- Land cover maps
- Vegetated areas and water bodies
- Bare soil and infrastructure
- Based on Sentinel-2 imagery
- Up to biweekly updates
- Add-on options with advanced analytics coming soon
The GeoService Silvisense from Science [&] Technology AS provides a detailed insight into the land cover of a selected area. The basic version of Silvisense, the land cover map, delineates forests, water bodies, bare soil / low vegetation and infrastructure. Additional information on forest health will come soon.
Subscribe to the service and get a one time map or retrieve up to biweekly updates for your area of interest. If you like to get more information on the upcoming add-on for forest health analysis, contact us or check the option in your subscription.
Land cover categories
- Do you want to run your analytics only on specific land cover classes?
- Do you need to monitor changes in land cover over time?
Using the differentiation between low / vegetation bare soil, forests, infrastructure, and water bodies, Silvisense is a perfect tool to simplify these tasks and support urban and land planners in their green space development, trails for recreational purposes, and other infrastructure planning applications.
Mask out with Silvisense all land cover you don't want to consider and focus only on the type of land cover you like to further analyse. This will speed up your processing, make your algorithms less complex and the same time the results more reliable.
Consultants and decision makers get a reliable and cost efficient land cover map and a monitoring service to detect changes: Year over year, month over month or even week over week. Silvisense is of special benefit for many forestry applications, from stand exams to watershed assessments and other numerous forest and agriculture-related attributes.
Portugal Wildfires, June 2017
Chile Wildfires, January 2017
The Land Cover Map provides a categorized description of land cover classes based on Sentinel-2 data. More specifically, the classification algorithm is using the RGB and NIR bands of each dataset after masking out any clouds.
The classifier operates in an object-based manner in order to produce smooth and accurate land cover maps. The pixels are first grouped into polygons of similar spectral signature, then the classifier assigns a class label based on an aggregated measure of the whole polygon.
The land cover map can be produced in 10m or 20m spatial resolution. It is delivered as a single-band GeoTIFF. Metadata are included in the band descriptions.
A classification accuracy of more than 90% makes the product highly valuable. The high degree of automation makes it very cost efficient.
Beyond the basic land cover map, more advanced analytics for forest health will become available soon. Please contact us if you are interested. We will connect you with the experts. Just contact us.
Please let us know if you encounter any problems at firstname.lastname@example.org. We are happy to get your feedback.
Pricing / Licensing
CloudEO Terms & Conditions apply.