DAPR-AM-Net: an end-to-end smart farming system powered by dual-attention progressive refinement and adaptive MixUp for explainable tomato leaf disease classification and forecasting

Ran Wang, Xiao Yu, Lina Lu, Cong Chen · 2026

Read DAPR-AM-Net: an end-to-end smart farming system powered by dual-attention progressive refinement and adaptive MixUp for explainable tomato leaf disease classification and forecasting by Ran Wang, Xiao Yu, Lina Lu, Cong Chen (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/PMC13455238.txt.

Description

Ran Wang, Xiao Yu, Lina Lu, and Cong Chen present, in a 2026 study, a smart farming system that names tomato leaf diseases from photos and explains its calls. Field diagnosis is hard: backgrounds clutter the frame, lesions look alike, and rare diseases get few training photos. Their network stacks four fixes for those problems.

Twin attention blocks steer the model toward diseased patches and away from soil and sky. A smart mixing method blends training photos in line with those attention maps, sharpening fuzzy boundaries. Features are refined step by step across layers, and the training loss reweights itself toward rare classes. On the team's own tomato dataset the system scored about 99.7 percent accuracy, and about 99.9 percent on a standard public set. The authors pitch it as a compact, explainable tool for growers.

Catalog record

CREATORRan Wang, Xiao Yu, Lina Lu, Cong Chen
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE89263
LICENSEhttps://creativecommons.org/licenses/by/4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/article/MED/PMC13455238

Browse the full library · About the Tower of Babel