Machine could replace plant sorters

PITTSBURGH, Dec. 18 (UPI) -- A machine with computer vision could replace human workers in sorting strawberry plants, engineers in Pittsburgh said.

To maintain good quality, commercial U.S. growers must replace several hundred million strawberry plants each year. To date, the plants have been sorted -- good from the bad -- by humans, engineers at Carnegie Mellon said in a release Thursday.

The engineers have developed a machine that inspects and grades plants, and then sorts them mechanically. The plant sorter uses computer vision and algorithms to teach it to classify plants of varying sizes, varieties and stages of growth, said Liz Ponce, head of Lassen Canyon Nursery, Redding, Calif., one of five strawberry plant producers sponsoring the project.

In a field test this fall in California, the machine classified and sorted plants more consistently and more quickly than workers could, with a comparable error rate, said Christopher Fromme, the project's lead engineer.

"The sorter can adapt to plants that vary from year to year, or even within the same growing season," Fromme said. "It's very flexible."