for 
g to 
the 
"wn 
; in 
ates 
yed 
ites 
are 
dle 
ata 
'en 
wn 
ree 
na) 
he 
rth 
vre 
nat 
ed 
lot 
Table The results of tests 
Number of RMSE of GCP 
GCP | CHP X Y 
With the 34 50 6.2 8 
STF rotation 1 5.6 7.1 
Incedent | correction 
25.0" (LJ 
26.2 9 (R) | With the 
rotation 
B/H=1.0 
correction 
Scene Status 
With the 
ephemeris 
data 
SIC 
Incedent 
14.8 9 (L) 
16.5 9 (R) 
B/H-0.5 |Without the 
ephemeris 
data 
With the 
STU ephemeris 
Incedent data 
1299 (L) 
8.9? (R) 
B/H=0.6 
Atittud 
Without the 
ephemeris 
data 39 2.0 5.3 
corrected previously. The reason is that there is 
unavoidable correlation between the additional 
parameters for the rotation correction and the 
element of sensor position during processing. 
In tests STC and STU , the orbital data are included. 
The interesting results of the two tests are useful to 
demonstrate the importance of application of 
ephemeris and attitude data during processing. The 
results show that the accuracy of the solution using 
ephemeris data is obviously improved. The 
accuracies obtained from five to nine points are 
comparable to those from 18 to 34 points. The tests 
also indicate that the stable solution of exterior 
elements of scene can be obtained using ephemeris 
data. 
In test STU, the attitude drift rates are employed. 
The result of the test proves that it is possible to 
process SPOT scenes without GCP at a special level 
of accuracy. The accuracy of the solution depends on 
that of the ephemeris and attitude data. 
CONCLUSION 
The method presented in this paper is a rigorous 
photogrammetric reconstruction of ground targets 
from SPOT stereo scenes. The mathematical model 
and programs can accommodate single stereopairs 
and aerotriagulation for SPOT scenes. According to 
the tests, the primary conclusion can be given below: 
1. It is necessary that the error due to earth rotation 
is removed before adjustment. That can separate the 
symmetric errors from adjustment system 
rigorously and ensure the reliability of solutions. 
  
71 
RMSE Of CHP (m 
X Y Z 
10 15.1 .0 
11. 6 
0.8 16.0 19.6 13.3 10 
2. The improvement effect using ephemeris data is 
obvious and tends to increase in case where the 
number of GCPs is small. The application of orbital 
constraints can avoid the high correlation between 
exterior elements and get stable solutions. The tests 
show that the accuracy achieved by using five to 
nine GCPs is sufficient to support topographical 
mapping and generation of orthophotos at 1:50,000 - 
- 1:100,000. 
3. The tests show that determination of ground 
targets without GCP is possible and indicate the 
important interesting of using ephemeris, attitude 
data and scene parameters in area where no GCPs 
can be employed. The accuracy of solutions will be 
improved with the improvement of reliability of 
orbital data. 
REFERENCES: 
[1] Li Deren, Chen Jiayu; 1988. Bundle 
adjustment of SPOT imagery, 16th 
I.S.PR.S. Congress, Com. III, Kyoto, pp. 
449-- 455. 
[2] CNES and SPOT Image, 1988. SPOT User's 
handbook, pp. A-06-4---A-06-12. 
[3] Chen, YING 1988. The Sequential collocation 
of attitude and altitude time - parameters 
for remote sensing imagery, ACTA 
Geodetica et Catographica Sinica NO.2 
[4] V.Kratky, 1989. Rigorous Photogrammetric 
processing of SPOT images at CCM Canada, 
ISPRS Journal of Photogrammetry and 
Remote Sensing, 44, pp.53--71.