The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part BL Beijing 2008 
Image Size 
18427 * 23136 pixels 
Resolution 
0.9*1.8 m 
Topography 
Flat 
imaging mode 
HIMAGE 
Time of Acquisition 
2008/03/15/11:05:03.5 
Product Type 
ssc 
Table 2. Basic information on the Shanghai COSMO-SkyMed 
imagery. 
4. THE TEST OF THE UNBIASED RPC ESTIMATOR 
Based on the RPC conclusions of Tao, the third order RPC 
model cases with an unequal denominator achieve the best 
accuracy when the rigorous sensor model is available (Tao et al., 
2001). This paper will only test the fitting of a third order RPC 
with an unequal denominator. 
2006 Annual Conference Proceedings, URL: http://socetset.com 
[Accessed: 1 Jan 2007]. 
Dolloff, J., 2004. Replacement Sensor Model Tagged Record 
Extensions Specification for NITF 2.1 APPENDIX C , URL: 
http://l 64.214.2.51/ntb/coordinationitems/ 
RSM%20TRE%20Appendiix%20C_July_23_04.pdf [Accessed: 
1 Jan 2007]. 
Kampes B., 2005. Delft Object-oriented Radar Interferometric 
Software User's manual and technical documentation, URL: 
http://enterprise.lr.tudelft.nl/doris/tAccessed: 1 Jan 2007]. 
Xinzhou Wang, Dingyou Liu, Qianyong Zhang, and Hailan 
Huang, 2001. The Iteration by Correcting Characteristic Value 
and Its Application in Surveying Data Processing, Journal of 
Heilongjiang Institute of Technology, 15(2):3—6. 
According to the thinking explained above, RPC fitting based Grodecki J., 2001. IKONOS stereo feature extraction-RPC 
on the rigorous sensor model was performed ( Kampes, 2005). approach, Proceedings of ASPRS, Conference, 2001, St. Louis 
These results are based on an established 3D Grid GCPs ( on cdrom). 
consisting of 15 elevation layers, each with 200 pixel * 200 
pixel grid points. The results are listed below. Danchao Gong and Yongsheng Zhang, 2003. The Solving and 
Application of Rational Function Model, Journal of Instituteof 
5. CONCLUSIONS Surveying and Mapping, 20(1 ):39-42. 
The RPC model has recently raised considerable interest in the 
photogrammetry and remote sensing community. The RPC is a 
generalized sensor model that is capable of achieving high 
approximation accuracy in comparison to the rigorous sensor 
models. In this paper, the unbiased RPC estimator is derived. 
The unbiased estimator is proposed to strengthen the solutions 
when the condition of the normal equations is poor. Based on 
numerous tests with terrain-independent scenarios, the 
following conclusions can be draw. 
The RPC model can achieve an approximation accuracy (1% 
pixel RMS ) that is extremely high for satellite data. The results 
support the view that the the RPC model, can be used as a 
replacement for sensor models for photogrammetric restitution. 
The RPC model is suited not only for push-broom imagery but 
also for high resolution SAR imagery. 
Check points 
Control points 
Y 
X 
planarity 
Y 
X 
planarity 
max 
RMS 
max 
RMS 
max 
RMS 
max 
RMS 
max 
RMS 
max 
RMS 
0.4 
0.1 
2.6 
1.6 
2.6 
1.6 
0.5 
0.1 
2.2 
1.2 
2.2 
1.2 
Table 3. The accuracy of the 3rd order RPC with unequal 
denominator on the Beijing TerraSAR-X imagery (lO^pixel). 
Liang-Chien Chen, Tee-Ann Teo, and Chien-Liang Liu, 2006. 
the Geometrical Comparisons of RSM and RFM for 
FORMOSAT-2 Satellite Images, Photogrammetric Engineering 
and Remote Sensing, 72 (5): 573-579. 
OGC (OPEN GIS CONSORTIUM), 1999. The OpenGIS 
abstract specification-topic 7: earth imagery, URL: http: 
www.opengis.org [Accessed: lJan 2007]. 
SPOT Imagery, 2002. SPOT SATELLITE GEOMETRY 
HANDBOOK, URL: http://ftp.spot.com [Accessed: 1 Jan 2007]. 
Tao, C. V. and Y. Hu, 2001. A comprehensive study of the 
rational function model for photogrammetric processing, 
Photogrammetric Engineering & Remote Sensing, 67(12): 
1347-1357. 
Toutin, T., 2004. Review Paper: Geometric processing of 
remote sensing images: models, algorithms and methods, URL: 
http://www.ccrs.nrcan.gc.ca [Accessed: 1 Jan 2007]. 
Guo Zhang, 2005. Rectification for High Resolution Remote 
Sensing Image Under Lack of Ground Control Points, Ph.D. 
Dissertation, Wuhan University. 
ACKNOWLEDGEMENTS 
Check points 
Control points 
Y 
X 
planarity 
Y 
X 
planarity 
max 
RMS 
max 
RMS 
max 
RMS 
max 
RMS 
max 
RMS 
max 
RMS 
1.5 
0.4 
-8.5 
2.9 
8.5 
3.0 
1.8 
0.4 
-9.6 
3.0 
9.7 
3.1 
Table 4. The accuracy of the 3rd order RPC with unequal 
denominator on the Shanghai COSMO-SkyMed imagery (10‘ 
4 pixel). 
REFERENCES 
de Venecia, K. J., F. Paderes, and A.S. Walker, 2006. Rigorous 
Sensor Modeling and Triangulation for Orbview-3, ASPRS 
Thanks are due for the support from the Natural Science Fund 
of P. R. China (No. 40601084 and No. 40523005), the Open 
Research Fund Program of State Key Laboratory of Satellite 
Ocean Environment Dynamics (No. SOED0602), Open 
Research Fund Program of the Geomatics and Applications 
Laboratory, Liaoning Technical University, open research 
subject of Key Laboratory of Geo-informatics of State Bureau 
of Surveying and Mapping (Project No. A1721), and China 
International Science and Technology Cooperation Project: 
High-Resolution Stereo Mapping Satellite: Field Geometric 
Calibration and Application (Project No. 2006DFA71570), and 
Commission of Science Technology and Industry for National