3.
on
9.
on
ots
er,
ed
ns
5
ad
S-
9-
jon
00,
90,
ni-
90,
ata
ng,
EVALUATION OF SPOT DATA FOR TOPOGRAPHIC MAP REVISION AT
THE NATIONAL LAND SURVEY OF SWEDEN
Bo Malmstróm
National Land Survey
S - 801 82 GÄVLE
SWEDEN
Agneta Green Engberg
Swedish Space Corporation
P.O. Box 4207, S - 171 04 SOLNA
SWEDEN
Presented at ISPRS, Washington D.C., August 2-14, 1992, Commission IV
ABSTRACT:
In a project carried out jointly by the National Land
Survey (NLS) and the Swedish Space Corporation
(SSC), SPOT data has been tested for map revision
purposes. The main application focused on is a frequent
revision of the new topographic map series at the scale
of 1:50 000. The map features of current interest are
roads, clearings and built-up areas. Comparison has
been made with aerial photos. Results show no or minor
differences in interpretations of roads and clearings
from aerial photos, SPOT PAN or SPOT XS regarding
the different accuracy measures: completeness, shape
and geometric accuracy. For urban areas, it is possible
to interpret the delimitations of new areas for revision.
The result also indicate that training is needed to fully
take advantage of the multispectral images of SPOT.
Costs and accessibility for SPOT data vs aerial photos
are evaluated. Current work involves digital revision
methods with SPOT data at NLS.
KEY WORDS: Topographic Map Revision, SPOT Data,
Satellite Remote Sensing Application, Image Interpre-
tation, Digital Revision Methods.
1. BACKGROUND
1.1 The new topographic map of Sweden
In 1990, the first map sheets of the new topographic
map series in Sweden at the scale of 1:50 000 was
published. The maps are produced by the National
Land Survey of Sweden (NLS, in Swedish: Lantmäteri-
verket) in Gävle. The new map series is produced
completely by digital methods. All objects are stored in
digital databases. The maps also contain an increased
amount of feature types compared to earlier maps, e.g.
clearings, bedrock outcrops and cultivated areas. The
new map is based on aerial photos and field control.
Introducing clearings in the map requires a rather
frequent revision interval. The planned revision cycle of
the new map series is: a frequent revision every 5 years
involving a few map feature types; an extended revision
every 10 years concerning a larger set of feature types;
and finally production of a new map including field
work after 20 years.
557
Revision here implies image interpretation without field
control. What type of image material, i.e. satellite
imagery or aerial photos from different altitudes, to be
used for these future revisions has not yet been decided.
1.2 The SPOT satellite
The French SPOT earth resources satellite carries high
resolution sensors producing data and images of the
earth. The panchromatic (PAN) mode has a pixelsize of
10 meters and the multispectral (XS) mode, in the
visible and near-infrared part of the electromagnetic
spectrum, has a pixelsize of 20 meters. Sweden coopera-
tes with France in the SPOT project, and the two main
receiving stations are situated in Toulouse, France and
in Kiruna, in the far north of Sweden.
SPOT is a near-polar orbiting satellite covering the
earth in 26 days. SPOT carries two sensors which can,
besides nadir viewing, be programmed to point off-nadir
to the side of the current satellite track. With nadir
viewing a certain satellite "scene" is revisited every 26
days. The further away from the equator, the closer the
distance is between the satellite tracks, causing overla-
ping scenes. In central Sweden, for instance, a particu-
lar area on the ground is covered approximately every
13 days with nadir viewing. Using the pointing capabili-
ty, however, the revisit cycle dramatically increases to
a matter of day or days, all depending on the latitude
and the pointing angle of the sensor.
2. AIM OF THE STUDY
The main objective for this study, carried out in coope-
ration between the National Land Survey and the
Swedish Space Corporation, was to find out if SPOT
images could be a used in the interpretation process for
the future frequent revision of roads, clearings and
built-up areas in the new topographic map at the scale
of 1:50 000.
The alternative method for this revision type would be
using aerial photos from high altitude (9 200 meters,
negative scale 1:60 000) or very high altitude (13 200
meters, negative scale 1:150 000).
Costs and accessibility were also to be investigated.