Research: Automatic tomato sorting using AI

Tomato factories are getting smarter. Researchers in Brazil have developed an artificial intelligence system that can automatically recognize and count tomatoes as they move along conveyor belts, helping producers improve quality control and reduce waste.

Using more than a thousand real images from the field and the factory floor, the system – based on the latest YOLOv11 computer vision technology – learns to distinguish ripe from unripe tomatoes and spot unwanted materials such as soil, rocks, or plant branches. Once trained, it processes video footage in real time, tracking each tomato as it passes and providing instant counts and quality data.

This innovation could transform the way tomatoes are sorted and processed, reducing human error and speeding up production. Beyond factories, the technology could also be used in precision agriculture to assess fruit quality directly in the field on harvester. With accuracy levels approaching 85%, it marks an important step toward a future where AI ensures that only the best tomatoes make it to your plate.

Reference: Cintas Donizette, A., & Rocco, C. D. (2025). Automated Tomato Sorting and Counting Using YOLOv11 for Industrial and Precision Agriculture Applications. Applied Artificial Intelligence, 39(1). https://doi.org/10.1080/08839514.2025.2576891

Full text at: https://www.tandfonline.com/doi/full/10.1080/08839514.2025.2576891