The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part Bl. Beijing 2008 
1066 
during the flight for both vertical and oblique images. In order 
to better use and visualize both oblique and ortho images, 
Electronic Field Study (EFS) has been developed at Pictometry. 
Both vertical and oblique images can be easily viewed in EFS, 
and spatial measurement such as distance and height of objects 
on the ground can be easily performed on both oblique and 
vertical images. The results can be exported into ArcGIS 
directly to update the existing geo-spatial information in the 
database. 
2.2. Camera Calibration 
Camera calibration is an important process in photogrammetric 
mapping to ensure extraction of accurate and reliable 3D 
information from imagery. In camera calibration, the principal 
length of the camera, the coordinates of the principal point in 
image coordinate system and the coefficients of lens distortion 
including radial distortion, tangential distortion and affinity and 
shearing are computed. Various calibration methods have been 
developed for calibration of digital mapping cameras and the 
performance of some medium format digital cameras can be 
found in Cramer (2004). At Pictometry, a calibration system has 
been developed for calibration of its digital cameras and the 
system was provided to EROS center of USGS at Sioux Falls 
for establishing a calibration system for calibration of various 
digital mapping cameras in mapping community (Pictometry, 
2002). Basically, the Pictometry's calibration system includes an 
indoor calibration cage with evenly distributed targets as shown 
in Figure 1 and software Australis which is the well-known 
calibration software (Fraser and Edmundson, 2000). 
(a) Calibration cage 
In calibration, the camera to be calibrated captures a number of 
images against the calibration cage from different locations to 
form a network, and Australis is then used to measure the 
targets in the images automatically and accurately and to 
perform a free network bundle adjustment to compute the 
interior orientation parameters and distortion coefficients of the 
camera. Figure 2 shows the calibration result of Pictometry 
digital camera. It can be seen that the radial lens distortion of 
the camera is very small in most area of the image and high 
accuracy of point measurement can be achieved after correction 
of lens distortion. 
The advantage of Pictometry's calibration system lies on its 
efficiency and reliability. It is very easy to run calibration of 
digital cameras with Pictometry's calibration system and very 
(b) Calibration target 
Figure 1. Camera calibration cage and target 
economic, compared with other approaches such as in-situ 
calibration method. Figure 3 shows the radial lens distortion of 
a Pictometry digital camera obtained at calibrations repeated in 
a short time period. It can be seen that the results are consistent 
and the difference between two calibrations is very small. It is 
very important that each digital mapping camera is calibrated 
regularly so that the changes of camera's interior orientation 
parameters and lens distortion parameters between two 
consecutive calibrations are within the defined tolerance and do 
not affect the mapping accuracy. 
Figure 3. Radial lens distortion at repeated calibrations