Deep Learning and IoT-Enabled Crop Disease Detection Approach for Precision Agriculture

Plant diseases significantly impact crop quality and yield, necessitating accurate and efficient detection methods. Traditionally, crop disease identification relies on visual inspection by experts, which is Lengthy and susceptible to errors. Recent progress in deep learning, especially Convolutional Neural Network, has demonstrated outstanding performance in image classification tasks. This study introduces a deep learning-driven model for automated tomato leaf disease classification in an IoT-based environment, utilizing Convolutional Neural Network. The model is evaluated on the Tomato Leaves Disease Dataset and is available on Kaggle, comprising 31,275 images of tomato leaves, including nine disease categories and one healthy class. Multiple data preprocessing and augmentation methods were implemented to improve model generalization. The proposed approach effectively classifies plant diseases with an accuracy of 94.70%, demonstrating their robustness and efficiency in real-world agricultural applications. The study comprehensively documents the implementation process, from dataset acquisition to model development, providing a scalable and practical solution for precision agriculture.