MobilePlantViT: A Mobile-friendly Hybrid ViT for Generalized Plant Disease Image Classification

2025-03-20

AI Summary

To address the low accuracy and inefficiency of plant disease image classification on mobile and edge devices, this paper proposes a lightweight hybrid Vision Transformer (ViT) architecture. The method integrates CNN-based local feature extraction with ViT’s global contextual modeling capability, incorporating progressive feature distillation, multi-scale data augmentation, and cross-dataset joint training. The resulting model contains only 0.69M parameters and achieves 80%–99% classification accuracy across multiple plant disease benchmarks—significantly outperforming comparably sized lightweight models such as MobileViT. Its novel hybrid backbone uniquely balances local sensitivity and global semantic modeling, enabling robust generalization across diverse crops and diseases. The implementation is open-sourced and optimized for edge deployment.

Technology Category

• Computer Vision: Large Vision Models
• Machine Learning: Learning on the Edge & Model Compression
• Search and Optimization: Learning to Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Web Mining and Content Analysis: Large pretrained models with web data
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

Plant diseases significantly threaten global food security by reducing crop yields and undermining agricultural sustainability. AI-driven automated classification has emerged as a promising solution, with deep learning models demonstrating impressive performance in plant disease identification. However, deploying these models on mobile and edge devices remains challenging due to high computational demands and resource constraints, highlighting the need for lightweight, accurate solutions for accessible smart agriculture systems. To address this, we propose MobilePlantViT, a novel hybrid Vision Transformer (ViT) architecture designed for generalized plant disease classification, which optimizes resource efficiency while maintaining high performance. Extensive experiments across diverse plant disease datasets of varying scales show our model's effectiveness and strong generalizability, achieving test accuracies ranging from 80% to over 99%. Notably, with only 0.69 million parameters, our architecture outperforms the smallest versions of MobileViTv1 and MobileViTv2, despite their higher parameter counts. These results underscore the potential of our approach for real-world, AI-powered automated plant disease classification in sustainable and resource-efficient smart agriculture systems. All codes will be available in the GitHub repository: https://github.com/moshiurtonmoy/MobilePlantViT

Problem

• Develop lightweight AI model for plant disease classification
• Optimize resource efficiency for mobile and edge devices
• Ensure high accuracy in diverse disease datasets

Innovation

• Hybrid Vision Transformer for plant disease classification
• Lightweight model with only 0.69 million parameters
• Optimized for mobile and edge device deployment