The Boundary Problems web site

Selected published articles - 3
 

Reproduced from Surveying World, Vol 8 No 2, Jan/Feb 2000,
with the kind permission of the Editor and publishers of Surveying World.
You may e-mail the Editor from here.
Jon Maynard  Better mapping than ever before

   By Jon Maynard
Where is topographic map-making going over the next ten years? That will depend on the technological and market influences that the industry faces but already there are some clear pointers, as Jon Maynard explains. C omputers are now involved in almost every stage of map production, from data capture to printing. Indeed, some maps are designed specifically for use on computers, and even some printed series maps, such as Landranger, are now offered as an equivalent raster product. Currently, vector digital map data specifications are no more than computer models of two-dimensional paper maps; ideally they should be computer models of the real world and for some applications, in 3-D.
    Map-makers would like to be able to collect data once only and to re-use it by automatically 		generalising it to smaller scales. For software, this remains an unrealised dream, and even the system used to create Landplan can only take periodic snapshots of the larger scale data from which it is derived. Yet customer expectations are rising when prices are expected to reduce and surveys are expected to be more current and immediately available to customers.
     Customers are increasingly using maps in digital applications, and customers are no respectors of map scale, viewing them at greatly enlarged or reduced scales. There is also an increasing demand for better height data than is afforded by products
    30  Surveying World January/February 2000  
 
 
'...it is now up to the customer's software to produce a fair plot, or screen image ...'

The concept of a map having a standard specification will disappear.

 derived from the contours on 1:10,000 scale maps.

More efficient data capture
Photogrammetry is likely to remain the most cost-effective means of data capture for mapping. But here, as everywhere, technology attempts to find faster, cheaper solutions. Expensive analytical plotters are being slowly replaced by softcopy systems on PCs (now cheaper than plotters). Automatic feature recognition software, when it is perfected, is expected to produce further savings. Airborne optical cameras are still the preferred source of raw data, although satellite imagery is now becoming available with a 1m resolution, but this is unlikely to be used for mapping at scales larger than 1:20,0000. In any case, it is more difficult to obtain satellite images of a specified area at a given time than it is to obtain air photos, unless you resort to images from radar satellites: again at 1m resolution but with the added advantage of being unaffected by cloud, haze, dust or smoke.
     The CCD technology of traditional satellite imagery is capable of further refinement, and eventually digital cameras will start to rival optical cameras for airborne photography (see elsewhere in this issue for developments on that front). This will give further economies in softcopy photogrammetry as it dispenses with film processing and scanning.
     Laser altimetry will increasingly be used to improve the knowledge of the third dimension in flatter areas. It will also be used to capture the three-dimensional ‘clutter’ of the urban environment, needed for the development of intensive cellular communications.
     Ground survey is still seen as being the fastest, cheapest way of capturing small pockets of topographical change. The need by some customers for maps of what is being built even before it is completed, places quite a strain on both the intelligence system and the manpower resources of a national mapping agency. This has led Ordnance Survey to experiment with incorporating third party design and as-built surveys into the National Topographic Database (see the reference to CODES on page 37).
     Technology change continues apace with survey instruments. However, the huge productivity increase represented by devices such as the Cyrax laser-scanner is unlikely to be realised in topographic mapping: the instrument produces far more data than these maps require. Ordnance Survey still uses graphical survey methods to collect much of its topographic detail. By replacing the drawing board with PIES - a portable interactive edit station (handheld pencomputer) - they have neatly combined survey and digitising into a single operation. It remains to be seen whether they will integrate GPS and/or total station output so as to integrate control and detail survey more or less into a single operation. But this would depend on the acceptance of on-board computing as a suitable alternative for post-processing of control survey.

Cartographic output
Vector digital maps no longer contain any human draughtsmanship - it is now up to the customer’s software to produce a fair plot or screen image.

As customers want to view the same data at different scales, so their software has to be capable of cartographic generalisation. Customers are also in the habit of overlaying datasets that have different spatial resolutions from each other. Improved automated generalisation will help, both here and with attempts to automatically derive maps as small as 1:625,000 scale from the large-scale data, resulting in a spatially consistent suite of map products across all scales.
     All of this will be greatly helped by delivery of the long-promised restructuring of Ordnance Survey vector data into polygons which represent meaningful real-world objects.

Moving it all around
The Internet is set to have a far-reaching effect on the mapping industry. Many corporate GIS are now being implemented using web technology, making it possible for any worker with a desktop PC to run spatial queries using data and software stored elsewhere on the corporate network, possibly without even realising that they are using GIS. Secondly, it is already possible to view topographic maps and images if you know which websites to go to. You can also obtain simple location maps in response to elementary queries such as, "where is the cinema that is showing the latest James Bond movie?’". It does not take a huge leap of imagination to see that this is the way that NLIS will be realised, but there are huge business and legal issues to address that far outweigh the technical hurdles.
     The Internet is already being used for map and image distribution. This is just in its infancy and the pace of development will be dictated by business and legal issues, as much as by technology. Satellite communications are expected to help here. Nevertheless, we can expect to see websites which are data warehouses: these will allow customers to pick and choose not just which dataset they want, but which data classes within it. The concept of a map having a standard specification will disappear. At present three organisations are producing complete or extensive air photo cover of Britain (and some of it is already being sold through the Internet). To a disinterested observer this looks like wasteful competition. Is it possible that mapping photogrammetrists will surf the web for suitable aerial or space imagery before commissioning their own imagery?
     Lastly, survey operations. With web access possible by mobile phone, surveyors can expect to be able to call up map and image files and control data after they arrive on site. Does this spell the end of reconnaissance? Customers can look forward to the day when they get access to surveys just minutes after they’ve been completed. The future is bright, because it will be better mapped than ever before.





Jon Maynard is a member of the Surveying World Editorial Board and a private practitioner specialising in property boundary disputes. He worked formerly for the Ordnance Survey.
Surveying World January/February 2000    31    
 
 

HOME PAGE

  

Top of page