Advancing Tomato Crop Health: A CNN Approach to Leaf Disease Detection
Bhoomika Bhoomika, Goldy Verma
Abstract
One of the most grown crops worldwide, tomato plants are quite vulnerable to many leaf diseases that could seriously lower quality and productivity. Effective control and reduction of the use of chemical pesticides depend on early identification of these illnesses. The implementation of a CNN model for identifying several tomato leaf diseases is investigated in this research article. The accuracy of 95% is attained with 994 images dataset split into 10 different classes. By the help of Transfer learning, the CNN model became optimized that enabled improved performance even for a limited dataset. Confirming the model’s great accuracy, recall, and F1-scores across all classes was the confusion matrix. This work emphasizes CNNs’ promise as a consistent tool for early and accurate tomato leaf disease diagnosis, hence improving crop management and environmentally friendly farming methods.