study performed by the French Air
Force on Toulouse-Francazal site (France).
2.2 Technical objectives
Technical objectives aimed at by SAVP partners can be split
into two main orientations: exploitation of images through
visual expertise and automatic (or semi-automatic) processing
techniques improvement.
The majority of SAVP partners uses mainly human
interpretation for extracting information from satellite images.
Their objective in the framework of SAVP was to assess the
information content of SPOT 5 as compared to previous SPOT
and/or to metric aerial photography, and to assess which
enhancement filters may help experts to analyse the images.
A smaller number of SAVP partners are interested in automatic
or semi-automatic processing, in order to be able to process
quickly and self-consistently a whole SPOT image,
guaranteeing that the same methodology was used on every
pixel.
Of course, several partners combine both approaches (visual
interpretation and automatic processing), and have performed
tests and evaluations in the two areas.
2.3 SPOT 5 products acquisition and production
Nearly 30 SPOT 5 products were tested through the SAVP.
Most of the investigators got a multi-resolution SPOT 5 dataset
including the 10 m colour, the 5 m black and white (B&W)
resolution and the 2.5 m B&W resolution products.
The 2.5 m colour resolution product was produced and tested in
the frame of six investigations upon the eight ones,
demonstrating a great interest of the users for this new product
for many applications.
All SPOT 5 products were acquired through normal priority
programming services.
Some difficulties due to meteorological conditions combined
with the schedule of realisation of the SAVP were encountered
in two areas above the eight ones.
In northern Europe in the winter season, no SPOT 5 image was
available because of bad weather conditions for the agricultural
test in Badden- Württemberg area.
In the case of Guatemala, the area was very difficult to cover,
and the radiometric quality of the resulting image was medium
because of the presence of clouds and haze. The programming
of a SPOT 5 stereopair was unsuccessful for the coverage of the
sub-areas of interest, due to the presence of clouds. A Digital
Elevation Model extracted from SRTM sources was finally
used for the orthorectification.
On the contrary, in Manila, Philippines, a SPOT-5 2.5 m
resolution stereopair was finally acquired in October 2003
despite of difficult meteorological conditions.
A similar satisfying result was also obtained in Nicaragua
despite of unfavourable meteorological conditions.
3. RESULTS AND PERFORMANCES OF SPOT 5
3.1 Cartography, Cadastre, Land Use/Land Cover
mapping
Thanks to the DORIS instrument, which allows to locate the
spacecraft very precisely on its orbit, the absolute location
accuracy of SPOT 5 images is better than 50 m RMS, without
any ground control point.
The absolute location accuracy of SPOT 5 orthoimages,
rectified on the basis of the digital elevation model issued from
the SPOT 5 High Resolution Stereoscopic (HRS) instrument, is
15 m RMS in planimetry without any ground control point.
100
penis pecu re = =
€ location measures
rms SPOTS : 50 m. |
i
ml:
aleng track (meters)
-100
-100 -75 50 -25 0 25 50 75 100
across track (meters)
Fig. 1 : SPOT 5 location performances without any ground
control point - 2002
Inte
em
3.1.
It h
geol
con:
topc
The
that
con
spa
scal
ima
In t
was
it is
sing
The
plot
Oth
vari
plai
dev
In |
Cer
Ima
Ma
the
con
topi
Chi
The
and
sca
20€
All
sho
sca
and
cari
are
wit
eff
bee
esti
In
lea
sup
pre
Fer
stat
dif
obt
sca
pre
per
an
lati
var
bui
sur
