AgroPages-Drone images detect two tomato diseases with 99% accuracy-Agricultural news

Drone images detect two tomato diseases with 99% accuracy

New technology being developed by University of Florida scientists identifies two dangerous tomato diseases with 99 percent accuracy. This finding is critical because diseases can cost growers millions of dollars annually in the state’s third most valuable crop. Thus, the earlier farmers detect those ailments, the better their chances of treating them before the diseases cause excessive damage.

Normally, growers walk through their fields and look at their crops to try to find and identify the types of diseases that have befallen their crops. Sometimes disease identification requires expertise that the farmer may lack.

To help tomato growers – and perhaps eventually other farmers – find diseases early and accurately, two UF/IFAS researchers used images from drone technology to see if they could distinguish between two perilous tomato diseases.

“Rapid and accurate identification of a plant pathogen is essential for implementing timely disease management tactics,” said Pamela Roberts, a UF/IFAS plant pathology professor.

For the new study, Roberts and Yiannis Ampatzidis, researchers at the UF/IFAS Southwest Florida Research and Education Center, flew drones over experimental tomato farms at the facility. The unmanned aerial vehicles took images of the crop and, from those pictures, the scientists were able to identify the diseases.

Through technology called a “multilayer perceptron neural network,” the scientists differentiated between bacterial spot and target spot, and did so with nearly perfect accuracy, Roberts said.

This technological advancement is critical for Florida tomato farmers, whose crop is the third most valuable in the state. In Florida, fresh market tomatoes bring in $400 million to $500 million annually, according to the National Agricultural Statistical Service, an arm of the U.S. Department of Agriculture.

Fresh market tomatoes face challenges from numerous diseases in Florida production.

The new UF/IFAS study is published in the journal Precision Agriculture: Detection of target spot and bacterial spot diseases in tomato using UAV-based and benchtop-based hyperspectral imaging techniques