Related Concept Videos
You might also read
Related Articles
Articles linked to this work by shared authors, journal, and citation graph.
Sort by
Same author
Segmentation of Plantar Foot Thermal Images Using Prior Information.
Sensors (Basel, Switzerland)·2022
Same journal
Serum vitamin D level and its association with vertigo frequency and severity in Meniere disease.
Scientific reports·2026
Same journal
PFA-Net: a physics-informed feature enhancement and attention network for interpretable bearing fault diagnosis under strong noise.
Scientific reports·2026
Same journal
Circulating inflammatory, redox, and apoptosis-related alterations in drug-naive idiopathic pulmonary fibrosis: an exploratory case-control study.
Scientific reports·2026
Same journal
A baseline-oriented dynamic aggregation approach for demand-side heterogeneous controllable resources.
Scientific reports·2026
Same journal
Temporal precision and accuracy in schizophrenia: an exploratory study.
Scientific reports·2026
Related Experiment Video
Updated: Jul 11, 2025
08:47
Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation
Published on: February 9, 2024
1.5K
Image-based crop disease detection with federated learning.
Denis Mamba Kabala1, Adel Hafiane2, Laurent Bobelin3
1INSA CVL, University of Orleans, PRISME Laboratory EA 4229, 88 Boulevard Lahitolle, 18000, Bourges, France. denis.mamba_kabala@insa-cvl.fr.
Scientific Reports
|November 6, 2023
Summary
Federated learning enhances crop disease classification accuracy using image analysis. ResNet50 models performed optimally in this federated learning approach, outperforming Vision Transformers.
Area of Science:
- Agricultural technology
- Computer science
- Machine learning
Background:
- Automated crop disease detection is vital for agricultural productivity and sustainability.
- Centralized machine learning models face challenges with data privacy, availability, and transfer costs.
- Federated learning offers a decentralized solution to these challenges.
Purpose of the Study:
- To explore the efficacy of federated learning for crop disease classification using image analysis.
- To compare the performance of Convolutional Neural Network (CNN) and Vision Transformer (ViT) models within a federated learning framework.
- To identify optimal model architectures and federated learning parameters for crop disease classification.
Main Methods:
- Utilized federated learning to train CNN (ResNet50) and ViT models on the PlantVillage open-access image dataset.
- Investigated the impact of varying numbers of learners, communication rounds, and local iterations on model performance.
- Analyzed computational time and communication costs associated with different model architectures.
Main Results:
- Federated learning model performance is sensitive to the number of learners, communication rounds, local iterations, and data quality.
- ResNet50 demonstrated superior performance among CNN models in federated learning scenarios.
- Vision Transformers (ViT_B16, ViT_B32) incurred higher computational time, making them less suitable for federated learning.
Conclusions:
- Federated learning is a viable and effective approach for decentralized crop disease classification.
- ResNet50 is a well-suited model for federated learning in crop disease detection due to its performance and efficiency.
- Further research is needed to optimize federated learning strategies for enhanced crop disease management.