Automated stomatal traits measurement in melon (Cucumis melo L.) based on vision transformers with dynamically composable multi-head attention

Yao Huang, Lelong Yan, Jiaxi Yang, Chunyan Cheng, Yuhui Wang, Ji Li, Qunfeng Lou, Yongbing Zhang, Chuntao Qian · 2026

Read Automated stomatal traits measurement in melon (Cucumis melo L.) based on vision transformers with dynamically composable multi-head attention by Yao Huang, Lelong Yan, Jiaxi Yang, Chunyan Cheng, Yuhui Wang, Ji Li, Qunfeng Lou, Yongbing Zhang, Chuntao Qian (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/PMC13555868.txt.

Description

Yao Huang leads a nine-author team automating measurement of leaf pores in melon plants in a 2026 study. These pores, called stomata, govern how plants breathe and lose water, yet counting them by hand is slow. The team built a melon image set with more than eight thousand labelled pictures. Their model pairs a vision transformer with attention heads that swap information flexibly, plus a module sharpening channel detail.

Detection precision ran high on the melon test set, topping ninety percent at loose overlap. Oval fitting then measured pore length, width, count, and area automatically. Those measures agreed closely with hand measurements by expert counters. The model also transferred fairly to cucumber, watermelon, pumpkin, and loofah, though each test set was small.

Catalog record

CREATORYao Huang, Lelong Yan, Jiaxi Yang, Chunyan Cheng, Yuhui Wang, Ji Li, Qunfeng Lou, Yongbing Zhang, Chuntao Qian
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE66112
LICENSEhttps://creativecommons.org/licenses/by/4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/article/MED/PMC13555868

Browse the full library · About the Tower of Babel