System Launch | Pixel- System 
Nominal 
Ground 
Resolution 
EarthWatch Stereo Fore/Aft 
- Earlybird 1996 3m Panchromatic 
- Quickbird 1997 1m Panchromatic 
Space Stereo Fore/Aft 
Imaging 1997 1m Panchromatic 
Orbview Stereo Fore/Aft 
1998 8m Multispectral 
1998 1 & 2m Panchromatic 
SPOT Stereo Fore/Aft 
5a 2001 5m Panchromatic 
  
  
  
  
  
  
Table 2: High spatial resolution sensor systems 
planned for 1996-2001 
3.6 Testing the potential 
OS, in collaboration with the Department of 
Photogrammetry and Surveying at University College 
London and the Department of Geography at the 
University of Southampton, has embarked on a project to 
investigate the potential of the new high spatial resolution 
systems. The project is part funded by the British National 
Space Centre through their second Application 
Demonstration Programme. 
The project is very much focused on seeking solutions to 
problems, for example: 
« seeking more cost-effective/efficient ways of 
updating certain geospatial data themes, 
e assessing the potential of greater automation 
in those processes, 
e employing economies of scale, where say 
one or more update processes can be 
combined and utilise a image to minimise 
production overheads. 
A common benefit the areas under investigation enjoy is 
that OS has extensive a priori knowledge in a digital form 
readily available to assist the process. This can be 
valuable in determining a ground surface DEM, for 
example, to minimise nugatory effort in populating areas 
where the desired height values are inaccessible. 
Test Sites: The project has various facets: topographic 
mapping, enhancement of Digital Elevation Models, 3D 
urban modelling and land use/cover determination. Test 
sites have been established in Lincolnshire (low lying 
coastal, the Lake District (mountainous) and 
Hertfordshire (rolling agricultural land). Unfortunately 
there is no suitable space imagery currently available, 
simulated or otherwise. OS is therefore creating 
simulated imagery from recent aerial photography and is 
acquiring imagery from one or more of the following 
sensors: ERS-1/2, Wide Angle Optoelectronic Stereo 
Scanner (WAOSS) and the Compact Airborne 
Spectrographic Imager (CASI) systems. 
104 
Where possible airborne system parameters will be 
matched as closely as possible to one of the proposed 
space platform systems. It is hoped that by the 
completion of the work (early 1997) at least one of the 
new high spatial resolution systems will be acquiring data 
from which we can validate the results of simulated test 
data. Specific areas of investigation are: 
Topographic Mapping An important issue when 
planning the deployment of survey staff and resources is 
to know what change exists and where it exists. Although 
OS already employs several approaches to solve this 
problem, it is believed that change detection methods can 
be improved. The nature and importance of change 
varies by topography and geography, e.g. a new fence 
dividing a rural land parcel may be much more important 
than the erection of a similar structure in an urban public 
park. Although urban change is often well reported, rural 
change is often more difficult to detect and hence is rarely 
fully documented. 
Mapping, whether new surveys or revisions are 
considered, is also being examined within the project 
using mono and stereo image techniques. 
Digital Elevation Models Now that the national high 
resolution terrain model digitising programme is coming to 
an end attention has focused on improving the 
maintenance and enhancement of that surface. Three 
areas of investigation within the investigation are: 
e automatic detection. of changes in terrain 
surfaces, 
e remodelling the surface following major 
engineering works (highways etc.) and 
e  improving the surface resolution where the 
environment is sensitive to floods etc. 
It is expected that existing topographic data will aid the 
process - for example, highway topography can be 
employed as breaklines and polygons defining buildings 
and woodland can be employed as masks either to 
prevent automated terrain modelling or as edit tools to 
remove spurious surface heights. 
Building and 3D Urban Modelling There is growing 
demand to model the world as we recognise it in three 
dimensions for specific applications. It is safe to say that 
the next generation of data users who have been 
introduced to realistic 3D PC games systems at an early 
age are unlikely to tolerate anything other than an 
interactive model to test urban planning scenario, 
concepts or environmental impact assessments. 
The benefit of an existing 2D dataset, and a supporting 
terrain surface, is seen as a major asset from which to 
start 3D or perhaps more realistically at this stage 2.5D 
modelling. Techniques are being explored to provide an 
upper building surface height and to more realistically 
model upper surface planes and structures. 
As with any project the data population phase is relatively 
trivial compared with ongoing maintenance. Currently 
demand does not warrant the conversion of all survey OS 
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B6. Vienna 1996 
tec 
the 
bei 
Lar 
the 
anc 
little 
are 
are 
fror 
pol 
If s 
ma 
tha 
pla 
suf 
que 
Mu 
org 
top 
nee 
affc 
sat 
Col 
tec 
coll 
imp 
aut 
In 
sta 
ens 
rea 
Rei 
Far 
- 0 
Phe 
40: 
Acc 
Infr 
Gilt 
196 
Ha 
ren 
Jot