Machines sort cotton seeds by size so scientists can test them in the field
Thanks to a student-developed machine that can “see” and sort seeds by size, a new North Carolina State University project is studying how seed size affects cotton plants' susceptibility to key pests. could shed more light on it. (Photo courtesy of NC State Extension)
RALEIGH, N.C. — Thanks to a student-developed machine that can “see” and sort seeds by size, a new project at North Carolina State University is studying how seed size affects cotton plants' susceptibility to key pests. could shed more light on how it affects
Five fourth-year electrical and computer engineering students completed two semesters of work training a computerized machine to automatically detect seed size and use robotics to separate seeds by size. Did. The team won him first place in the machine learning category of the undergraduate senior design competition.
The students' machines will play a key role in upcoming field trials at the Sandhills Research Station in Montgomery County.
Anders Huseth, an associate professor in the North Carolina Department of Entomology and Plant Pathology and a faculty member in the North Carolina Plant Science Institute, said the study showed that small seeds caused by thrips, a pest that causes crop losses, are more susceptible to larger seeds. Throughout the US Cotton Belt, which said it was designed to determine whether it causes damage.
When a farmer receives a bag of seeds, some contain a mix of large and small seeds, while others contain more stable seeds. Huseth suspects that small seeds and their seedlings may be more susceptible to thrips because the plants grow slower and are more susceptible to injury over longer periods of time. However, this hypothesis has not been tested in practice. Huseth said it's difficult to manually separate seeds by size.
The students solved the dilemma.
How to use
“Their machine was designed to automatically separate seeds by size into labeled wells in custom-printed trays. A conveyor moved the trays into position in front of a barcode reader. , records the ID of the tray,” Huseth explained. “A camera takes an image and calculates the dimensions of each individual seed. The data is compiled into a database containing the dimensions and estimated mass of each individual seed. The tray is moved to the end of the conveyor and adjusted to suit the user. They will be stacked automatically.”
Team members were Jessica Elvegard, Hayden Peek, Gabe Redding, Justin Riley, and James Sober. They created the machine in the NC State Botanical Science Museum's makerspace by 3D printing plastic parts and using other equipment to fabricate the electronics and support structure.
After graduating in May, Elvegard, the team leader, hopes to enter the fields of robotics, mechatronics, materials science, or embedded systems, which combine both computer software and hardware. She said, “This project introduced me to working with embedded systems and reminded me how much fun mechanical design can be.”
Elvegard hopes the machine will accelerate Huseth's research, and Huseth believes it will too.
He said the machine was a “huge step forward for insect research and seed characterization in general.” Using high-throughput phenotyping platforms, this and other NC Plant Sciences Initiative projects aim to advance agricultural research and technology.
Such projects “will help close the loop between entomology and plant physiology from the lab to the field,” Huseth said.
–Dee Shore, North Carolina State University
