How it works
AfriScan combines Earth observation, GIS, weather and environmental information with suitable AI-assisted analysis. The technology follows the question, not the other way around.
Four clear stages
We begin with the farm, landscape or project boundary. Then we clarify the change, condition, risk or reporting need the analysis must support.
We identify suitable satellite observations, spatial data, weather or forecast context, client information and analytical methods. AI or machine learning is used where it adds value and can be interpreted responsibly.
We analyse spatial and temporal patterns, compare places, dates or scenarios and check whether the evidence is strong enough to support the intended interpretation. Deep-learning models are being integrated into CropIntel for more advanced pattern analysis as suitable validation data become available.
You receive clear maps and explanations that distinguish observed conditions, forecasts and model-generated estimates, showing what was observed, what may happen next and what may need closer review.
Satellite observation
Interpretation matters
Cloud, data resolution, timing, crop type and field conditions can affect interpretation. AI or model output does not automatically establish a diagnosis. AfriScan states dates, model status and limitations clearly and recommends field verification when the cause cannot be established remotely.
Begin with the basics
