(2). In case c, the strong correlation between X,,Yo,f 
and X,,Y,,Z, is greatly weakened. Figure 1.b shows 
that their correlation coefficients are very small. 
Therefore, the determination values of the 
orientation elements are almost a constant for 32 
photographs and the accuracy is very high. The 
overall accuracy of the elements of exterior 
orientation is quite close to that of case a, the 
accuracy of the elements of interior orientation 
reach +10 um and all 32 determined values are not 
obviously different to each other. 
(3). Similarly, high precision elements of interior 
orientation are obtained in case d. The results are in 
accordance with that of case c. However, the need 
for ground control points in case d are much less 
than that in case c. 
4. CONCLUSION AND SUGGESTION 
(1). The test has confirmed that the two kinds of 
GPS-supported determination methods described in 
this paper for interior orientation elements of aerial 
camera are correct in theory and efficient in practice. 
The camera calibration can be performed in 
photogrammetric adjustment. 
(2). In photogrammetry, the application of GPS can 
not only reduce the number of ground control points 
and replace the hard work of field measurement, but 
also have enormous potentiality on camera 
calibration. This amply demonstrates that GPS can 
be widely applied in photogrammetry. 
(3). If several pictures with carrier phase 
measurements are taken over a high precision test 
field before and after an air photographic mission, 
the elements of interior orientation of aerial camera 
can be determined by using GPS-supported single- 
image resection in space. The test field method will 
eventually replace the conventional 
photogrammetric operation method where the 
camera calibration is independent of photo 
orientation. 
(4). The determination method of interior orientation 
elements in  GPS-supported bundle block 
adjustment can be widely applied in 
photogrammetry. This processing method does not 
pose any major problem for photogrammetric 
adjustment but can efficiently correct the systematic 
errors caused by the high dynamic aerophotography. 
The dynamic method of camera calibration can take 
full account of the actual conditions in 
aerophotography. There is no question that the 
method is helpful for improving the accuracy of 
photogrammetric densification. 
216 
ACKNOWLEDGMENT 
| should like to express my hearty appreciation to 
my supervisor Prof. Li Deren for his guidance and 
encouragement in the completion of the research 
project which leads to this paper. 
REFERENCES 
[1] Ackermann,F.,1991. GPS for Photogrammetry. 
Proceedings of ISPRS, pp 17~69. 
[2] Friess,P., 1991. Aerotriangulation with GPS-- 
Methods, Experience, Exception. Proceedings of the 
43rd Photogrammetric Week. 
[3] Li Deren, 1991. Application of Global Positioning 
System in High Accuracy Point Determination 
Airborne Remote Sensing. Remote Sensing of 
Environment (in Chinese), 6(3), pp 216-223. 
[4] Li Deren, Yuan Xiuxiao, 1992. Bundle Block 
Adjustment with Small Format Air Photographs on 
Surveying and Mapping On Antarctica. Antarctic 
Research (in Chinese), 4(2), pp 27-35. 
[5] Li Deren, Yuan Xiuxiao, 1995. GPS-supported 
Bundle Block Adjustment. Acta Geodaetica et 
Cartographica Sinica (in Chinese), 24(3), pp 1-7. 
[6] Wang  Zhizhuo, 1990. Principles of 
Photogrammetry (with Remote Sensing). Press of 
WTUSM, Publishing House of Surveying and 
Mapping, Beijing. 
[7] Yuan Xiuxiao, 1994. GPS-supported Combined 
Bundle Block Adjustment Based on Differential 
Carrier Wave Phase Observations (in Chinese). MS. 
Dissertation, WTUSM. 
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B1. Vienna 1996