Nonprofit journalism dedicated to creating a Human Age we actually want to live in.

This AI tool predicts a crop’s future growth, health, and yield based on a single image

DAILY SCIENCE

This AI tool predicts a crop’s future growth, health, and yield based on a single image

It could enable precision agriculture and sustainable farming practices, ranging from minimizing fertilizers and pesticides to reducing food waste.
June 21, 2024

Let the best of Anthropocene come to you.

A picture is worth a thousand words, or so the old adage says. Now crop scientists have developed a technology that proves it: their AI-powered tool can accurately predict a crop’s future growth, health, and yields, based on just a single snapshot of the plant. 

This is more than just another AI gimmick, the University of Bonn researchers say: their new tool could help guide farmers on how to raise crops more sustainably without sacrificing yields—for instance, pinpointing places in the field where adding less fertilizer could result in the same productivity, or more. 

To build their tool, they trained a machine learning algorithm—often described as ‘artificial intelligence’ or AI—with thousands of images of growing crops. These images were gathered from previous studies on three different sets of crops, photographed by overhead cameras or drones: thale cress test plants grown in a lab; a field of growing cauliflowers; and a mixed field of wheat interspersed with faba beans. Overall, the researchers plugged more than 100,000 images into the algorithm, which captured the three sets of growing crops over a period of months to years.

Through this data-crunching, the algorithm learned to link certain visual features of a crop in its early stages, with how its growth unfolded over time. The researchers also trained the algorithm to accurately identify specific crop traits, such as leaf area and estimated biomass, which can be linked to yields. 

After this intense training period, the researchers made a striking discovery: if they presented the algorithm with a single image of a crop in the early phases of its growth, it could use this as a foundation to generate dozens of artificial images, which predicted how the crop would look at different stages of its future growth. These artificial images were strikingly accurate, closely comparing to real images of those crops in the field, says Lukas Drees, a doctoral student at the University of Boon, and lead author on the new research. 

 

Recommended Reading:
Can AI make crop production more sustainable?

 

Interestingly, the researchers also found that the algorithm generated more accurate visual predictions of crop growth if it was fed a little bit of extra information about the growing condition of those crops, such as the cultivar and how densely the plants were sown in the field or lab experiment. 

This proof of principle provides a peek at what the model’s immense processing power could achieve. For instance, the algorithm could be supplied with images that show how crops fare under different fertilizer levels. Then, by snapping a single photo of a young crop, a farmer could pick the right course of fertilizer application to ensure the best growth down the line. 

As a tool, the algorithm could make precision agriculture easier to implement on the field, helping to guide more sustainable practices in general. “This applies not only to fertilization, but also for example to irrigation and pesticides,” says Drees. “Instead of applying such valuable resources to the entire field, the model, which has a geo-coordinate, can predict images showing which regions are likely to develop worse, so that targeted and resource-saving interventions can be made.”

The smart imaging may even help farmers to plan the timing of their harvests more accurately, he adds: “This makes it possible to negotiate fair prices and minimize food waste, both of which contribute to sustainability.”

Drees et. al. “Data-driven crop growth simulation on time-varying generated images using multi-conditional generative adversarial networks.” Plant Methods. 2024.

Photo by stevanovicigor/Envato Elements

 

Our work is available free of charge and advertising. We rely on readers like you to keep going. Donate Today

What to Read Next

Anthropocene Magazine Logo

Get the latest sustainability science delivered to your inbox every week

Newsletters

You have successfully signed up

Share This

Share This Article