Transforming Tea Agriculture with AI: A CNN Approach to Leaf Disease Classification

In order to solve the major economic effect and agricultural difficulties presented by tea plant illnesses, this research article investigates the application of Convolutional Neural Networks (CNNs) in tea leaf disease classification. Conventional approaches of disease diagnosis depend on hand inspection, a labour-intensive and prone to human mistake process. By means of image analysis, CNNs provide an automated method to precisely identify and classify tea leaf illnesses. Using a strong dataset from Johnstone Boiyon farm, Koiwa location, Bomet county, the study consists in approx. 1000 photos across eight disease categories. Techniques of preprocessing guarantee diversity and quality of data, so improving model performance. Fine-tuned with transfer learning, the CNN model shows a quite excellent 73% accuracy in disease categorization. This approach emphasizes CNNs' ability to transform tea production by means of real-time disease diagnosis and focused management techniques, hence improving crop health and output.