Strawberry plant sorter outperforms workers

Liz Ponce, CEO of Lassen Canyon Nursery in Redding, Calif., is looking forward to using a plant sorting machine developed by researchers at Carnegie Mellon Univ.’s Nat’l. Robotics Engineering Center. The machine uses computer vision and machine learning to inspect and grade harvested strawberry plants. It then mechanically sorts them by quality. These tasks could previously be done only manually.
In a successful field test conducted in October, the machine classified and sorted harvested plants more consistently and faster than workers could with a comparable error rate. The machine’s learning algorithms allow it to be taught how to classify the plants of different sizes, varieties and stages of growth.
The NREC project is being funded by 5 strawberry plant producers that represent about 85% of the California strawberry plant nursery market. In addition to Lassen Canyon, the sponsors include Driscoll Nursery Associates, Nor Cal Nursery, Plant Sciences and Crown Nursery.
To maintain good strawberry yields, commercial berry growers replace their plants every year. During the fall harvest season, strawberry plant nurseries use manual labor to sort several million plants into good and bad categories.