Skip to content
Open access

Fine-Grained Tomato Leaf Disease Classification Using Swin Transformer V2 on the Plant Village Dataset

Sep 2026 · International Journal of Innovative Research in Engineering & Multidisciplinary Physical Sciences · 0 citations · 24 references

Abstract

Tomato leaf diseases remain a major cause of yield loss in one of the most widely cultivated vegetable crops, and manual, expert-dependent diagnosis is slow, inconsistent, and often inaccessible in resource-limited farming regions. This study proposes a fine-tuned Swin Transformer V2 model for automated tomato leaf disease classification using the PlantVillage dataset, retaining the full ten-class taxonomy. The model was trained using an ImageNet-pretrained SwinV2-Tiny backbone with cross-entropy loss and label smoothing. On a test split of 2,403 images, the model achieved 99.71% accuracy, with weighted precision, recall, and F1-score of 99.71%, misclassifying only six images. The misclassifications primarily occurred between visually similar disease classes. Compared with related studies using comparable PlantVillage tomato disease taxonomies, the proposed approach achieved higher test accuracy. These results demonstrate the potential of hierarchical vision-transformer architectures for fine-grained tomato leaf disease classification and their possible application in automated agricultural disease diagnosis.

Read PDF

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.