University of Florida scientists' newly published research shows artificial intelligence (AI) can improve leaf wetness detection. 

Although Florida's strawberry season starts in December, UF/IFAS works year-round to find ways to manage strawberry diseases.

Continuous moisture and temperatures higher than 65 degrees, combined, give growers a sign that damaging diseases such as botrytis and anthracnose are imminent.

To help growers in the state, the University created the Strawberry Advisory System (SAS), a tool that works with data generated by Florida Automated Weather Network stations near farms and identifies leaf wetness duration to help growers estimate the risk of their fruit getting infected with a fungal disease. Farmers use the tool to know when to spray fungicides to prevent plant diseases. 

Professor of plant pathology Natalia Peres, and Won Suk "Daniel" Lee, professor of agricultural and biological engineering, developed a system that took photos of a reference plate that detects water more directly than the method already used by SAS. 

The data were collected from 30 October 2023 to 29 March 2024, comprising 9,429 images. These images were divided into a 90% training set (including the validation set) with 8,485 images and a 10% test set with 944 images.

The university blog states that "nearly 96% of the time, the algorithm found moisture on the reference plate in comparison with manual observations, and a nearly 84% accuracy rate was observed when comparing with the current sensors and models in SAS."

The study was done at Plant Science Research and Education Center in Citra, the Gulf Coast Research and Education Center in Balm, and at farms in Dover and Plant City. 

Most of Florida’s 13,500 acres of strawberries grow in Hillsborough, Polk, and Manatee counties.