Drones and AI used to measure soybean maturity

Drones and artificial intelligence (AI) are now being used to measure soybeans’ maturity in less time while doing so accurately. Current methods require people to spend hours in the sun every day to check the soybeans. The drones have cut the time down to just two days without the need for anyone on the ground.

Researchers at the University of Illinois have turned to drones to improve the efficiency and accuracy of measuring the maturity of soybeans.

Nicolas Martin, assistant professor in the Department of Crop Sciences at Illinois and co-author on the study said:

Assessing pod maturity is very time consuming and prone to errors. It’s a scoring system based on the color of the pod, so it is also subject to human bias. Many research groups are trying to use drone pictures to assess maturity, but can’t do it at scale. So we came up with a more precise way to do that. It was really cool, actually.

To detect most of the soybeans, a drone passes over them a few times taking photos each pass. The images are then compared to one another to see the difference in each one. This was done five times across three seasons in two countries to ensure the drones can produce accurate data.

Rodrigo Trevisan, a doctoral student added:

Let’s say we want to collect images every three days, but one day, there are clouds or it’s raining, so we cannot. In the end, when you get the data from different years or different locations, they will all look different in terms of the number of images and the intervals and so on. The main innovation we developed is how we can account for whatever we are able to collect. Our model performs well independent of how often the data was collected.

The test images were fed into a deep convolutional neural network, that allows it to think similarly to a human brain. The color, shape, and texture of each soybean were pulled from the images and then compared to previously taken images to give a result.

Martin finished with:

We had industry partners on the study who definitely want to use this in the years to come. And they made very good, important contributions. They wanted to make sure the answers were relevant for breeders in the field making decisions, selecting plants, and for farmers.

Photo: Kelly Sikkema