Generalisation and Multiple Representation

Production, Generalization, and Conflation

6704.4 - Producing maps from multiple sources using automated data conflation, generalisation and advanced styling

Thursday, July 6
5:10 PM - 5:30 PM
Location: Virginia B

Volunteered Geographic Information (VGI), such as OpenStreetMap data (OSM), has proven its worth many times over in relief & emergency situations. This is primarily due to its currency, but also to its ever-increasing completeness and accuracy. This has made high-level providers of geographic information, NMA’s and commercial publishers, start looking at using VGI as a data source, in order to be able to better deal with the growing demands (currency) and constraints (budgets) they are faced with.
Beyond the use case of updating a topographic database, ingesting VGI or other external data offers the opportunity to quickly produce thematic maps. With the push from government agencies to open up their data, more and more authoritative data are available, but often in forms that are hard to interpret. Integrating them with topographic data and making a map to present them is a very powerful way to bring their value out and give it some visibility.
To do so, the first challenge is to integrate the data in a clean way. This process is often referred to as “data conflation”. In the context of bringing new thematic data in, there are two main scenarios: the new data brings some additional information to features already present in the topographic database, or the data contains entirely new features. In the former case, a matching needs to be performed to identify which feature needs to be enriched by what bit of information, usually by adding new attributes to it. In the latter, new features are created, but they often need adjusting to ensure consistency between layers. If the aim of the conflation is to update a layer, then the challenge is again to perform an effective matching to identify features which have been added, removed or changed, and update the topographic data accordingly. The key to a successful conflation process is to have access to the rules that guide the matching, and the transfer of the data, as these are specific to each case.
Once the dataset has been enrichened, the next difficulty is to make it fit for portrayal. The conflation is usually done with high resolution data, but the result is often best presented at smaller scales, so that a larger geographic extent can be seen on screen or on a standard size sheet of paper. Generalisation is a complex problem, hard to achieve using standard GIS. Some tools to perform specific generalisation tasks are often present, but automating their use requires expertise and heavy configuration.
Finally, the visual impact of a map depends on the styling. Good styling that highlights the important information and relevant context is key to producing a map fit for its purpose.
At 1Spatial, we have made a proof of concept using a combination of 1Integrate, 1Generalise and 1Publish, to update a topographic dataset from the Dutch Kadaster using OpenStreet Map, and transfer some attribution about speed limits that was available on OSM. We then have used the enhanced dataset to produce a small scale map of speed limits in a Dutch city. All this was done automatically, after having configured the software for each step in the process.
The next exciting step will be to cut down on the configuration required for such process. For this we’ll use the capability of our software to propose preconfigured solutions that can be tweaked, so that the configuration remains simple and still offers great flexibility. We’ll then look into making it available as a service, so that potential users can use it as and when they need it, and only pay for what they get out of it.

Nicolas Regnauld

Product Manager
1Spatial

Nicolas Regnauld has many years of experience in researching solutions for automating generalisation. After a PhD on generalisation at IGN (Paris), and a post Doctorate at the University of Edinburgh, he led the research team on automated generalisation at Ordnance Survey for 11 years. During this time, his team successfully developed a fully automated generalisation process that was used for the creation of OS VectorMap District. Nicolas is now Product Manager at 1Spatial, responsible for 1Generalise and 1Publish. These two products form the core of 1Spatial map derivation solutions, for deriving high quality maps from large geospatial databases.

Presentation(s):

Send Email for Nicolas Regnauld

Paul Duré

Senior account manager
1Spatial

Paul Duré has been working in the domain of high-end geospatial production since 1993, first as trainer-consultant and later as manager of the Mercator cartographic software (Barco / STAR-APIC). In his present position of senior account manager with 1Spatial, he continues to serve mapping agencies, publishers and other organisations that need productive and reliable systems for the production of data and mapping with advice and assistance. Paul can be contacted at paul.dure@1spatial.com.

Presentation(s):

Send Email for Paul Duré

Nicolas Regnauld

Product Manager
1Spatial

Nicolas Regnauld has many years of experience in researching solutions for automating generalisation. After a PhD on generalisation at IGN (Paris), and a post Doctorate at the University of Edinburgh, he led the research team on automated generalisation at Ordnance Survey for 11 years. During this time, his team successfully developed a fully automated generalisation process that was used for the creation of OS VectorMap District. Nicolas is now Product Manager at 1Spatial, responsible for 1Generalise and 1Publish. These two products form the core of 1Spatial map derivation solutions, for deriving high quality maps from large geospatial databases.

Presentation(s):

Send Email for Nicolas Regnauld


Assets

6704.4 - Producing maps from multiple sources using automated data conflation, generalisation and advanced styling



Attendees who have favorited this

Please enter your access key

The asset you are trying to access is locked. Please enter your access key to unlock.

Send Email for Producing maps from multiple sources using automated data conflation, generalisation and advanced styling