strengthen the
of the still video
ion of the drive
location of these
f the factory floor
of camera stations,
ide than expected
"or the first epoch
1ges to the initial
were included to
weaker network,
.2 exposures per
pted to overcome
it. The original
, but in this case
ected in the first
were used on the
1anually on site to
‘get and station
the basis of the
remaining frames
entirely automatic
ed to orientate the
using an intensity
re computed in a
the pixel intensity
on via bundle
Juares estimation
ita from 2D image
ooviding a self-
ons of the target
IS
cs, changes in the
> quality of some
y different results.
e precision for the
ch 1 — Epoch2
44 0.032
2 18
3 62
16 0.17
‚000 / 1:50,000
07 0.07
56 1.01
rk computations
96
In an apparent contradiction, the average object space precision
for the second epoch is slightly degraded, caused by an
unforseen loss of target images. Due to the few days between
the two epochs. many of the targets were smeared with rust and
oil. reducing the retro-reflective response despite attempts at
cleaning. changing exposures and the outright replacement of
some targets.
The precisions of the derived centres and radii of the circles
ranged from 0.1mm for the flange to 0.3mm for the bottom
plate. which are also in broad agreement with the predicted
object space precisions. The top plate and flange certainly met
the accuracy specification of +0.5mm with a large degree of
confidence. whilst the bottom plate met the accuracy
specification only at a 1.7 sigma level.
The results of the first epoch of measurement indicated that the
design dimensions were out of tolerance. The height of the
hopper, using the separation of the top and bottom plates, was
some 6.3 mm in excess of the design. To verify that this was
not an error in the overall scale of the photogrammetric
network, spirit levelling was used to independently determine
the separation of the plates. The result of the levelling was a
excess separation of 5.5mm with a precision of 1mm, which
agreed with the videometric determination.
The design dimensions for the critical flange were also in out of
tolerance, although this was expected as the surface had been
deliberately left "green" with some excess material. The amount
of surface material to be removed at each target location was
computed. The average thickness to be removed was 4.4mm.
The flange was to be reduced in thickness using a "bolt on"
milling machine. The milling machine incorporates a calliper-
like measuring system to determine cutting depth. The
measurements made with the calliper disagreed with the
photogrammetric determination of the planarity of the surface,
in the worst case by Imm. The face was therefore milled to
within Imm of the design dimension.
After the second epoch of measurement, constant dimensions
were verified and the excess in the design dimensions for the
flange had reduced as expected. The over dimension of the
height of the hopper was verified at 6mm and the centre of the
critical flange was Imm in excess of design. Again the
photogrammetric and calliper measurements could not be
reconciled, so the flange face was milled at a compromise depth
which resulted in an average of 0.4mm of under-cut. Although
this problem was never resolved, the flange was nevertheless
within the £0.5mm design tolerance according to both systems.
CONCLUDING REMARKS
Developments in digital photogrammetric equipment and
techniques will no doubt have a significant effect on a variety of
metrology applications. The development of large area CCD
sensors and corresponding improvements in target location
187
algorithms have placed the modern digital photogrammetric
system at a point where it can offer traditional users of
triangulation systems a very attractive alternative. The case
study examined is but one example of how the digital
photogrammetric alternative is finding increased acceptance.
What is perhaps even more significant is the obvious savings
available both in terms of time and money. Given that the
accuracy achievable is commensurate with that of a theodolite
triangulation system it is clear that in due course the digital
photogrammetric system will find its appropriate niche in the
metrology market.
REFERENCES
Allan, A. L., 1988. The principles of theodolite intersection
systems. Survey Review, 29(227) : 226-234.
Beyer, H.A., 1995. Digital photogrammetry in industrial
applications. Proceedings, ISPRS Intercommission
Workshop "From Pixels to Zurich,
Switzerland, pp 373-378.
Fraser, C. S., 1984. Network design considerations for non-
topographic — photogrammetry. Photogrammetric
Engineering and Remote Sensing, 50(8) : 1115-1125.
Fraser, C.S. and Shortis, M.R., 1995. Metric exploitation of
still video imagery. The Photogrammetric Record, 15(85)
: 107-122.
Fraser, C.S., Shortis, M.R., and Ganci, G. 1995. Multi-sensor
system self-calibration. Proceedings, SPIE International
Symposium , Videometrics IV, Philadelphia, USA,
Ganci, G. and Shortis, M.R., 1995a. Videometric as-built
surveys during the manufacture of a furnace hopper.
Proceedings,
Monitoring Surveys, Melbourne, Australia, pp 59-69
Ganci, G. and Shortis, M. R., 1995b. The Use of Digital
Photogrammetry For Large Scale Metrology. Proceedings,
36th Australian Surveyors Congress, Singapore, pp 247-
261.
Roberts, T. P., and Moffitt, N. M., 1987. Kern system for
positioning and automated coordinate evaluation : A real
time system for industrial measurement. Proceedings,
Volume 7, /987 ASPRS-ASCM Annual Convention,
Baltimore, U.S.A., pp 57-61.
Shortis, M. R. and Fraser, C. S., 199]. Current trends in close
range optical 3D measurement for industrial and
engineering applications. Survey Review, 31(242) : 188-
200.
Shortis, M. R.,, Clarke, T. A. and Short, T., 1994. A
comparison of some techniques for the subpixel location of
discrete target images. Proceedings, Videometrics III,
SPIE Vol. 2350, pp 239-250.
Shortis, M. R., 1992. The use of theodolite intersections to
calibrate a wind tunnel probe.
37 (4) : 295-301.
Woodward, C. A. W., 1987. Practical experiences with the
Wild-Leitz RMS2000 Proceedings, 2nd
Industrial and Engineering Survey Conference, London,
England, pp 42-53.
Sequences",
3rd Symposium on Surveillance and
The Australian Surveyor,
system.
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B5. Vienna 1996
