U-FedTomAtt: Ultra-lightweight Federated Learning with Attention for Tomato Disease Recognition

TL;DR

This paper introduces U-FedTomAtt, an ultra-lightweight federated learning framework with attention mechanisms, designed for accurate tomato disease recognition on resource-limited edge devices across distributed farms.

Contribution

The paper presents a novel ultra-lightweight neural network with dilated bottleneck modules and a linear transformer, along with a federated dual adaptive weight aggregation algorithm, improving accuracy and efficiency in federated tomato disease diagnosis.

Findings

Achieves over 99% accuracy on benchmark datasets.

Uses only 245K parameters and 71 MFLOPS, suitable for resource-constrained devices.

Demonstrates effective federated learning in distributed agricultural environments.

Abstract

Federated learning has emerged as a privacy-preserving and efficient approach for deploying intelligent agricultural solutions. Accurate edge-based diagnosis across geographically dispersed farms is crucial for recognising tomato diseases in sustainable farming. Traditional centralised training aggregates raw data on a central server, leading to communication overhead, privacy risks and latency. Meanwhile, edge devices require lightweight networks to operate effectively within limited resources. In this paper, we propose U-FedTomAtt, an ultra-lightweight federated learning framework with attention for tomato disease recognition in resource-constrained and distributed environments. The model comprises only 245.34K parameters and 71.41 MFLOPS. First, we propose an ultra-lightweight neural network with dilated bottleneck (DBNeck) modules and a linear transformer to minimise computational and memory overhead. To mitigate potential accuracy loss, a novel local-global residual attention (LoGRA) module is incorporated. Second, we propose the federated dual adaptive weight aggregation (FedDAWA) algorithm that enhances global model accuracy. Third, our framework is validated using three benchmark datasets for tomato diseases under simulated federated settings. Experimental results show that the proposed method achieves 0.9910% and 0.9915% Top-1 accuracy and 0.9923% and 0.9897% F1-scores on SLIF-Tomato and PlantVillage tomato datasets, respectively.