The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part Bl. Beijing 2008 
Domain 
Item 
Contents 
Test field 
All GCPs 
117 points 
Platform 
Helicopter 
AS-350BA 
Flight 
Speed 
130 km/h 
Altitude 
1270 m 
GPS 
Rate 
1 sec 
IMU 
Rate 
200 Hz 
Laser 
Scan rate 
70 kHz 
Scan angle 
40 degree (full) 
Digital Camera 
Focal length 
60 mm 
Sensor size 
9 x 9 pm 
Image size 
5440 x 4080 
Table 1. Data component 
Figure 4. Centre image 
Figure 6. Image quality 
3.4 Accuracy Evaluation 
Table 2 shows root mean square error for 26 check points, and 
permissible errors means restrictions for check points which are 
established by GSI in generating each scale map using direct- 
georeferencing system. RA (Relative Accuracy) means the 
value which was computed from equation (5) 
RA = 
1 
77 
(5) 
where FFFlight Height 
enroot mean square error for x, y, z 
It can’t be found significant differences between both 
calibration models, but it can be confirmed that the horizontal 
and vertical accuracy are improved by the simultaneous 
adjustment. Furthermore, it can be found that the both 
calibration accuracy compare with permissible error show less 
than 1/1000 and more than 1/5000. Consequently, it is 
concluded that calibration using pseudo GCPs is practical 
method in direct-georeferencing, and the accuracy was 
improved by triplet restriction. 
T yP e 
o Y 
RA 
Using Pseudo 
GCPs 
0.422 
0.422 
0.449 
1/1701 
Simultaneous 
adjustment 
0.402 
0.372 
0.433 
1/1819 
Permissible error 
1/500 
1/1000 
1/2500 
1/5000 
0.150 
0.150 
0.200 
1/4356 
0.300 
0.300 
0.300 
1/2444 
0.750 
0.750 
0.500 
1/1083 
1.500 
1.500 
1.000 
1/542 
H =1270m in RA 
Table 2. RMSE for check points 
5. CONCLUSION 
376 
In order to develop new calibration methods combined with 
GPS/IMU and Laser Scanner data, simultaneous adjustment 
was investigated in this paper.