ISPRS Commission III, Vol.34, Part 3A „Photogrammetric Computer Vision“, Graz, 2002 
  
quite high stability of physical boresight alignment over longer 
time periods, assuming optimal GPS/inertial data processing 
and the use of a correct mathematical approach for modelling of 
physical reality of image formation during AT. 
4.3 DG based on long term boresight angle calibration 
Within the preceding sub-section the stability of boresight 
alignment parameters and self-calibration terms was analysed 
and certain variations in some parameters have been seen. In 
order to simulate the later practical use of direct georeferencing, 
where the calibration parameters from system calibration should 
be used for several mission flights ideally, the long term quality 
of system alignment is checked using the 21 available control 
points as independent check points for overall quality checking. 
Only the mean boresight calibration is applied since position 
offsets as well as influence from self-calibration are varying 
and cannot be corrected in advance. The performance analysis 
from check point differences from over-determined forward 
intersection is given in Table 5. 
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
RMSAX | RMSAY | RMS AZ 
Day #Flight nil im) [m 
Jan 29 1+2 0.174 0.091 0.536 
Jan 31 5+6 0.211 0.066 0:575 
Feb 05 8+9 0.194 0.112 0.385 
Feb 18 107-11 0.076 0.170 0.365 
Feb 19 12-13 0.184 0.167 0.380 
Feb 21 1415 0.072 0.078 0.270 
Feb 24 1617 0.073 0.077 0.463 
Mar 12 | 18+19 0.088 0.094 0.436 
Mean 0.134 0.106 0.426 
  
Table 5. Quality of DG based on long term calibration. 
As to be expected, the maximum deviations are present in 
vertical component and raise to values up to 50cm, the mean 
RMS is about 4dm. In horizontal component the differences are 
within 2dm maximum, the mean RMS is about 12cm. 
Comparing these numbers to the values obtained from direct 
georeferencing with optimal system alignment, the accuracy is 
significantly worse, which shows the influence of non-optimal 
overall system alignment mainly due to remaining global 
position offsets. Nevertheless, such global errors can be easily 
overcome if integrated sensor orientation with minimal number 
of GCPs is applied. 
5. CONCLUSIONS 
The results presented above have reconfirmed that GPS/inertial 
data integration and overall system calibration is the most 
critical factor during direct georeferencing. Besides the need for 
consistently high GPS/inertial positioning and attitude quality, 
which has to be guaranteed throughout the whole mission 
duration, the estimation of physical relevant and correct 
calibration parameters is the crucial task during system 
calibration. Especially the correlations appearing between 
different parameters used in calibration are eminent since they 
compensate the impact from other physical effects, which might 
cause trouble when the calibration is transferred to the mission 
site. System calibration is “the” challenging task, where the 
silver bullet for the most efficient calibration procedure is — 
unfortunately — not found yet. 
Nonetheless, results from this real flight test underline the 
highly operational use of GPS/inertial components for direct 
georeferencing. The exclusive correction of mean boresight 
angles is sufficient for object point accuracy within 4-5dm 
(RMS) if strong image overlap, i.e. block geometry is given. No 
AT process (except for calibration) is necessary to reach this 
quality. The boresight angles remain constant within a certain 
interval and can be used for longer time periods. Although the 
overall quality is less compared to the well controlled 
GPS/inertial accuracy tests, the results are quite remarkable for 
the first long term test in true production environment. 
ACKNOWLEDGEMENTS 
The authors would like to thank Hansa Luftbild German Air 
surveys (Münster, Germany) and IGI GmbH (Kreuztal, 
Germany) for providing the data. Special thanks needs to be 
expressed to Anke Steinbach from Hansa Luftbild and Jens 
Kremer from IGI for their support and cooperation. 
REFERENCES 
Brown, D. C. (1971): Close-range camera calibration, 
Photogrammetric Engineering 37(8), pp. 855-866. 
Brown, D. C. (1976): The bundle adjustment — progress and 
prospects, Congress of the International Society for 
Photogrammetry, Invited Paper of Commission I, Helsinki, 
Finland. 
Cramer, M. (1999): Direct geocoding — is aerial triangulation 
obsolete?, in Fritsch/Spiller (eds.): Photogrammetric Week 
1999, Wichmann Verlag, Heidelberg, Germany, pp. 59-70. 
Cramer, M. (2001a): On the use of direct georeferencing in 
airborne photogrammetry, in Proceedings 3rd. International 
Symposium on Mobile Mapping Technology, January 2001, 
Cairo, digital publication on CD, 13 pages. 
Cramer, M. (2001b): Performance of GPS/inertial solutions in 
photogrammetry, in Fritsch/Spiller (eds.): Photogrammetric 
Week 2001, Wichmann Verlag, Heidelberg, Germany, pp. 49- 
62. 
Ebner, H. (1976): Self-calibrating block adjustment, Congress 
of the International Society for Photogrammetry, Invited Paper 
of Commission III, Helsinki, Finland. 
Fraser, C. (1997): Digital camera self-calibration, in ISPRS 
Journal on Photogrammetry and Remote Sensing, 52(4), pp. 
149-159. 
Grün, A. (1978): Accuracy, reliability and statistics in close- 
range photogrammetry, Inter-congress symposium, 
International Society for Photogrammetry, Commission V, 
Stockholm, Sweden. 
Heipke, C., Jacobsen, K. and Wegmann, H. (2002): Analysis of 
results, Final report on the OEEPE test on integrated sensor 
orientation, to be published. 
Kremer, J. (2001); CCNS and AEROcontrol: Products for 
efficient photogrammetric data collection, in Fritsch/Spiller 
(eds.): Photogrammetric Week 2001, Wichmann Verlag, 
Heidelberg, Germany, pp. 85-92. 
Reid, D. B. and Lithopoulos, E. (1998): High precision pointing 
system for airborne sensors, in Proceedings IEEE Position 
Location and Navigation Symposium (PLANS), pp. 303-308. 
Nilsen, B. (2001): Can map compilation rely on GPS/INS 
alone?, OEEPE workshop on integrated sensor orientation, 
Hannover, Germany, September 17-18, 2001.