Parcel- or pixel-level crop-type maps for 28 European countries, made by deep learning on Sentinel-2 time series, for agribusiness and public bodies

Create a free account to discover solutions related to this one.
Sign up for freeAlready have an account? Log in
Create a free account to discover solutions related to this one.
Sign up for freeAlready have an account? Log in
Nimbo Crop ID is Kermap's crop-type mapping product, sold under its Nimbo brand. Official parcel registers arrive late: in France the RPG is published about a year after the season, so agri-businesses, public bodies and sustainability programmes lack timely data on what is grown where. Kermap's deep-learning models classify crops from dense Copernicus Sentinel-2 time series at 10 m resolution, with data available from the end of the growing season. The product covers 28 European countries, with the rest of the world on demand, and distinguishes 22 crop classes in Europe and up to 34 in France, where winter and summer crops are separated. The minimum detection area is 0.15 ha.
Results come at pixel level as TIFF, for production-area statistics at regional, national or continental scale, or along field geometries as FlatGeobuf. The geometries can be the buyer's own parcels, official registers such as the RPG, or Kermap's AI-derived field boundaries. Samples and area statistics can be explored in Nimbo Earth Online; data is ordered by quote request, and an API integration is offered with support from the Nimbo team. Kermap's specification sheet gives an average detection accuracy of 86%, and the company reports 2 billion hectares monitored since 2018.
Kermap crosses its parcel-level crop identification with irrigation and soil-cover indicators in EPIC, a platform delivered with partners to the French State under France 2030 (2024-2026). It also supplied crop identification for two seasons to the public water utility Eau du Grand Lyon.
KermapEvaluating this solution?
Create a free account to save solutions, compare alternatives, and request introductions to providers.
Sign up for freeQuote-based
Pricing is not published; the provider quotes per customer or project (contact sales, request a quote).
Space Capability
Processing
Insight Generation
Converts analysis into context-rich, actionable insights.
Delivered as
System Integration
Geographic Scope

Kermap
Kermap is a French geospatial AI company, founded in 2017 and based in Cesson-Sévigné (Brittany), that turns satellite and airborne imagery into geographic information using deep learning. It produces land-use, vegetation, crop and change data, spatial analyses and interactive map platforms, notably Nimbo (monthly cloud-free global satellite basemaps and crop analytics) and Nos Villes Vertes (urban trees). Its customers include regional and local authorities such as Région Grand Est and Rennes Métropole, public bodies such as France's agriculture ministry and INRAE, and companies such as Nestlé France and McCain, across agriculture, environment and urban planning.
2017
Founded
11-50
Team size
France
Country
Embeds data and insights into existing systems and workflows.