thod of testing is adopted, the critical value of the
overall model test is depending on the degrees of free-
dom of the adjustment (Baarda, 1968).
To set the criteria for e.g. testing in the orientation
procedures the assesment of the precision of photo-
grammetric measurement in section 3 is used as a
starting point. The criteria for these tests are set to
two times the standard deviation of the quantity being
tested. In other words: a confidence level of 95% was
adopted. It is assumed that only normally distributed
quantities are involved.
In the sections to follow each part of the photogram-
metric process is briefly discussed with emphasis on
its guidelines. In the plotting phase a distinction is
made between mapping (production of a new digital
map) and map revision. If digital photogrammetric
workstations are deployed there is no fundamental
difference in the procedures or the quality control.
However, in case of map revision in flat terrain
‘mono-plotting’ is accepted as an efficient technique
for map revision. The last step of all surveying proces-
ses is the connection of point fields. In case of photo-
grammetric mapping this is the integration of the new-
ly plotted data with the existing map. In the HTW a
separate chapter deals with this often undervalued
step.
4.1 Photoflight
The execution of the photoflight is the bottleneck of
the photogrammetric process, as there is only a limited
number of days available for good photography
(ASPRS, 1995). The quality of the aerial photographs
is crucial for the mapping process and the quality of
the digital map. In figure 3 the first part of the photo-
grammetric process - the data acquisition - is depicted.
The two major parts are the photoflight itself and the
setting-up of ground control, possibly in combination
with the computation of the coordinates of the projec-
tion centra from GPS phase observations. The control
points are usually measured by GPS in a separate
ground control survey. The positioning of the camera
by GPS is only advantageous if it is difficult to esta-
blish and measure control points in the terrain (Bur-
man, 1995), which is rarely the case in The Nether-
lands. One often uses natural control points. Instead of
signalising points in the terrain, suitable well defined
points are chosen from the photographs. Points that
are visible in several photographs can usually also be
measured by GPS.
The guidelines for the photoflight (and flightplanning)
include for instance:
specification of flying conditions (weather,
time of the year, elevation of the sun);
photo scale (dependent on map scale);
overlap (60% in flying direction, 30%
tween strips);
tolerances for deviations of the flightplan;
the location of ground control points.
be-
If GPS is used to determine the positions of the projec-
904
| objectives and limiting conditions |
Y
( flightplanning )
C ground control planning 1
/
signalising ground control
or
choosing ground control photographs
Y
C ground control survey S
| coordinates of control points |
digital images
Figure 3: The data acquisition step of the photogram-
metric process.
tion centra it is advised to fly two strips perpendicular
to the block that contains control points only in its
corners. With the current developments in GPS-positio-
ning (such as on the fly ambiguity resolution) it is
expected that these extra strips will be superfluous in
the near future.
In the digital photogrammetric process scanning is a
critical operation because it can result in geometric
and radiometric distortion of the information in the
photographs. Therefore guidelines have been develo-
ped for the scanning of aerial photographs.
These guidelines are split up in guidelines for the pho-
togrammetric scanner and guidelines for the scanning
result: the digital aerial photograph. The main require-
ments for the scanner and the resulting digital image
are:
- geometric precision of 5 um standard deviati-
on or better, independent of the pixel size;
the pixel size is between 10 x 10 um? and
20 x 20 um?, allowing good interpretation.
A number of checks on the geometry and radiometry
of the image is provided in the HTW. One of the
checks on the geometry is part of the interior orientati-
on procedure (see next section).
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B4. Vienna 1996
