Multi-scale closed-loop tuning via spatial frequency collaborative sensitivity for rice leaf disease detection

Yandong Song, Kang An, Lidong Wang, Bin Zhou · 2026

Read Multi-scale closed-loop tuning via spatial frequency collaborative sensitivity for rice leaf disease detection by Yandong Song, Kang An, Lidong Wang, Bin Zhou (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/plos-pone-0351727.txt.

Description

Yandong Song and colleagues engineered a vision model, MCCA-YOLO, to spot rice leaf diseases early, when the marks are faint and farmers' routine spraying either wastes pesticide or arrives too late. Rice feeds more than half the world, so small accuracy gains scale enormously. Their design gives the detector two feature-extracting backbones, a module that sharpens leaf texture in the frequency domain, and a feedback loop that checks its own work across scales.

The architecture targets exactly the failure mode of field photos: tiny, low-contrast lesions on cluttered leaves. Cross-scale fusion lets a pinhead spot borrow context from the whole leaf before the model commits. For extension services and drone scouts, a detector tuned for the earliest, faintest stage is the difference between spot treatment and blanket spraying, with the chemical savings and yield protection that follow.

Catalog record

CREATORYandong Song, Kang An, Lidong Wang, Bin Zhou
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE84940
LICENSECC_BY
RIGHTSCC BY
SOURCEPLOS

Source: https://doi.org/10.1371/journal.pone.0351727

Browse the full library · About the Tower of Babel