Table 4 Summary and description of tomato leaf datasets used in the study.
Dataset name | Web Source | Dataset Description |
|---|---|---|
PlantVillage | https://www.kaggle.com/datasets/abdallahalidev/plantvillage-dataset | A large and diverse dataset with over 54,000 images. It includes 38 plant species and 11,870 tomato samples. The tomato samples cover 10 different disease classes Some disease types are underrepresented which introduce mild class imbalance Well-labelled & organized Images captured under controlled laboratory conditions with uniform backgrounds (white/grey) and consistent illumination |
Tomato Leaves Dataset | Over 20,000 samples of tomato leaves with 10 diseases and 1 healthy class are available Collected from both lab scenes and in-the-wild scenes. This helps the model generalize better for practical use Field images suffering from variable lighting, poor focus are available. Need intensity correction and adjustment | |
PlantDoc dataset | https://www.kaggle.com/datasets/abdulhasibuddin/plant-doc-dataset | Comprises real in-field images of diseased and healthy plant leaves with natural backgrounds and illumination Contains 577 tomato images across 8 classes with separate test samples available Enhances robustness evaluation under non-laboratory conditions |
Tomato-Village dataset | Focuses on nutrient deficiency and viral disorders such as Magnesium Deficiency, Nitrogen Deficiency, Potassium Deficiency, and Spotted Wilt Virus |
