Machine Learning-Based Prediction of Bacterial Diseases in Bell Pepper Leaves

An effective and automated detection system is required because bacterial infections in bell pepper leaves have a substantial impact on crop output and quality. Real-time monitoring is challenging due to the laborintensive and inaccurate nature of traditional identifying techniques. Using a dataset of $\mathbf{5, 0 1 5}$ photos of bell pepper leaves, this study uses machine learning techniques tocreate an automated illness prediction system. We have extracted the features from the image dataset instead of regular statistical data for prediction. Ten machine learning models were trained and assessed, including XGBoost, Random Forest, SVM, and Decision Tree. An ensemble model that combined all classifiers scored 94% accuracy, indicating its robustness, while the bestperforming individual modelgot $98 \%$ accuracy. The suggested system provides a scalable and dependable precision agricultural solution by facilitating early identification and efficient disease management. The findings demonstrate the potential of machine learningbased methods to increase agricultural output, decrease human labor, and improve plant disease diagnostics.