— 
0 1 KM 
  
A Horizontal control point 
Q vertical control point 
€ New GPS point for aerial triangulation frame 
  
  
Figure 2. GPS frame for aerial triangulation, controlling an engineering project. 
square error, including the eight control points 
within a one point fixed adjustment, is +0.025 m 
in total position and +0.021 m in height. (The 
maximum m.s.e. in position is +0.041 m, and in 
height is +0.034 m). 
After completing an aerial triangulation for 40 
models, situated in 4 strips (average scale is 
1:6,000), one model was chosen - randomly - for a 
test. Nine premarked signals and fourteen, well 
identified details were measured in the field by 
the combination of GPS and total station, based 
on the GPS frame of the aerial triangulation. 
Subsequently, the measured signals and details 
were read by five different operators in five 
different stereoplotters (Wild SD-2000 and ADAM 
ASP-2000), when the orientation of the model was 
based on the appropriate model control (pricked 
on diapositives) previously determined by aerial 
triangulation. The comparison between field 
measurements and single readings in 
stereoplotters is given in tables 1 and 2. The 
differences between geodesy and photogrammetry, 
in absolute sense, both for premarked signals and 
identified details are smaller then 0.045 m in 
position and 0.08 m in height, (with clear 
systematic character). The accuracies (m.s.e. of 
a single geodesy-photogrammetry difference) are: 
for premarked signals 40.06 m in position and 
10.07 m in height, for identified details +0.12 m 
in position and 20.08 m in height. Presumable, 
that the general accuracy of the field geodetic 
measurements were, both in position and height, 
on a +0.03-0.05 meter level. 
6.2 National GIS - Topoqraphic Data Base 
In 1991 a meaningful decision was made at the 
Survey of Israel to create the topographic data 
base of the national geographic information 
system by digital photogrammetric remapping of 
the country (Peled et al., 1991). The chosen 
photo scale is 1:40,000 for most of the regions. 
The pilot - the digital mapping of a nearly 40X40 
km area - is just being in execution. The 
26 
absolute accuracy standards for the well 
identifiable mapped details are rather rigorous: 
2 meters in position and 2 meters in height for 
the mapped details. Aerial triangulation was 
carried out for the block, composed of 81 models 
in 8 strips. From geodetic point of view, the 
aerial triangulation was controlled, mainly, by 
existing and identified third (or higher) order 
classical triangulation points and benchmarks. 
The first results show, that horizontal accuracy 
is sufficient - but not the vertical one. The 
height accuracy is regionally inhomogeneous and 
differences between geodetic and photogrammetric 
heights exceed the standard, sometimes by 30-50%. 
The clear consequence is that the control frame 
for aerial triangulation must be completely 
measured by GPS. This method will ensure both 
local and global accuracy and harmony along this 
important, national dimensioned project. 
7. SUMMARY AND CONCLUSION 
GPS is the best tool today for geodetic control 
of digital photogrammetric mapping. It is fully 
digital in its nature, efficient, sure, rapid, 
accurate and relatively inexpensive. Until the 
subdecimeter accurate airborne GPS receiver, 
combined with some kind of cheep inertial system, 
will reduce or totally cancel the need of ground 
control (Baustert et al., 1991; Cannon et al., 
1991; Dorrer and Schwiertz, 1990), GPS will be 
undoubtedly dominant in this activity. 
8. ACKNOWLEDGMENTS 
The assistance of the personnel of the Research, 
Photogrammetry and Field Divisions of the Survey 
of Israel is appreciated. Particular thanks to 
Ron Adler, Yossi Melzer and Sari Grossman for 
their support. Thanks to Mapping Technologies 
LTD., Armi Grinstein Geodetic Engineering LTD., 
and other private firms for their contribution to 
the paragraph 6.1. 
Ac 
Be 
Ce