International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B-YF. Istanbul 2004 
  
()e2473170.dst F:\KaltengSurvel\Gps\DATA\TEAMO2(2003111 31RAW 90KB DAT File 
3 B2473170.eph | F:\KaltenaSurvei\Gps\DATA\TEAMO2\20031 1 13\RAW 11KB EPH File 
ti B2473170on  F:\KaltengSurveilGps\DATA\TEAMO2120031 1 131RAW 1KB ION File 
1 82473170.mes F:\KaltengSurvei{Gps\DATAITEAMO2\200311 13)RAW 2KB MES File 
3D Position Best PDOP Position [ 2.3] Mean Position [918] 
Latitude: 1:35'38.25980" $ 1:35'38.25572" S 
Longitude: 113:18'87.23415" E 113:18'07.23219" E 
Height [m]: 83.0 84.8 
Fig. 6. GPS raw data file and absolute 
1.6 Identification Validation 
Validations begin with the comparison between air photo (fig 7 
and fig. 8) and snapshots (fig. 9) and between GPS empirical 
route to the planed route. When the data is valid, then the 
GCP will be drawn on AutoCAD with the air photo 
background (ungeoreferenced) with special prepared Auto 
LISP program, to guarantee the consistency in style, numbering 
- system and filename. 
Although identification is done carefully, from about 500 
GCPs, 50 GCPs were in wrong location (misidentification). It 
means at least 6 working days with helicopter etc. Although 
until the end of the project, the conventional way is continuing 
to be used, a new approach should be developed. 
There is idea with cross-strip (new aerial photography). The 
problem is then how to identify the new air photos to the old 
one. Cross strips could be reducing the demand of GCP, but 
the error (misidentification) in GCP of cross-strips will 
influence very large to the whole project. 
In addition, the solution is still the minimum nine GCP 
measurements for each map sheet, but the way of survey 
should be improved 
  
Fig. 7. GCP on air photo 
  
Fig. 9. The actual place (12 years later) 
2. MULTI-TRIANGULATION 
To speed up the work, following strategy could be done: (1) a 
stable or low. vibrating helicopter should be selected. (2) A 
small digital metric camera will be mounted on helicopter. (3) 
The camera should be connected to kinematical-differential- 
GPS (KiDGPS). (4) Over the potential GCP, the helicopter 
will fly a round the point and make several oblique snapshots 
photographs. (5) The projection centre for each snapshot will 
be derived by adjustment of all KiDGPS measurement using 
geodetic GCP. on base camp. (6) The photo-GCP will be 
identified using the actual snapshots and the old aerial 
photographs, where the potential photo-GCP should be found 
and its coordinate will be derived by forward resection using 
all projection centres of all-snapshots of the photo-GCP. (7) 
The computation could be in one bundle adjustment includes 
KiDGPS-base-line … measurement, new-snapshots-forward- 
resection; and old-photo aerial triangulation. 
There are also three kind of triangulation in apply: 
(1) The triangulation to get the the center coordinates of all 
snapshoot. This is done by the use of all Kinematic 
Differential GPS data and might be additional information 
from inertial system if any. 
a]