FieldFinder Iowa SPOT

Airbus and Agrimetrics have developed an algorithm based on state-of-the-art artificial intelligence and computer vision technology that automatically delineates agricultural fields visible in a satellite image, producing a layer of polygons to describe them. Using high-resolution imagery, we can provide growers, agri-businesses, retailers and institutions with instant access to a number of territories.

The FieldFinder algorithm uses a deep learning instance segmentation model to extract field polygons from SPOT satellite imagery on-demand.

Supported workflows

Data platform:

• Catalog -> SPOT 6/7 Display -> Order and get the asset from storage of UP42
• In Projects workflows, Processing from Storage -> DIMAP GeoTIFF conversion -> FieldFinder Iowa SPOT

Data blocks:

• SPOT 6/7 Display (Streaming) -> FieldFinder Iowa SPOT

Performance and Geographies

The FieldFinder Iowa model has been developed using agricultural environments typical for Iowa. For the state of Iowa, the Iowa model has achieved a mean field F1 score of 0.78.

Users can apply the Iowa model to other territories with similar agriculture environments but Airbus cannot guarantee the algorithm has the same level of accuracy if applied to other territories.
For Europe, please use the FieldFinder Europe SPOT block.

Input parameter description input data format / query parameters

• Requires a minimum AOI size of 512 x 512 pixels.

Output format description - output data format / error codes

• Field polygons as a shapefile

Algorithm training data

The multi-territory FieldFinder algorithm has been developed to apply a single model to the state of Iowa by using a variety of high-resolution satellite images to ensure a high level of accuracy and to account for different locations that have distinct features. A number of different scenarios were represented in the training data, including: different points in the growing season, all possible terrain types, all possible features that could be encountered (including those unrelated to agriculture), and the use of data augmentation. This was to ensure the training data was of a good quality, so the algorithm produced results with high precision and recall when applied to SPOT imagery, and to reduce the number of false positives by including the identification of non-agricultural features.

Use cases use case blog / case study page

• Traditional field boundary capture methods such as ground survey or digitising from aerial photography, can be very time consuming and expensive to produce. FieldFinder provides delineated fields timely and remotely from your desktop removing inefficiencies associated with manual field boundary data capture.
• Scaling some traditional methods over wide areas can be a prohibitively expensive elongated exercise. FieldFinder provides consistent, good quality field boundaries for an entire country with the same high level of accuracy throughout.
• FieldFinder delineates boundaries using high-resolution satellite imagery providing a reliable source of information, depicting even very small agricultural fields.
• Digital field boundaries are often offered as a dataset and are, therefore, already derived from imagery not chosen by the user. By making FieldFinder available as a processing block, users can choose the input imagery data and the temporal dimensions. Using the FieldFinder algorithm is a dynamic and flexible method for users to access the field boundaries for the specific time in the growing season they require.
• FieldFinder helps users overcome these issues by providing an automatic digital method of boundary delineation using high-resolution satellite imagery. The GIS data output is presented in polygons and compatible with many different agriculture products and applications. This capability can be used to deliver field boundaries to growers, retailers, institutions and developers.
• The algorithm can be developed to delineate boundaries on additional territories and image resolutions. Contact us if the country you require is not available.