Loss-Guided Model Sharing and Local Learning Correction in Decentralized Federated Learning for Crop Disease Classification

TL;DR

This paper proposes a decentralized federated learning framework using validation loss to improve crop disease classification accuracy, convergence, and robustness while preserving data privacy in agricultural settings.

Contribution

It introduces a novel Loss-guided model sharing and local correction method in decentralized federated learning for plant disease detection.

Findings

Improved classification accuracy and convergence speed.

Enhanced generalization and robustness across heterogeneous data.

Effective privacy preservation in agricultural applications.

Abstract

Crop disease detection and classification is a critical challenge in agriculture, with major implications for productivity, food security, and environmental sustainability. While deep learning models such as CNN and ViT have shown excellent performance in classifying plant diseases from images, their large-scale deployment is often limited by data privacy concerns. Federated Learning (FL) addresses this issue, but centralized FL remains vulnerable to single-point failures and scalability limits. In this paper, we introduce a novel Decentralized Federated Learning (DFL) framework that uses validation loss (Loss_val) both to guide model sharing between peers and to correct local training via an adaptive loss function controlled by weighting parameter. We conduct extensive experiments using PlantVillage datasets with three deep learning architectures (ResNet50, VGG16, and ViT_B16), analyzing the impact of weighting parameter, the number of shared models, the number of clients, and the use of Loss_val versus Loss_train of other clients. Results demonstrate that our DFL approach not only improves accuracy and convergence speed, but also ensures better generalization and robustness across heterogeneous data environments making it particularly well-suited for privacy-preserving agricultural applications.