Every day, drones are used to make people’s lives easier. They have a number of applications across dozens of industries, and are used to make tasks more time efficient, financially efficient, and in some cases, to save lives. Drone’s have the capability to change how we work for the better. To put it simply - drones are tools, not toys.
Agrowing Ltd. and VetorGEO have collaborated to develop an AI for multispectral sensors that can detect when citrus trees are infected. The Huanglongbing (HLB) disease is incurable and detrimental to citrus trees around the globe.
The only strategy for farmers to combat this disease has been to quickly remove infected trees as soon as they begin showing symptoms. However, this has proved ineffective for several reasons. First, visual inspections of the citrus trees are expensive and time consuming. Second, it can take the citrus trees around 6 months - 2 years after contracting the disease before they start showing any symptoms. This means that these trees can be infected and transmitting the HLB disease long before farmers are able to detect physical symptoms.
The AI is developed by uploading hundreds/thousands of images of infected citrus trees. With each additional image that is uploaded, the AI becomes better equipped to detect an infected tree. When the multispectral sensors and AI are equipped to a drone, farmers will be able to detect infected trees at a much earlier stage, even prior to when the human eye can detect symptoms.
“Finding new symptoms, so the infected trees can be removed as quickly as possible, is critical,” said Dr. Kelly Pazolini, a scientific researcher with VetorGEO. “This allows them to both be replaced and limits the damage they would otherwise cause to the rest of the orchard.”
Agrowing Ltd. is currently working on making this AI more universally applicable across the agriculture field. Their goal is to design the AI to be able to detect disease among all sorts of plants and crops, not only the citrus trees.