The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part B4. Beijing 2008
Number of
RMSE
RMSE
RMSE
RMSE
GCLs
(P)
(a)
(X)
00
fi
6
0.56m
0.049 0
0.91m
1.26m
parameters
10
0.34m
0.052°
1.03m
1.06m
15
0.30m
0.052°
1.03m
1.04m
8
parameters
6
0.50m
0.061°
1.10m
1.34m
10
0.48m
0.066°
1.17m
1.03m
15
0.47m
0.066°
1.15m
1.03m
DLT
10
0.94m
0.042°
2.16m
1.73m
15
0.90m
0.045°
1.77m
1.01m
Table 1. RMSE for check-lines and check-points
In order to test the stability of the estimated parameters, another
experiment is conducted using the 6 parameters transformation
model with different sets of GCLs. For each experiment, the
transformation parameters are computed. Results in table 2
shows the minimum, maximum, and mean values of the
calculated transformation parameters. In addition, table 2 shows
the transformation parameters calculated using the point-based
6 parameters transformation model with all the 12 GPS points
used as GCPs. The table shows that the differences between the
mean values of the transformation parameters and the point-
based transformation parameters are insignificant. Table 3
shows the statistics for the differences in the parameters of the 8
parameters transformation model using different sets of GCLs
and the values computed using the point-based transformation
model.
Min
Max
Mean
Point-
based
al
0.8840
0.8811
0.8827
0.8827
a2
-0.0020
-0.0075
-0.0041
-0.0043
a3
0.6913
-1.3231
0.0163
0.000
a4
0.0529
0.0514
0.0522
0.0520
a5
0.9150
0.9131
0.9138
0.9146
a6
0.1972
-0.4049
0.0319
0.000
Table 2. Statistics for the differences in the parameters of the 6
parameters transformation model using different sets
of GCLs
Min
Max
Mean
Point-
based
al
0.881
0.883
0.882
0.882
a2
-0.007
-0.001
-0.004
-0.004
a3
-1.283
0.685
0.025
0.431
a4
0.051
0.530
0.052
0.052
a5
0.913
0.914
0.913
0.914
a6
-0.516
0.332
0.023
0.074
a7
0.000
0.000
0.000
0.000
a8
0.000
0.000
0.000
0.000
Table 3. Statistics for the differences in the parameters of the 8
parameters transformation model using different sets
of GCLs
5. CONCLUSIONS
This research presents the potential of using straight lines to
rectify a single panchromatic IKONOS image. The research
showed the process of developing the line-based 2D
transformation models. In addition, the research investigates the
use of the line-based 6-parameters, 8-parameters, and DLT
transformation models to rectify panchromatic IKONOS images.
Results showed less than 1.5 meters RMSE in the horizontal
direction using 6 to 15 GCLs. Moreover, the results showed that
the computed parameters are stable and equivalent to the
parameters computed using the point-based transformation
models and the differences are insignificant. The results suggest
the use of linear features to rectify IKONOS images and other
high-resolution satellite images such as QuickBird. This will
allow reducing the number of ground control points and will
eventually reduce the time and cost of the surveying effort.
REFERENCES
Barakat, H.F., Emam, H.A., and Abdel-Wahab, M.S., 2004.
Assessment of a developed combined point/line-based
projective equations”, The International Archives of
Photogrammetry, Remote Sensing, and Spatial Information
Sciences, Istanbul, Turkey, Vol. XXXV, Part 3.
Blasco, F., Bellan, M.F., Barbaroussi, V., and Miliaresis, G.,
2004. Use of orthophotos as ground truth IKONOS image
processing”, The International Archives of Photogrammetry,
Remote Sensing, and Spatial Information Sciences, Istanbul,
Turkey, Vol. XXXV, Part 4.
Di, K., Ma, R., and Li, R., 2003. Rational functions and
potential for rigorous sensor model recovery. Photogrammetric
Engineering and Remote Sensing, Vol. 69(1), pp. 33-41.
Dial, G. and Grodecki, J., 2005. RPC replacement camera
models. Proceedings of the ASPRS 2005 Annual Conference,
Baltimore, Maryland, (available on a CD ROM).
Fraser, C. and Hanley, H.B., 2003. Bias compensation in
rational functions for IKONOS imagery. Photogrammetry
Engineering and Remote Sensing, Vol. 69(1), pp. 53-57.
Gruen, A., Li, Z., and Eisembeiss, H., 2005. 3D processing of
high resolution satellite imagery. Proceedings of the ASPRS
2005 Annual Conference, Baltimore, Maryland, (available on a
CD ROM).
Habib, A. and Al-Ruzouq, R., 2005. Semi-automatic
registration of multi-source satellite imagery with varying
geometric resolutions. Photogrammetric Engineering & Remote
Sensing, Vol. 71(3), pp. 325-332.
Habib, A., Morgan, M., Kim, E.M., and Cheng, R., 2004.
Linear Features in Photogrammetric Activities. The
International Archives of Photogrammetry, Remote Sensing,
and Spatial Information Sciences, Istanbul, Turkey, Vol.
XXXV, Part 2.
Kaczynski, R., Majde, A., and Ewiak, I., 2004. Accuracy of
DTM and Ortho Generated from IKONOS Stereo Images. The
International Archives of Photogrammetry, Remote Sensing,
and Spatial Information Sciences, Istanbul, Turkey, Vol.
XXXV, Part 4. <
Kumar, M. and Castro O., 2001. Practical aspects of IKONOS
imagery for mapping. Proceedings of the 22 nd Asian Conference
on Remote Sensing, Bangkok, Singapore, 5-9 November, 2001.
Lutes, J., 2004. Accuracy analysis of rational polynomial
coefficients for IKONOS imagery”, Proceedings of the ASPRS
1308