From Table 10, it can be seen that model parallax converges as
PPVG value increases; this is expected because more tie points
allows the adjustment to refine the EO such that tie point
residuals can be minimized. However it is also clear the value
of parallax is increasing while RMS is kept stable. Thus, higher
residual tie points are obtained when PPVG value increases,
and the solution is degraded. Thus, for Integrated Sensor
Orientation, using less tie points is better.
6. CONCLUSIONS
This paper reviews several uses of Integrated Sensor
Orientation. First, it can be used in conjunction with direct
georeferencing as a Quality Assurance / Control Tool to
calibrate boresight mis-alignment or any datum shift of the
DGPS/INS data. Second, it can be used with a high end direct
georeferencing system to achieve better accuracy in large scale
mapping projects, where the DGPS position error is not always
sufficient to meet the desired ground accuracy. Finally, when
flown in a block configuration with a minimum of 2 strips, ISO
can be used in conjunction with a lower accuracy system to
achieve similar performance of a high end system, all without
the use of any ground control point and hence lower the cost of
direct georeferencing system,
From the EO analysis presented in this paper, it is easy to
understand why a less accurate DGPS/INS system is not
suitable for high-accuracy direct georeferencing applications:
the ground error of a POS AV 310 is 2 times larger than the
POS AV 510, and parallax can be as large as 4 pixels RMS
versus | pixel for the POS AV 510. However, if a block of
photos with at least 2 strips is always available in the projects,
the advantages of Integrated Sensor Orientation can be
exploited with the lower accuracy system. Using an advanced
automatic tie point collection module such as ISAT by Z/l,
highly accurate tie point can be collected instantly using the
seeded EO from the DGPS/INS system. Then, without the help
of any ground control, assisted triangulation can be performed
on the collected tie point to refine the EO data to achieve
similar level of accuracy as those obtained form a high end
DGPS/INS system. Given the cost difference between the lower
accuracy POS AV 310 and high end POS AV510 (about a
factor of 1.5 to 2), Integrated Sensor Orientation seems to make
sense; however this is only true if the additional processing
time required to do the tie-point matching and ISO can be
minimized. M inefficient software and workflow are used in the
ISO process, the cost savings in the system is not realized The
results presented above show that the processing time can be
optimized with proper use of EO in the tie point matching
software (as is done in ISAT), and in the tie point collection
strategy. Results from both POS AV 510 and simulated POS
AV 310 data shows that having more tie points will in fact
degrade the assisted triangulation results. This is due to the fact
that more noise from collected tie points is included into the
assisted triangulation as number of tie point used increases.
This is an important conclusion: only the minimum of tie points
must be used to perform Integrated Sensor Orientation, which
in turn also helps reduce the processing time.
Although the results presented provide a good insight into the
use of Integrated Sensor Orientation, further work is required to
be performed before any firm conclusions can be made. This
includes conducting tests using an actual POS AV 310 system
on a film camera
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B5. Istanbul 2004
7. ACKNOWLEDGEMENTS
The authors would like to thank the financial support by
National Science and Engineering Research Council of Canada
(NSERC) and Applanix Corporation for supporting this
research work, and software support by Zl/Imaging.
8. REFERENCES
Cramer, M., Stallmann, D., 2001a. OEEPE Test on "Integrated
Sensor Orientation" — IFP Results and Experience. OEEPE
Workshop, Integrated Sensor Orientation, Hannover, Germany,
Sept 17 — 18, 2001.
Cramer, M., Stallmann, D., 2001b. On the use of GPS/Inertial
Exterior Orientation Parameters in Airborne Photogrammetry.
OEEPE Workshop, Integrated Sensor Orientation, Hannover,
Germany, Sept 17 — 18, 2001.
El-Sheimy N., 1996a. A Mobile Multi-Sensor System For GIS
Applications In Urban Centers. The International Society for
Photogrammetry and Remote Sensing (ISPRS) 1996,
Commission Il, Working Group 1, Vol. XXXI, Part B2, pp. 95-
100, Vienna, Austria, July 9-19, 1996, Best Young Authors
Award Winning Paper.
Heipke, C., Jacobsen, K., Wegmann, H., 2001. Analysis of the
Results of the OEEPE Test “Integrated Sensor Orientation".
OEEPE Workshop, Integrated Sensor Orientation, Hannover,
Germany, Sept 17 — 18, 2001.
Ip, A., Mostafa, M.M.R., El-Sheimy, N., 2004. Fast Orthophoto
Production using the Digital Sensor System. Map India 2004,
New Delhi, India, Jan 28-30, 2004
Madani, M., Mostafa, M.M.R., 2001. ISAT Direct Exterior
Orientation QA/QC Strategy using POS Data. OEEPE
Workshop, Integrated Sensor Orientation, Hannover, Germany.
Sept 17 — 18, 2001.
Mostafa, M.M.R., 2003. Design and Performance of the DSS.
Proceedings, 49th Photogrammetric Week, Stuttgart, Germany,
September 1-5, 2003.
Mostafa, M.M.R., 2001. Digital Multi-Sensor Systems —
Calibration and Performance Analysis. OEEPE Workshop.
Integrated Sensor Orientation, Hannover, Germany, Sept 17 —
18, 2001.
Mostafa, M.M.R., Hutton, J., Lithopoulos, E., 2001. Airborne
Direct Georeferencing of Frame Imagery: An Error Budget. The
39 International Symposium on Mobile Mapping Technology.
Cairo, Egypt, January 3-5, 2001.
Reid, D.B., E., Lithopoulos, J., Hutton, 1998. Position and
Orientation System for Direct Georeferencing (POS/DG), ION
54% Annual Meeting, Denver, Colorado, June 1-3, pp. 445-449.
Skaloud, J., 1999. Problems in Direct-Georeferencing by
INS/DGPS in the Airborne Environment. ISPRS Commission
111, WG III/I Barcelona, Spain, November 25-26.
Skaloud, J., M. Cramer, K.P., Schwarz, 1996. Exterior
Orientation by Direct Measurement of Camera Position and
Attitude, International Archives of Photogrammetry and
Remote Sensing, Vol. XXXI, Part B3, pp. 125-130.
n jan e P Py
Ld
e NS
