HISTORICAL
Digital orthoimagery (DOI) appeared in France around 1985 when the national mapping institute IGN implemented a system based on a Scitex scanner and I2S orthorectification equipment. The organization, however could not devote major resources to developing DOI as they were also heavily engaged in establishing a series of national databases. In 1989 IGN carried out a large contract from 1/14,500 color photography for SIVOA, a river valley authority south of Paris.
      Until 1992 the DOI market developed very slowly with IGN carrying out about one or two projects a year. In 1993 the city of Le Havre purchased a full cover of orthoimagery for utilization as a mapbase for vectorization while the Conseil GŽnŽral du Vaucluse (Avignon, South of France) obtained imagery to overlay on their newly digitized cadaster.
      In 1994 IGN encountered its first competition when France's second biggest urban area, Lyon, chose GeomŽditerranŽe to provide full cover color DOI of their 700km2 area. Shortly after, a mining authority in the north awarded a sizeable contract to the Belgian company Eurosense.
      This year things have been generally slower due to local elections although numerous contacts have appeared from Compagnie de la Navigation du Rh™ne. The company IMA/GEO has also obtained its first contracts to cover the Paris suburb of Issy les Moulineaux and a planned expressway route to the west of the capitol.
      Perhaps of more significance is the appearance of a contract from SIVOA to resurvey its territory six years after the first contract. The fact that a French user has deemed DOI a tool worthy of renewal is encouraging news and hopefully an indicator that the market is approaching maturity.

ORTHOIMAGERY USE
Typically most European authorities and companies have been less interested in digital orthoimagery than their North American counterparts. General orthocover has been established only in Belgium and in the autonomous Catalonia province of Spain (at 1/5,000 in both cases).
      In some European countries such as the United Kingdom, the existing digital mapping cover is so detailed that DOI is often considered a luxury. This is not however the case in France where off the shelf digital mapping at even 1/25,000 scale (IGN's B.D. Topo) is still limited. In Figures 1 and 2 it can be seen that both in urban and in rural areas detail is dense enough to justify treatment at larger scales.
      Given this, it should be relatively easy to prove the case for DOI as a complimentary source of digital information especially for administrations that have large land areas to administer, so we are left asking why have so few projects been implemented? Perhaps the fault lies as much with us, the producers as with the users themselves. In our efforts to educate the mapping community we have a tendency to put the accent on explaining the production process, and at least some of us also try not to gloss over the pitfalls. The result is that decision makers realize that DOI is not quite as easy as it looks without really understanding what they might be able to achieve with it.
      So in fact what have French users done with DOI till now?
      Most of them seem to have judged the experience as largely positive. Take SIVOA for instance, an organization administering 33 communes with a population of 350,000 inhabitants over a mixed 200km2 area. After four years of usage in conjunction with the local cadaster, their project officer Mr. Bordoni notes several advantages:
      ¥ DOI is a valuable aid in interpreting and digitizing existing mapping.
      ¥ It supplies information about zones which are poorly mapped (typically river valley bottoms with sometimes dense vegetation).
      ¥ It is readily accessible by different users such as planning and architectural services who are able to work with it using simple image processing software such as Photoshop.
      Michel Pochon, project officer of Le Grand Lyon, a 700km2 area with a population of over 2 million, notes that DOI is an excellent communications tool which enables his organization to illustrate the richness of the city's sizeable digital database. It also serves as an indicator of change and here an interesting experiment was carried out using the same GEOimage raster mapping package that had generated the orthoimagery. The existing building files from the city's GIS were imported to the software and sampled to build up a set of characteristics based on color and texture. A supervised classification was then carried out to locate areas of similar characteristics not covered by the database and, the result displayed in bright pink clearly showed the existence of building shaped structures not included in the original dataset.
      For their part, Le Havre, the largest port in Northern France has demonstrated that even in a European city DOI can be used as a map base enabling effective vectors to be produced for the towns GIS.

CONCLUSION
Digital orthoimagery has been slow off the ground in Europe but this year with backing from the European community several large contracts have been issued. In France the government has not been a front runner in promoting the technology, rather has it been left up to local administrations to justify the choice directly. Judging by the interest we have seen building up through inquiries in the last six months, once the new administrations have settled in later this year there will be a major increase of activity in the '96 budgets and by '98 DOI should be well established as another tool in the GIS kit.

About the Author:
Miles Taylor is with IMA/GEO, a Paris based company specializing in image processing that works closely with GEOMEDITERRANEE on the C™te d'Azur. He can be contacted by telephone at (33 1) 46 72 33 69, by fax at 46 72 42 56.

Back