Automatic Approach Based on Deep Learning for Tea Leaf Disease Detection

Tea production is an essential agricultural business, but it is vulnerable to various illnesses that can reduce productivity and quality. This study uses Convolutional Neural Networks (CNNs) to provide a unique technique for automated categorization of tea leaf illnesses. An extensive collection of 900 high-resolution pictures illustrating eight different types of leaf spot (Red leaf spot, Algal leaf spot, Bird's eyespot, grey blight, White spot, Anthracnose, and Brown spot). Tea leaf diseases (blight.) were curated. A cutting-edge CNN model architecture was created and trained to utilize a deep learning framework and obtained 84.5% classification accuracy. The model has standard criteria such as accuracy, precision, recall, and F1-score were used to assess performance. The outcomes are essential for the tea business since the automated disease classification system can give realtime illness categorization. Tea estates can be monitored in real-time for early disease identification and intervention. This may result in increased tea production, improved crop management, and decreased economic losses for tea producers.