ly calibrated with respect to the measuring marks of the instrument.
2.3 All 4 sets of data were processed through bundle block adjustment
with 12 block-invariant additional parameters (program PAT-B).The pre-
cision results (oy) and the accuracy results (u u u x are sum-
marized in table 2. The absolute accuracy values (1 ard obtained as
r.m.s. coordinate differences between the photogrammetric and the inde-
pendent geodetic determination of 107 planimetric check points (verti-
cal check points were not available). The differences contain the inac-
curacy of the geodetic coordinates which is known to be 9 mm in the
terrain.
Table 1: Sumamry of the 4 sets of data, block Schnürpflingen
Version | measurement | type of no. of tie-points no. of xyz- no. of
points photo terrain control points check points
photo terrain terrain
a operator signalized 405 104 39 12 107
PK1
b operator signalized 405 104 39 12 107
C 100
c digital 1) |natural 1) 405 104 39 12 107
C 100
d digital 1) |signalized 1)|400 104 39 12 107
C100
1) only tie-points
Table 2 : Block Schnürpflingen, precision and accuracy results after
bundle-blockadjustment with 12 additional parameters
'tie-points from check-points
Version| measurement point transfer % ux uy Mx y
[um ] [um] [ um] [um ]
a operator no 2.4 2.7 3.2 3.0
PK 1 (sign. points)
b operator no 2.7 3.0 3.5 3.3
C 100 (sign. points)
C dig. correlation yes, with 4.0 3.6 3.8 3.7
C 100 natural points
d dig. correlation yes, with 3.9 4.5 3.9 4.2
C 100 sign. points
2.4 The evaluation of the results is primarily concerned with the com-
parison of versions b) and c). Version a) is to be compared with ver-
sion b). In the latter case all features being identical except for the
measuring instrument, the slight differences show the addtional error
effects caused by the planicomp,amounting to 1.2 um in gg ànd 1.4 ym in
u ;
X,y
Considering first the precision indicator 6o of the bundle block ad-
justment we notice values of 2.7 um and 4.0 um for the versions b) and
c) respectively. The 2.7 um of case b) contain the total error budget