while digital photogrammetric systems 
are in themselves not cheap, the 
superimposition feature is purely a 
software function within them and 
thus incorporated in the purchase 
cost. In spite of the latter, if 
such superimposition is to be in 
stereo there are very significant 
introductory costs. 
The question must therefore be asked 
"is full stereo superimposition 
essential for revision of 
large-scales data?" This leads 
directly to consideration of using 
the digital photogrammetry principles 
within a monoscopic solution via a 
digital orthophoto, a process which, 
depending on the availability of a 
suitable scanning facility within an 
organisation, can offer a cheaper 
option. While such methodology may 
well Seem anathema to the 
photogrammetric  purist cost is a 
major consideration for any 
organisation. Fitness for purpose is 
thes critical factor and! 1f-à Less 
costly system can revise a database 
without loss of data precision then 
its potential cannot be ignored. 
Theres is. ‘nothing 0 new imvithis, 
monoplotting technology being well 
established. However its widespread 
use for revision purposes has to date 
been largely confined to scales of 
1:25,000 and smaller. 
In practical terms the principles are 
the same regardless of scale, but in 
the case of larger scales, such as 
1:2500, the degree of refinement in 
respect of information needed to 
produce the digital orthophoto is 
more demanding. 
Five main stages apply to the 
updating process using the 
mono-plotting system: 
- acquisition of appropriate, up to 
date, vertical air photography 
= scanning of the resultant aerial 
film to produce the digital 
raster image 
- J— incorporation of digital 
elevation data to produce the 
digital orthophoto 
36 
- superimposition of the existing 
vector database for updating 
from the digital orthoimage 
- subsequent databasing of the 
updated result 
While some of these are largely 
independent of the scale of the 
database to be revised, others have a 
major influence on the accuracy of 
the final updated product. Taking 
each of the above stages in turn the 
following observations apply. 
i. Aerial Photography 
Given the right flying conditions, a 
properly specified aerial camera and 
appropriate flying parameters, there 
is no problem in acquiring suitable 
vertical aerial photography. Until 
digital aerial cameras come into 
common usage the digital 
photogrammetry process will commence 
with film, or diapositives, produced 
in this way. 
ii. Scanning 
This operation has a major influence 
on the accuracy of the final updated 
product and in digital photogrammetry 
the accuracy achievable increases in 
direct relation to the resolution of 
the scanning process. The smaller 
the pixels produced the more accurate 
the result. This in turn directly 
influences the standard of the 
eventual orthophoto and consequently 
governs the match achievable with the 
vector database at superimposition. 
Most proprietary scanning systems 
will include software for input of 
camera parameters, etc and to deal 
with all aspects of the orientation 
and scanning operations. The desired 
result is a digital image, 
incorporating all this necessary 
input, at a suitable resolution and 
in the required co-ordinate system. 
While scanning accuracies in the 
order of 14-20 microns would appear 
necessary for successful revision of 
large scales data, more practical 
work needs to be done in respect of 
using such results in a production 
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B6. Vienna 1996