Map generalisation – or cartographic generalisation – is the process of making smaller-scale maps from larger detailed maps. By doing this you can fit a larger geographic area into a smaller dataset.
Map generalisation challenges
It’s not easy generalising geospatial datasets manually. Some of the challenges are:
- Traditional and highly manual production processes produce inconsistent map data.
- High volumes of data are required in different formats for multiple complex products.
- Manual production requires high levels of quality control.
- The currency of the data is affected. There might be a long time between the point the data was captured and the manually finished product.
Why automate the map generalisation process?
When done manually, the generalisation process can take up a lot of time and resource. Automating the process will give you:
- Data is more accurate and up-to-date: You'll reduce production time and increase the speed to market, giving your customers even more value from the data.
- Improved production and product quality through processing data. The different types of processing includes partitioning, national datasets, edge-matching and creating change-only updates.
- Increased consistency and quality of data output: This is perfect for both contextual and analytical products.
- Consistent and current data enables the creation of a consistent zoom-stack at a variety of resolutions.
- Reduced hardware and support costs because you can distribute processing in the cloud.
- Eliminate manual input, freeing up production resources.
Improve your generalisation process
This map data of Sheffield is the original, without generalisation.
Automatically generalised map data. Some buildings have been merged (e.g. see top right).
We provide a 100% automated derived map production for your country's national dataset. We:
- Process the dataset entirely, rather than tile by tile. This produces a machine-readable, fully-analytical data product with national coverage.
- Ensure our process minimises the need for bespoke coding and no manual input or editing is required.
- Ensure edge-matching across adjacent tiles. Often when creating tile-based generalisation, the bigger picture is lost. This causes place names to half disappear or building outlines or roads to not meet. Our process ensures the edges of tiles match, restoring the bigger picture across multiple tiles.
- Include details on environment build and configuration; data load and processing; and we publish the process. This information means we have a 100% fully automated end-to-end production process.
We can also provide consultancy services to advise you on how you can improve generalisation techniques and processes. We’ll work with you to look at:
- Improving data modelling for derived output.
- Delivering scale and performance production for large volumes (national coverage) of data.
- Developing and implementing distributed processing in the cloud
Our latest updates
We understand how important automated geospatial dataset production is for modern national mapping agencies.
Following years of research into automated generalisation, Ordnance Survey has invested in an automated workflow for fast processing of geospatial information.
- Learn more about fast automation for polygon datasets
We have also automated a national scale dataset, with the ability to update nearly half a billion features in less than 12 days.
- Learn about automation of a national vector and raster dataset
Why Ordnance Survey?
- We’ve been working on work-flow automation for almost 20 years.
- We’ve successfully automated our own products.
- You'll get access to our experienced geospatial engineers.
- Your data is stored in the cloud – we don’t process or store any data on our premises.
Want to improve your generalisation processes? Get in touch and we can help you.