Attention mechanisms and transfer learning for robust peach leaf damage classification under domain shift Journal: Expert Systems with Applications
Adrián Cánovas-Rodriguez, Miguel A. González-Illán, Maria Fernanda García-Cruz, Pedro Nortes Tortosa, José Salvador Rubio-Asensio, Miguel A. Zamora Izquierdo, Juan Antonio Martínez Navarro, Antonio F. Skarmeta · 2026
Read Attention mechanisms and transfer learning for robust peach leaf damage classification under domain shift Journal: Expert Systems with Applications by Adrián Cánovas-Rodriguez, Miguel A. González-Illán, Maria Fernanda García-Cruz, Pedro Nortes Tortosa, José Salvador Rubio-Asensio, Miguel A. Zamora Izquierdo, Juan Antonio Martínez Navarro, Antonio F. Skarmeta (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/arxiv-2606-02045.txt.
Description
Cánovas and co-authors measure how crop models fail when they leave the lab. Models for leaf damage are trained on neat public photos. Real orchards look quite unlike those curated sets. Peach leaves show stress from heat, drought, and pests. The signs of each cause can look much the same.
The team built a set of 1,366 peach leaves by hand. Six damage types are covered in that set. A second set holds 180 field photos from a real orchard. It serves as an unseen target for the models. Eleven network backbones and three attention variants are compared. One model tops the source tests with 93.3 percent accuracy. On field photos, scores drop by 0.21 F1 points on average.
Catalog record
| CREATOR | Adrián Cánovas-Rodriguez, Miguel A. González-Illán, Maria Fernanda García-Cruz, Pedro Nortes Tortosa, José Salvador Rubio-Asensio, Miguel A. Zamora Izquierdo, Juan Antonio Martínez Navarro, Antonio F. Skarmeta |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 71 KB |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2606.02045