tanbul 2004
as the ISAE
ible size of
ages.
ed on the
on: dll. of
| function
CESS.
1erated with
100m. The
asis of their
ed by auto-
le 1:26 000,
RMS Z = +
IM, taking
:d above is
to measure
ainous and
filtration of
eli as the
analysis of
'ogramming
ig systems
iccuracy of
ountainous
terrain.
T4
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part Bl. Istanbul 2004
Test region Percentage of pixels with accuracy better than Percent of bad
DLR-DTM im 2m 5m 10m correlation Bemarks
01 17.0 16.1 30.8 15.2 20.8
02 20.1 17.2 25.9 14.1 227
03 20.9 18.0 30.7 15.9 14.4
04 19.7 16.7 27.4 13.6 22.6
05-1 4.9 4.6 10.2 14.3 66.0 very bad correlation
05-2 8.0 7.1 16.1 13.8 55.0 very bad correlation
06 14.9 14.1 31.7 25.1 18.2
Table 5. Accuracy of DTM for the test regions (raw DTM 20m pixel size interpolated to reference DTM)
Test region (DLR-DTM)
Accuracy of DTM
03 04 031 05 2 06
RMSE +26 13
t24 4.2.5 + 14.2 t67 12.8
Table 6. The RMSE of DTM generated for different test regions from HRS SPOT 5
5. THE ACCURACY OF DTM GENERATED ON THE
BASIS OF PANCHROMATIC HRS SPOT 5 STEREO
IMAGES
Epipolar images and OV's have been generated for two stereo
images: 01-02 and 03-04. PCI Ortho Engine has been used for
matching and DTM generation. Matching and DTM generation
have been performed for both stereo scenes with pixel 20m as
16-bit image data in GeoTiff format. DTM have been extracted
only for the 7 test regions. Next, the DTM have been
interpolated to the pixel size as given in reference DTM data.
Generated DTM has neither been verified in order to eliminate
gross errors nor edited. The result of comparison tests is shown
in table 5.
For five of the test regions 60% of results of DTM is in the
range from Im to 5m. The detailed analyses of histograms of
images representing DTM and results are shown in table 6.
6. CONCLUSION
The use of well-elaborated methodology of correlation
measurement on the panchromatic HRV SPOT 4 images
permits to achieve DTM with the accuracy similar to the
accuracy of raw DTM generated on the basis of the
panchromatic HRS SPOT 5 images. SPOT 4 Pan stereo
data can be used for generation of contour lines with 20
m interval and orthophotomaps in scale 1:25 000.
Technology for generation of DTM for orthophotomaps
in scale 1:25 000 from HRV SPOT 4 Pan stereo data can
be used for mapping and up-dating topographic maps in
Africa and in Europe including Poland. This technology
can be also adopted for new satellite data like IKONOS
and QuickBird stereo data.
To increase the accuracy of raw DTM generated on the
basis of the panchromatic HRS SPOT 5 images the use of
suitable editing and filtering techniques as well as
additional supplementary measurements (breaklines,
etc.), especially for mountainous and hilly terrain is
recommended.
The use of proper methodology for correlation measurement on
the basis of the panchromatic HRS SPOT 5 and for HRV SPOT
4 stereo images will permit to generate DTM with accuracy in
elevation better than 2m.
REFERENCES
References from Journals:
Cooper, P. R.. Friedman, D. E, Wood, S. A. 1937.
The Automatic Generation of Digital Terrain Models
from Satellite Images by Stereo. Acta Astronautica,
15 (3), pp. 171 — 180.
Day, T. Muller, J. P., 1988. Quality Assessment
of Digital Elevation Models Produced by Automatic
Stereo — Matching from SPOT Images Pairs,
Photogrammetric Record, 12 (72), pp. 797 — 808.
El-Manadili, Y., Novak, K., 1996. Precision Rectification
of SPOT Imagery Using’ the Direct Linear
Transformation Model. Photogrammetric Engineering
and Remote Sensing, Vol. 62, pp. 68-70.
Krupnik, A. 2000. Accuracy Assessment
of Automatically Derived Digital Elevation Models from
SPOT Images. Photogrammetric Engineering and
Remote Sensing, Vol. 66 (8), pp. 1017-1023.
Toutin, Th., 1999. Stereo-Mapping with SPOT-P and
ERS-1 SAR Images. International Journal of Remote
Sensing, 20 (15).
Westin, T., 1990. Precision Rectification of SPOT
Imagery. Photogrammetric Engineering and Remote
Sensing, Vol. 56 (2), pp. 247-253.
References from Books:
Linsenbarth, A., 1987. Satelitarne systemy teledetekcyjne.
Wydawnictwa Politechniki Warszawskiej.
Kaula, W. M., 1966. Theory of Satellite Geodesy.
References from Other Literature:
Angleraud, C. Becek, K. Trinder, J. C. 1992. DEM
determination from SPOT. In: The International Archives
of the Photogrammetry, Remote Sensing and Spatial
Information Sciences, Washington D. C., USA, Vol. 29/B4,
pp. 969-973.