fbpx

Detecting fusarium head blight using a smartphone

30-03-2023 | |
Fusarium head blight causes a lot of economic losses in wheat, and the associated toxin, deoxynivalenol (DON), can cause issues for human and animal health. - Photo: Mark Pasveer
Fusarium head blight causes a lot of economic losses in wheat, and the associated toxin, deoxynivalenol (DON), can cause issues for human and animal health. - Photo: Mark Pasveer

A new University of Illinois project is using advanced object recognition technology to keep toxin-contaminated wheat kernels out of the food supply and to help researchers make wheat more resistant to fusarium head blight, or scab disease.

According to Jessica Rutkoski, assistant professor in the Department of Crop Sciences, part of the College of Agricultural, Consumer and Environmental Sciences (ACES) at Illinois, fusarium head blight causes a lot of economic losses in wheat. “The disease has been a big deterrent for people growing wheat in the Eastern U.S. because they could grow a perfectly nice crop, and then take it to the elevator only to have it get docked or rejected.”

Quantify kernel damage using cell phone images of grains

Increasing resistance is normally done through phenotyping, which is a long, repetitive process. Together with AI experts Junzhe Wu, doctoral student in the Department of Agricultural and Biological Engineering (ABE), and Girish Chowdhary, associate professor in ABE and the Department of Computer Science (CS), Jessica Rutkoski wanted to test whether it’s possible to quantify kernel damage using simple cell phone images of grains.

“A system that could automatically score kernels for damage seemed doable because the symptoms are pretty clear,” Rutkoski said. Algorithms were trained to detect minutely damaged kernels with good enough accuracy using just a few images.

Also read: On-the-go grain quality analysis

Only 60% as accurate as humans

When the team tested the machine learning technology alone, it was able to predict deoxynivalenol (DON) levels better than in-field ratings of disease symptoms, which breeders often rely on instead of kernel phenotyping to save time and resources. But when compared to humans rating disease damage on kernels in the lab, the technology was only 60% as accurate.

The researchers are still encouraged, though, as their initial tests didn’t use a large number of samples to train the model. They’re currently adding samples and expect to achieve greater accuracy with additional tweaking.

Rutkoski says the ultimate goal is to create an online portal where breeders like her could upload cell phone photos of wheat kernels for automatic scoring of fusarium damage.

Join 17,000+ subscribers

Subscribe to our newsletter to stay updated about all the need-to-know content in the agricultural sector, two times a week.

Claver
Hugo Claver Web editor for Future Farming