Drone counts strawberries from the air
Belgian research institute for the manufacturing industry, Flanders Make, has developed a fruit counting drone. The flying robot can count the number of strawberries from the air thanks to artificial intelligence, helping crop growers to predict their harvest more accurately.
Flanders Make developed drones that automatically take stock in a warehouse and an algorithm that can, for example, count the number of flowers and fruits on a piece of agricultural land to make a harvest prediction, using cameras and sensors.
“For this case, we collected aerial images of thriving strawberry crops,” says Rob Heylen, a researcher at Flanders Make. “Our algorithm analysed the images and counted the number of fruits and/or flowers that were visible.”
Text continues underneath images
A Flanders Make researcher flies the drone in a strawberry field. - Photo: Flanders Make
An algorithm analyses the images and counts the number of fruits and/or flowers that were visible. - Photo: Flanders Make
Margin of error
By also doing a manual count during our research, Heylen says they were able to determine a margin of error between the number of digitally recognised flowers and the actual number of flowers. “Because we now know this margin of error, we are able to determine the actual number of flowers very accurately based on quickly collected drone images.”
Text continues underneath video
Predict harvest yield much faster
This application was created in collaboration with the Belgian Experimental Fruit Growing Center. They were already doing manual counts of strawberry flowers to predict the harvest yield of through-bearing strawberries over a 3-week period. Thanks to the drone, this can now be done much faster.
Artificial intelligence key in acceptance of drones
A key factor in the successful acceptance of new drone applications in the business world is the use of artificial intelligence, says Flanders Make. “This technology enables drones to perform tasks completely independently, or to analyse collected imagery at lightning speed.”
To comment, register here
Or register to be able to comment.