lated distances amd angles, taking into ac- 
count the ship driving speed. 
The final checking and determination 
of the buoy coordinates was made by the di- 
rect intersection with two theodolites 
where the azimuth angles are read simulta- 
neously. The error of buoy location was of 
t 0.50 m, accuracy allowed by the designers 
taking into account the river current and 
the wind influence which may displace the 
buoys during the determination. 
The error registered at the marking 
of the navigation channel was of x lm 
and it does not influence the determina- 
tion of drawing the convoy in the curve 
( Fig.6). 
The photograms processing was made 
with four phototheodolites UMK 16/13 18/F 
operating synchronoudy from the two bases 
located so as each base may cover half a 
curve in the position of normal photogra- 
phing, deviated on the left and on the 
right. The photography was made at 3 mi- 
nutes each for three convoy passing along 
the channel, namely two upstream passing 
and one downstream passing, with a speed 
of 8 km/h. The main purpose was to register 
the convoy imagine as more centred as 
possible on each photogram. 
The photo-sensitive material used was 
the aerial film JLFORD with a sensitivity of 
24 DIN. 
The bird's eye view of the photogram- 
"metric images was shown in the photogra- 
phic diagram of Fig. 7. 
The stereomodel plotting was made by 
ihe autograph ZEISS 1318, where the coordi- 
nates of the base ends and the buoy ones 
were used for the absolute orientation of 
the model. 
There were also placed fiducial 
screens with known coordinates on the bank 
for control. 
The convoy position during photogra- 
phing at the times settled in relation to 
the buoys marking the navigation channel 
may be seen in Fig. 8. 
Depending on the coordinates of the 
convoy ends and the buoys ones, there were 
determined the distances of drawing in the 
curve taking into account the buoy position 
and the navigation channel banks. 
63 
The average error in determining the 
coordinates was of mx = 0.90 m and my = 
- * 0,76 m. 
The value of convoy immersing was de- 
termined by reading the levels of the two 
ends of the convoy considering as datum le- 
vel "O" the flat surface of ihe lake, non- 
disturbed by the waves created by the con- 
voy. The convoy immersing during its dis- 
placement in the three passings along the 
channel is shown in the diagram of Fig.9. 
The phototheodolite was placed as for 
an usual photographing having a base ratio 
b/y as large as possible so as to get a 
better accuracy of the stereoscopic measu- 
rement. : 
We consider admissible the relative 
error of measuring the distance y 
dy. 
y < 
As for the photogrammetric plotting 
at an usual plotter y < - it results 
that for the measurements considered we 
have to meet the condition: 
ls D... 1. 
0 SNS 
1 
1000 
at dp, = + 0.01mm 
and the local coefficient of the photo- 
theodolite UMK 10/1318 it results that 
de, id . 
ee Y. Py 2000 to 
Hence, in order to meet the condition 
dy € loue 
y 5 1000 
in the case of photographing by UMK 10/1318 it 
should have the expression: 
boy 
The phototheodolite used in the synchro- 
nous photographing had a focal coefficient of 
99,27 m. These coefficients can not be record- 
ed.in the stereoautograph and we had to apply 
the affine rectification. 
At this rectification we used a new fo- 
cal coefficient, obtaining by multiplying the 
focal distance of the photographing photo- 
theodolite by the coefficient u = 2. 
The space-time representation by rectifi- 
cation at the autograph was made at the scale 
of 1:2000depending on the stereo-models at the 
scale of 1:2000 and 1:5000. 
For & correct plotting in the plane xy it 
was necessary a reduction on the direction y 
by the coefficient 1/n = 1/2