A collaboration between Pula and Tetra Tech under Enabling Crop Analytics at Scale project.
Bird's-Eye project is a collaboration between Pula and Tetra Tech under the Enabling Crop Analytics at Scale Project. This work, funded by the Bill & Melinda Gates Foundation, is meant to foster the collection of Ground truth georeferenced agriculture labelled datasets that would improve predictive analytics in small-holder agriculture, through the use of artificial intelligence and machine learning (AI/ML) and remote sensing.

With accurate crop masks data, yield forecasting will then be possible, of which a ripple effect on mitigation measures and preparedness on food security policy can be achieved.
What is the project about?
There has been a scarcity of training data points (Ground truth data) that can be used for ML.
This data is key to building systems capable of accurately modelling:
A total of 4000 data points have been collected. 1000 out of the 4000 data points include high-resolution drone imagery.
For each data point, the following attributes are recorded:
Geo-tagged farm boundaries.
Crop Yields(MT/Ha).
Crop type cultivation (monoculture, crop rotation, intercropping).
Planting and harvesting dates.
Type of crop.
The project has been implemented in two countries:
1. Zambia
2. Kenya

The project covers the following crops:
1. Maize
2. Potatoes
3. Millet
4. Sorghum
5. Beans