Automated Diagnosis of Tomato Leaf Diseases with Attention-Enhanced Feature Aggregation

Tomato crop yield can be enhanced using agricultural technologies if the diseases associated with the plant leaves are detected early. This article proposes a new tomato disease classification model, known as the Multiscale Parallel Feature Aggregation Network with Attention Fusion (MPFAN-AF), which can classify diseases using leaf images. This model comprises parallel convolutional branches that extract multiscale features to retrieve rich data and fuse them through an attention-based fusion module, which performs global average pooling and weights the channels. This model enables the network to optimise patterns associated with the disease while disregarding the diseased area's background noise. The generalisation ability is increased using dropout and L2 weight decay. The refined features are passed through a lightweight multi-layer perceptron for classification. Evaluated on a benchmark tomato leaf disease dataset, MPFAN-AF outperforms conventional Convolutional Neural Networks (CNNs) and existing attention-based models across accuracy, precision, recall, and F1-score. Overall, MPFAN-AF delivers an efficient, accurate, and interpretable solution for automated disease diagnosis in precision agriculture.