A novel lightweight hybrid CNN-ViT for maize leaf disease classification.

Abstract

Maize is a vital global crop, but its productivity is often threatened by plant diseases, highlighting the need for precise and timely diagnostic methods. Traditional manual inspection is inefficient and prone to errors, motivating the development of automated solutions. Recent advances in computer vision and deep learning have enabled effective automated plant disease diagnosis. While Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs) have shown promise in plant disease classification, CNNs struggle to capture global contextual information, and ViTs require large datasets and high computational resources. Inspired by mixture-of-experts (MoE) architectures, we propose a lightweight hybrid model that integrates CNN and ViT components, adaptively emphasizing local or global features based on input characteristics. Evaluated on a novel, real-world dataset of full maize plant images, our approach achieves 99.90% classification accuracy, significantly outperforming state-of-the-art baselines such as MobileViT, PiT, EdgeNeXt, and DeiT. These results demonstrate that lightweight hybrid architectures can deliver high-performance disease diagnosis suitable for practical agricultural deployment. The code is available at: https://www.github.com/sabermehdipour/MXiT .

Plant phenotyping relevance

トウモロコシ全身画像から病害状態を推定する軽量CNN-ViT手法を開発・評価しており、植物表現型取得・判定が中心的です。

Code and data availability

The paper's authors' MXiT analysis code is publicly available via a GitHub URL stated in the abstract, and the PlantVillage image dataset used for evaluation is publicly available. The Plant Scanner maize dataset is paper-specific but only available upon request, so it is listed as request_only.