A deep learning - driven convolutional neural network framework for automated detection and classification of tomato leaf diseases to enhance precision agriculture and crop health monitoring

Techniques for identifying diseases in tomato leaves involve visual inspection, which is time-consuming, labor-intensive, and prone to human error. This paper suggests a CNN system based on deep learning for disease diagnosis and classification in order to overcome these limitations. Food safety and agricultural productivity are significantly impacted by tomato plants' vulnerability to certain diseases. The suggested method uses CNN architectures and image processing techniques, such as baseline CNN models and transfer learning models like Inception-V3, to reliably forecast a variety of tomato leaf diseases. The data set spans ten distinct disease classes and consists of 18,345 training photos and 3,875 validation images. To improve model performance, preprocessing methods such feature extraction, normalization, and picture augmentation are applied. The field-use monitoring systems will be built using Sphere. Future studies on these crops ought to concentrate on smartphone apps or similar technologies.