Revolutionizing Crop Health: CNN and XGBoost for Accurate Disease Detection in Smart Agriculture

Smart agriculture is revolutionizing conventional agricultural methods by utilizing new technologies for improved crop monitoring and health management. This research introduces a hybrid model that integrates Convolutional Neural Networks (CNN) with XGBoost for precise detection of root rot disease, with the objective of transforming crop health management. The CNN model is employed for automated feature extraction from leaf photos, identifying complex patterns linked to root rot illness. The features are subsequently input into the XGBoost classifier, recognized for its efficiency and accuracy, to deliver reliable illness prediction. Comprehensive experiments utilizing the Coconut Tree Disease Dataset revealed that the proposed hybrid method attained a remarkable average accuracy of 91%, exceeding the performance of conventional machine learning (ML) models, including CNN, MLP, SVM, and XGBoost, which achieved accuracies of 86%,81%, 83%, and 88%, respectively. The suggested framework establishes a basis for a scalable and intelligent plant health monitoring system, promoting sustainable and smart agricultural practices.