The proliferation of low cost, high latency satellite data has ushered in an era of innovation across a number of critical industries and is especially useful for observing phenomena in agriculture. These collections of planet-scale data provide comprehensive coverage at frequent intervals, high resolutions, across multiple sensor types, and in multiple spectral bands. This data is crucial for applications such as detections relevant to monitoring change in agricultural and land-resource based systems.

Similarly, current commercially available AI capabilities to recognize and predict features in satellite imagery are generally trained on images of real world, terrestrial objects. These tools were not designed for the inherent patterns and complexities found in satellite imagery. As a result, existing approaches often do not perform well when applied to satellite data.

A number of R&D efforts have chipped away at this problem in the geospatial realm by creating labeled training datasets for geospatial AI research. These initiatives released valuable training datasets, yet there are still challenges that need focused attention to create the landscape for a commercial boom in geospatial AI:

• Training data efforts are local or regional scale, usually in Europe and the US
• Data is from many different sources, lacking interoperability:
• Training datasets are released without community feedback loops or tools for other contributions:

AI/ML experts whose research uses earth observation data for understanding agricultural issues such as food security have noted that satellite imagery should be considered a distinct modality and identified the urgent need for technical development to bridge the gap between AI capabilities, satellite and earth observation data, and the needs and opportunities of practitioners.

The Taylor Geospatial Engine’s Innovation Bridge for Field Boundaries seeks to contribute to the collective effort to advance the state of the art in the application of AI and computer vision for agricultural field boundaries by bringing together world renowned experts in data science, artificial intelligence, satellite imagery with stakeholders who stand to benefit from the this work.

Cloud-Native
Geospatial
Foundation