Tomato Plant Leaf Disease Prediction and Suggestion Using Deep Learning

Detecting plant leaf diseases at an early stage is crucial for a thriving agricultural economy like India, given its need to sustain a large population. The ability to identify and monitor leaf diseases in plants effectively is crucial for disease prediction and the provision of preventive measures. The novelty of this work lies in not only detecting diseases but also suggesting suitable fertilizers. The dataset utilized for this work comprises tomato leaf images against a consistent background obtained from a plant village, featuring ten categories, including nine disease categories and one representing healthy leaves. Three different CNN models, namely AlexNet, InceptionV3, and VGG16, have been used to train the dataset. This approach enables the system to automatically learn and identify disease patterns in plant leaves. The efficacy of the model under consideration, which manifests in an attainment of 89% accuracy, will introduce techniques to identify a broader range of plant diseases.