An Ensemble Technique for Tea Leaf Disease Prediction Using Machine Learning

The tea growing industry ranks among the major agricultural sectors across the global market yet improper handling of tea leaves deteriorates both product freshness and quality. The paper demonstrates a fundamental discovery solution for detecting tea leaf conditions through SVM alongside RF and ANN, KNN, MLP (Multiple Layer Perceptron) neural network and Hybrid Model machine learning algorithms. Color along with texture and shape features are extracted from tea leaf images using image processing collection methods. The system achieves superior performance throughout the large dataset according to accuracy along with precision, recall measures, and F1-score. Evidence shows the ensemble model produces superior results compared to its independent classifiers making it an effective method for automated disease diagnosis. The paper presents an effective and scalable real-time disease monitoring system serving to support sustainable tea agriculture and smart farming practices.