An efficient non-parametric feature calibration method for few-shot plant disease classification

The temporal and spatial irregularity of plant diseases results in insufficient image data for certain diseases, challenging traditional deep learning methods that rely on large amounts of manually annotated data for training. Few-shot learning has emerged as an effective solution to this problem. T...

Full description

Saved in:
Bibliographic Details
Main Authors: Jiqing Li, Zhendong Yin, Dasen Li, Hongjun Zhang, Mingdong Xu
Format: Article
Language:English
Published: Frontiers Media S.A. 2025-05-01
Series:Frontiers in Plant Science
Subjects:
Online Access:https://www.frontiersin.org/articles/10.3389/fpls.2025.1541982/full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849733690580533248
author Jiqing Li
Zhendong Yin
Dasen Li
Hongjun Zhang
Mingdong Xu
author_facet Jiqing Li
Zhendong Yin
Dasen Li
Hongjun Zhang
Mingdong Xu
author_sort Jiqing Li
collection DOAJ
description The temporal and spatial irregularity of plant diseases results in insufficient image data for certain diseases, challenging traditional deep learning methods that rely on large amounts of manually annotated data for training. Few-shot learning has emerged as an effective solution to this problem. This paper proposes a method based on the Feature Adaptation Score (FAS) metric, which calculates the FAS for each feature layer in the Swin-TransformerV2 structure. By leveraging the strict positive correlation between FAS scores and test accuracy, we can identify the Swin-Transformer V2-F6 network structure suitable for few-shot plant disease classification without training the network. Furthermore, based on this network structure, we designed the Plant Disease Feature Calibration (PDFC) algorithm, which uses extracted features from the PlantVillage dataset to calibrate features from other datasets. Experiments demonstrate that the combination of the Swin-Transformer V2F6 network structure and the PDFC algorithm significantly improves the accuracy of few-shot plant disease classification, surpassing existing state-of-the-art models. Our research provides an efficient and accurate solution for few-shot plant disease classification, offering significant practical value.
format Article
id doaj-art-a46af6f54bb6408ab520175f9327736f
institution DOAJ
issn 1664-462X
language English
publishDate 2025-05-01
publisher Frontiers Media S.A.
record_format Article
series Frontiers in Plant Science
spelling doaj-art-a46af6f54bb6408ab520175f9327736f2025-08-20T03:07:58ZengFrontiers Media S.A.Frontiers in Plant Science1664-462X2025-05-011610.3389/fpls.2025.15419821541982An efficient non-parametric feature calibration method for few-shot plant disease classificationJiqing LiZhendong YinDasen LiHongjun ZhangMingdong XuThe temporal and spatial irregularity of plant diseases results in insufficient image data for certain diseases, challenging traditional deep learning methods that rely on large amounts of manually annotated data for training. Few-shot learning has emerged as an effective solution to this problem. This paper proposes a method based on the Feature Adaptation Score (FAS) metric, which calculates the FAS for each feature layer in the Swin-TransformerV2 structure. By leveraging the strict positive correlation between FAS scores and test accuracy, we can identify the Swin-Transformer V2-F6 network structure suitable for few-shot plant disease classification without training the network. Furthermore, based on this network structure, we designed the Plant Disease Feature Calibration (PDFC) algorithm, which uses extracted features from the PlantVillage dataset to calibrate features from other datasets. Experiments demonstrate that the combination of the Swin-Transformer V2F6 network structure and the PDFC algorithm significantly improves the accuracy of few-shot plant disease classification, surpassing existing state-of-the-art models. Our research provides an efficient and accurate solution for few-shot plant disease classification, offering significant practical value.https://www.frontiersin.org/articles/10.3389/fpls.2025.1541982/fulldeep learningfew-shot learningplant disease classificationfeature calibrationimage classification
spellingShingle Jiqing Li
Zhendong Yin
Dasen Li
Hongjun Zhang
Mingdong Xu
An efficient non-parametric feature calibration method for few-shot plant disease classification
Frontiers in Plant Science
deep learning
few-shot learning
plant disease classification
feature calibration
image classification
title An efficient non-parametric feature calibration method for few-shot plant disease classification
title_full An efficient non-parametric feature calibration method for few-shot plant disease classification
title_fullStr An efficient non-parametric feature calibration method for few-shot plant disease classification
title_full_unstemmed An efficient non-parametric feature calibration method for few-shot plant disease classification
title_short An efficient non-parametric feature calibration method for few-shot plant disease classification
title_sort efficient non parametric feature calibration method for few shot plant disease classification
topic deep learning
few-shot learning
plant disease classification
feature calibration
image classification
url https://www.frontiersin.org/articles/10.3389/fpls.2025.1541982/full
work_keys_str_mv AT jiqingli anefficientnonparametricfeaturecalibrationmethodforfewshotplantdiseaseclassification
AT zhendongyin anefficientnonparametricfeaturecalibrationmethodforfewshotplantdiseaseclassification
AT dasenli anefficientnonparametricfeaturecalibrationmethodforfewshotplantdiseaseclassification
AT hongjunzhang anefficientnonparametricfeaturecalibrationmethodforfewshotplantdiseaseclassification
AT mingdongxu anefficientnonparametricfeaturecalibrationmethodforfewshotplantdiseaseclassification
AT jiqingli efficientnonparametricfeaturecalibrationmethodforfewshotplantdiseaseclassification
AT zhendongyin efficientnonparametricfeaturecalibrationmethodforfewshotplantdiseaseclassification
AT dasenli efficientnonparametricfeaturecalibrationmethodforfewshotplantdiseaseclassification
AT hongjunzhang efficientnonparametricfeaturecalibrationmethodforfewshotplantdiseaseclassification
AT mingdongxu efficientnonparametricfeaturecalibrationmethodforfewshotplantdiseaseclassification