The major aerial camera manufacturers have announced
digital aerial cameras with an information content similar to
the existing large size aerial film cameras. This will speed
up the photogrammetric workflow by saving the time for
film scanning, which is not negligible.
Laser scanning has become a standard tool for pho-
togrammetry with, new possibilities in forest and coastal
areas and also cities. The problem of the removing of
objects not belonging to the bare soil has been solved par-
tially but still has to be improved, as also the correct
matching of neighbouring flight strips.
Interferometric Synthetic Aperture RADAR -systems
(INSAR) can determine the Digital Elevation Model (DEM)
of large areas in a short time. Depending upon the wave-
length used, usually only the visible surface will be deter-
mined, which means that the effect of the vegetation and
buildings has to be removed. With the Shuttle RADAR
Topographic Mission (SRTM) from the Space Shuttle, a
large part of the land area has been imaged by INSAR.
Based on this, in a few years a homogeneous DEM will be
available, though the absolute height accuracy is only
about 16 meters.
Platform technology has been improving so as to make
spacecraft with lower weight and less power. Another
important development is to make the platform agile
enough to point at any point of interest. Flight navigation
by GPS has become a standard for satellite position deter-
mination. The ending of GPS Selective Availability has
improved the accuracy, without relative positioning.
Aerial cameras are being supported by relative kinematic
GPS-positioning and also Inertial Measurement Units (IMU),
allowing a direct sensor orientation. Direct sensor orienta-
tion has come into practical use. An accuracy of 20cm on
the ground can be reached.
With a number of earth observation cameras having simi-
lar characteristics operating from different platforms, it will
be necessary to integrate the data from different sensors
for proper interpretation. Time series will be essential for
certain applications, such as NDVI changes for proper
understanding of the natural phenomenon happening,
extraction of yield etc. This calls for a reliable radiometric
calibration programme before launch and in orbit. Simi-
larly, complex high resolution systems with along-track
stereo capability, will place high demands on geometric
calibration. This means that
(a) Procedures for radiometric and geometric calibration
(both laboratory and inflight) of digital camera systems
should be investigated with the objective of deriving
recommendations for standards.
(b) Large area test sites with highly accurate ground truth
should be established and maintained world wide to
ensure reliable verification of geometric lab-derived
calibration parameters of different digital camera sys-
tems.
(c) For vicarious calibration, ground test fields with known
spectral reflectance should be established world-wide.
ISPRS
— m
(d) Field campaigns for radiometric intercalibration of data
of different satellites should be conducted.
In the context of the increasing relevance of, and depend-
ence upon, fusion of data from various sensors, it is nec-
essary to identify a set of sensor parameters, which, if
standardised globally, would ensure maximal retrieval of
information. In order to make inter-comparison more
meaningful it is also necessary to formulate standard pro-
cedures for measurement of these parameters.
IN summary, the past few years have witnessed a number
of innovations in the field of sensors and platforms which
could make the remotely-sensed imagery useful for sev-
eral new tasks. With a large number of sensors it is neces-
sary to investigate independently from the manufacturers
information performance parameters. This calls for a uni-
form code for specifying sensor parameters and evalua-
tion methodology. This should be one of the major tasks of
the next commission.
4. Resolutions
Standardisation of Sensor Parameters
Resolution 1.1
The Congress
Noting
- that a number of earth observation sensors with sim-
ilar capabilities are available and planned by various
space agencies/manufactures that users will have to
use data from more than one sensor for their specific
applications
Recognising
- that such usage requires good understanding of the
sensor parameters
- that there is no uniform way of specifying sensor
parameters
Recommends
- the generation of a common set of parameters to be
specified for each camera / sensor, in conjunction with
manufacturers
Radiometric and Geometric Calibration
Resolution 1.2
The Congress
Noting
- that the number of high resolution, multispectral and
hyperspectral imaging sensors in space is increasing
- that radiometric calibration of this data is essential for
quantitative environmental and ecological research
with multispectral image data
- that data from various sensors have to be used for
long-term observations and for change detection
- various test fields exist or are planned for calibrations
- that the accuracy potential of high resolution digital
camera systems in space is better than 10 m and
thus appropriate for the production or updating of
topographic (image) maps of scale 1:50,000 and
larger
- that accurate and reliable geometric calibration
parameters of those digital camera systems are a
precondition to taking full advantage of their accuracy
potential to produce high quality photogrammetric
products, such as DEM, orthoimages, etc.
International Archives of Photogrammerty and Remote Sensing. Vol. XXXIII, Part A. Amsterdam 2000. — — uU AA