Open Menu Close Menu


Carnegie Mellon Invention Grades Strawberry Plants

Engineers at Carnegie Mellon University have invented a machine that can inspect and grade strawberry plans and then sort them by quality. The invention--sponsored by California strawberry growers and developed by the university's National Robotics Engineering Center (NREC)--uses computer vision and machine learning to do its sorting, which is traditionally performed manually by farm workers. In a successful field test this fall, the machine classified and sorted harvested plants more consistently and faster than workers could, with a comparable error rate.

The United States is the largest strawberry producer in the world. Commercial industry production was estimated to total about $1.9 billion in 2008, according to the U.S. Department of Agriculture.

Commercial berry growers replace their plants every year. During fall harvest, the farms use manual processes to sort several hundred million strawberry plants into good and bad categories. The strawberry plant sorter uses computer vision to examine harvested plants as they pass by on a conveyor belt. Learning algorithms in the software allow the machine to "learn" how to classify strawberry plants of different sizes, varieties, and stages of growth, with more nuance than simply a "good" or "bad" classification.

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

During a 10-day field pilot in October 2009, NREC engineers tested the strawberry plant sorter under realistic conditions, where rain and frost change plants' appearance, and roots can contain mud and debris. On average, the machine sorted 5,000 plants per hour, several times faster than human sorting. Researchers said they hope to boost that speed to 20,000 to 30,000 sorts per hour. While the sorter's overall error rate was close to that of human workers, it inspected and sorted plants more consistently, according to the researchers.

The field trial was the second of a five-phase program to ready the device for commercial use. In phase 3, the engineers will develop better methods to separate harvested strawberry plants for inspection, improve the sorter's robustness and ease of use, and integrate it into the nurseries' harvesting and packaging processes.

"We're looking forward to using the system," said Liz Ponce, CEO of Lassen Canyon Nursery in Redding, CA, one of five strawberry farms sponsoring the project. "All of our stakeholders feel that it has a lot of potential." Sponsors include four other producers, which, together with Lassen, represent about 85 percent of the California strawberry plant nursery market.

About the Author

Dian Schaffhauser is a former senior contributing editor for 1105 Media's education publications THE Journal, Campus Technology and Spaces4Learning.

comments powered by Disqus