Rapid Classification and Deep Learning-Based Development Estimation of the Seeds of Helianthus annuus
Fami A. Mume, Daniil S. Ulyanov, Temur R. Muratov, Andrey O. Blinkov, Alina A. Kocheshkova, Sergey M. Avdeev, Pavel Yu. Kroupin, Gennady I. Karlov, Mikhail G. Divashuk · 2026
Read Rapid Classification and Deep Learning-Based Development Estimation of the Seeds of Helianthus annuus by Fami A. Mume, Daniil S. Ulyanov, Temur R. Muratov, Andrey O. Blinkov, Alina A. Kocheshkova, Sergey M. Avdeev, Pavel Yu. Kroupin, Gennady I. Karlov, Mikhail G. Divashuk (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13364111.txt.
Description
Fami A Mume and a large team present a 2026 study on counting developed and aborted seeds on dried sunflower heads from bench photos. Counting by hand eats about an hour of hand work per head. They photographed 1,093 heads under fixed light and marked each seed by class. Then they trained a YOLO detector, locking its confidence setting on a validation set before a fair test on 109 fresh images.
Counts landed within about twelve percent of hand counts on average, though only a small share of heads came within two percent. Bigger models and extra image tricks did not help, and the detector struggled to place seeds in the densest zones. The authors release their data and model, framing the tool as a fast estimate that still needs sharper aim and field-ready versions.
Catalog record
| CREATOR | Fami A. Mume, Daniil S. Ulyanov, Temur R. Muratov, Andrey O. Blinkov, Alina A. Kocheshkova, Sergey M. Avdeev, Pavel Yu. Kroupin, Gennady I. Karlov, Mikhail G. Divashuk |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 38365 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |