NeuralNet AgroGuard: A Deep Learning Approach to Tomato Leaf Disease Detection

This study presents a novel hybrid model combining Convolutional Neural Networks (CNNs) and Random Forests for the accurate leaf diseases prediction, this model is trained on a dataset of 23,371 high-resolution images. The customized CNN model is used for feature extraction from the images, for capturing difficult patterns from various leafs. The extracted features are processed by a Random Forest classifier, leveraging its robustness and capacity to manage high-dimensional data. The images are preprocessed like, including image normalization and rotation; enhance the model's learning capabilities. The proposed model achieved an accuracy of 95%, with individual class evaluations demonstrating strong predictive performance across all categories. The results illustrates that the model not only excels in identifying specific diseases but also maintains consistency and reliability.