Finding cocoa swollen shoot before it spreads
DEMO-13, a cocoa farm at Asuopri Forest Edge in Western North: a farmer's voice note, two drones, an AI first look, a check on the ground, a lab test, and a replanted farm.
The 60-second cut
The same story, shortened: the survey, the AI's first look, the check on the ground and the replanting.
What happens in the film
- Ama sees red veins on new cocoa leaves and sends a voice note to her hub.
- The hub turns it into a case and books a drone survey and a technician visit the same day.
- Two drones go up: one maps the farm, the other carries a hyperspectral camera, in a trial with CRIG, the Cocoa Research Institute of Ghana.
- The AI compares every crown with healthy ones and flags the suspect trees. A flag is a first look, not a diagnosis.
- A technician walks to each flagged tree: red vein banding, swollen branches, mealybugs, the insects that carry the virus.
- A rapid test confirms it in under an hour, a sample goes to CRIG's laboratory, and every step is sealed in the farm's passport.
- Under COCOBOD's programme, and with the farmer's agreement, the infected trees and the trees around them are cut and the stumps treated.
- Tolerant hybrid seedlings go in with plantain for shade and income. Each season the drones fly again.
What the film shows, and what it doesn't
The farm is a real place, drawn from a real drone survey. The trees, people, drones, the disease and the numbers on screen are simulated, and every frame says so.
Finding swollen shoot with a hyperspectral drone is not proven yet: we found no published drone study with ground truth. The pilot plans a trial with CRIG, and the film labels that step "Trial with CRIG · not yet proven".
Nothing is cut on the strength of an image. A technician and a test confirm first, and the farmer decides.
How crop disease detection works · The same steps in the app · Every video
Put the engine to work with us
For partners, investors, government programmes and operators that want to run a pilot. Your message goes straight to the Aionautics team.










