ra 
WO 
are 
pic 
1 a 
res 
ate 
WO 
by 
ata 
nd 
le. 
ric 
ric 
'ell 
es, 
on 
ite 
ing 
aw 
ato 
the 
'ell 
in 
ver 
rol 
PS 
of 
ard 
Or 
ng 
aw 
ich 
ing 
ich 
nd. 
of 
ind 
rol 
for 
ing 
rce 
ing 
rol 
om 
ige 
the 
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B4. Istanbul 2004 
process effects the accuracy of the digital orthophoto. The 
absolute accuracy of an orthoimage depends also upon the 
quality of the ground control information. 
Some ASTER, IKONOS, QuickBird, and SPOT images may be 
delivered as image tiles. If the image tiles were cut from a strip 
of data acquired on the same day in a single pass of the 
satellite, the tiles can be tied with same pass constraints and 
the orbital data is rebuild for the whole strip. This process 
offers some advantages: 
° Fewer ground control points (GCPs) to collect since 
fewer math models are computed. 
* More coverage by the math model by bridging over 
obscured areas, such as areas under cloud cover, where 
GCPs cannot be collected. 
3. METHODOLOGY 
In this study, the affect of the same pass constraints on the 
adjustment results was investigated. The aim of the study is to 
investigate could the usage of these constraints reduce the 
needed GCP number or with the same number of GCP could a 
better accuracy reached A block with two columns each of 
which consists of three SPOT frames from the same pass was 
used. At the first step, the block is adjusted using the 29 
ground control points and 102 tie points distributed 
homogeneously over the block. GCP distribution is shown in 
the Figure 1. Some GCPs are measured on several images. So, 
the number of GCP measurement is 40. The results of this 
block adjustment are accepted as reference. The adjustment 
results are given in the Table 1. 
meu A Eod joi Eee r E E EA : 
/ ee 
/ a / A / / : 
/ 5 j fo d 
/ e / / a 
/ a / / A / 
' / / y Fan 
/ E ® ja SS / 
/ / 
/ / A / / 
Fo _ ere re 
b f / Try armee | 
/ 5 / / / 
7 e / |e | 
/ aj A sf x / 
m Lo / / 
A E E 
Figure 1. SPOT block with all GCPs 
625 
  
  
  
  
  
  
  
  
  
X Y Z 
GCPs | Average 
Error 3.524 3.097 2.606 
(meter) 
RMSE 4.337 3.734 3.818 
(meter) 
Tie Average 
Point | Error 0.17 0.27 
s ixel) 
RMST 0.27 0.39 - 
(pixel) 
  
  
  
Table 1. Reference adjustment results 
At the second step, only the ground control points over the two 
images at the north of the block was used. The number of 
GCPs is 13 and the number of GCP measurements is 15. GCP 
distribution is shown in the Figure 2. The adjustment results 
are given in the Table 2. 
di 
Frame PARA Wr MN dee / 
as ! 
Pe LT ES e. ln TE 4 
Figure 2. SPOT block with fewer GCPs 
  
  
  
  
  
  
  
  
  
  
X Y Z 
GCPs | Average 
Error 4.091 5.428 0.667 
(meter) 
RMSE 4.655 6.202 1:121 
(meter) 
Tie Average 
Point | Error 0.29 0.25 - 
8 (pixel) 
RMSE 0.40 0.34 - 
(pixel) 
  
  
Table 2. Adjustment results with fewer GCPs