Quantifying Domain Shift in Peach Leaf Damage Classification: An Evaluation of Attention Mechanisms and Fine-Tuning Strategies

Leaf damage classifiers are usually trained and evaluated on public images collected elsewhere, and their behaviour in a new orchard is rarely measured. We quantified this gap for peach leaves. Eleven ImageNet-pretrained convolutional backbones, each with and without a Convolutional Block Attention Module (CBAM), were trained on 1415 public leaves (six classes). They were evaluated, with photograph-grouped cross-validation and confidence intervals, on 182 leaves from 122 photographs taken in a commercial orchard (four classes). On the public hold-out set, macro-F1 ranged from 0.670 to 0.772, and the ranking of the best models was not resolved; CBAM brought no improvement (pooled change −0.009 to −0.001). Without retraining, mean orchard macro-F1 was 0.634, against 0.719 on the same four public classes, but only after masking the two classes absent from the orchard; without masking, one-third of the leaves were assigned to them. Label-free normalisation-statistics alignment did not help, whereas fine-tuning raised the mean macro-F1 to 0.72, and a few tens of labelled leaves sufficed to match zero-shot transfer. On the shared public classes, full fine-tuning lost more source performance than linear probing, and joint training preserved it. Light models and a modest site-specific annotation effort are sufficient for this task.

Authors

Institutions

Publication Details

Journal
Applied Sciences
Published
2026-11-17
DOI
https://doi.org/10.3390/app162211400
Primary Topic
Smart Agriculture and AI
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Quantifying Domain Shift in Peach Leaf Damage Classification: An Evaluation of Attention Mechanisms and Fine-Tuning Strategies

J.S. Rubio-Asensio, P. Tortosa, Adrián Cánovas-Rodríguez, Antonio F. Skarmeta et al.
4 links
Applied Sciences
Smart Agriculture and AI
article

Quantifying Domain Shift in Peach Leaf Damage Classification: An Evaluation of Attention Mechanisms and Fine-Tuning Strategies

J.S. Rubio-Asensio, P. Tortosa, Adrián Cánovas-Rodríguez, Antonio F. Skarmeta, Juan Antonio Martínez Navarro, Maria Fernanda García-Cruz, Miguel A. González-Illán, Miguel Ángel Zamora Izquierdo
article en

Abstract

Leaf damage classifiers are usually trained and evaluated on public images collected elsewhere, and their behaviour in a new orchard is rarely measured. We quantified this gap for peach leaves. Eleven ImageNet-pretrained convolutional backbones, each with and without a Convolutional Block Attention Module (CBAM), were trained on 1415 public leaves (six classes). They were evaluated, with photograph-grouped cross-validation and confidence intervals, on 182 leaves from 122 photographs taken in a commercial orchard (four classes). On the public hold-out set, macro-F1 ranged from 0.670 to 0.772, and the ranking of the best models was not resolved; CBAM brought no improvement (pooled change −0.009 to −0.001). Without retraining, mean orchard macro-F1 was 0.634, against 0.719 on the same four public classes, but only after masking the two classes absent from the orchard; without masking, one-third of the leaves were assigned to them. Label-free normalisation-statistics alignment did not help, whereas fine-tuning raised the mean macro-F1 to 0.72, and a few tens of labelled leaves sufficed to match zero-shot transfer. On the shared public classes, full fine-tuning lost more source performance than linear probing, and joint training preserved it. Light models and a modest site-specific annotation effort are sufficient for this task.

Applied SciencesVol. 16(22)
Consejo Superior de Investigaciones Científicas (ES), Centro de Edafología y Biología Aplicada del Segura (ES), Universidad de Murcia (ES)
European Commission, NextGenerationEU
Zero hunger
Smart Agriculture and AI
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon