SEEDS INDORE 
EER Tae AE: 
A total of 22 control points were deter- 
mined with the help of a 1:500 survey of 
the Palace carried out by the French Ar- 
chaeological School of Athens several 
years ago. À local co-ordinate system was 
used and the determination of the control 
points' plan co-ordinates was carried out 
on a Calcomp 9100 digitizer with an esti- 
mated uncertainty of + 20cm. Their 
heights were subsequently estimated from 
parallax measurements on the stereopair. 
This determination was by no means ideal, 
but it would not affect much the result- 
ing accuracy of the height differences. 
The camera was calibrated with a photo- 
graphic calibration in an outdoor test 
field according to the procedure 
described in 2a and with the help of the 
FullCal module. For the observations on 
the photographs the Calcomp 9100 
digitizer was used. The original nega- 
tives were printed with four times enlar- 
gement, in order to indirectly increase 
the pointing accuracy of the digitizer. 
These observations were carried out using 
a mirror stereoscope simply set up over 
the pair in order to ensure good point 
recognition. Apart from the 22 control 
points a further 1500 detail points were 
observed and it turned out that no mis- 
take occurred in point recognition. 
Using the second module (PHOS) and 9 of 
the 22 control points the exterior orien- 
tation parameters of the two photos were 
determined. The other 13 control points 
were kept for checking the achieved accu- 
racy. Their co-ordinates, along with 
those of. the 1500 detail points, were 
determined using simple photogrammetric 
intersection provided by module PHOS. The 
differences of the co-ordinates  deter- 
mined from the initial ones gave a strong 
indication of the high accuracy and reli- 
ability of the method used. In other 
words the accuracy achieved was well wi- 
thin the expected limits, considering the 
quality.of .the _initial.co-ordinates._.In 
Table 2 these results are summarised. 
  
No of CP's |oX (m) oY (m) oZ (m) 
  
xz 0.15 + 0.15 |: 6.43 
0.28 £0.19 |+ 0.58 
  
bp 
w 
I+ 
  
  
  
  
  
  
Table 2 
Considering the conditions of the photo- 
graphy and the quality of the control 
points, these results are encouraging. 
Further investigation into the  improve- 
ment of the accuracy is already underway. 
The ground co-ordinates of the detail 
points were stored into an ASCII file by 
module PHOS along with the appropriate 
codes for the automated restitution. 
Using the AutoCAD restitution module the 
plot was produced within the AutoCAD en- 
vironment. Some editing was necessary in 
order to rectify certain mistakes of the 
digitizing stage. The final plot is pres- 
ented in Figure 2. The time necessary for 
  
each of the various stages to produce 
this plot appear in Table 3. 
  
Action Time (min) 
  
Digitizer observations 
  
  
  
  
  
  
  
  
(for 100 points) 30 
Photogrammetric adjust- 
ments — PHOS 15 
Restitution - SURVEY 2 
Editing — AutoCAD 30 
Plotting 3 
Table 3 
4. CONCLUDING WORDS 
The package described above does not 
claim an award of originalitv. It is, 
however, built. in. such a way to enable 
modular use by the users. This means that 
each of the described modules may be im- 
plemented independently and  irrespecti- 
vely of the origin of the input data: 
Thus the FullCal module may be used on 
its own for the calibration of a camera. 
The second module may be used for 
resecting single photographs as well as 
for determining spatial co-ordinates. 
Finally the third module accepts ‚Input 
from the previous one, but may well be 
used with a compatible ASCII file pro- 
duced in any other way (eg. classical 
geodetic observations). 
Furthermore, it is intended to be used 
with any kind of camera, be it a metric 
aerial survey camera or a phototheodo- 
lite, or a semi-metric reseau camera or 
even a non-metric camera. Thus its uses 
are only limited by the user's imagi- 
nation, as the package could be  imple- 
mented in cases of small aerial surveys 
and in close range applications. 
It is believed, that it provides an ideal 
integration of the traditional pointwise 
photogrammetric restitution with the pow- 
erful editing facilities of today's stan- 
dard CAD software, promoting at the same 
time the idea of automation. 
Finally, it is considered to be a posi- 
tive solution to today's demands, which 
are simplicity, ease of use, economy and 
effectiveness. FotoCAD ensures quick pro- 
ductivity for non-photogrammetrists, thus 
making Photogrammetry available to them. 
It could be stated that it is a decisive 
contribution to a new concept approach 
(Georgopoulos 1990), which will change 
Photogrammetry as we have known it up to 
now.