DETECTION OF COFFEE LEAF DISEASE USING DEEP LEARNING: A CASE STUDY OF WORAMIT AGRICULTURAL RESEARCH CENTER IN AMHARA REGION

Coffee leaf disease refers to a range of diseases that affect the leaves of coffee plants, causing damage to the plant and potentially leading to significant losses in coffee production. Some of the most common coffee leaf diseases include: Coffee Leaf Rust, Coffee Berry Disease, and Coffee Wilt Disease. These diseases can cause symptoms such as leaf spots, premature leaf drop, wilting, and berry lesions, ultimately affecting the plant's health and productivity. due to this,we have developed a detection of coffee leaf disease using a deep learning approach: A Case Study of Woramit Agricultural Research Center in the Amhara Region, Ethiopia. A VGG16 deep learning model is employed to identify common diseases affecting coffee plants, such as Coffee Berry Disease (CBD), Coffee Leaf Rust, and Coffee Wilt Disease. These diseases severely impact coffee yield and quality, threatening the livelihoods of farmers in the region. The dataset for this study was collected from the Woramit Kebele and was preprocessed to enhance accuracy and minimize errors. Data augmentation techniques, such as zooming, rotation, and flipping, were applied to expand the dataset and improve the model’s generalization capabilities. A random search algorithm was used for hyper parameter optimization, and the model’s performance was evaluated using various metrics, including accuracy, precision, recall, and F1-score.The study compared the performance of the VGG16 model with five other machine learning algorithms: Artificial Neural Networks (ANN) at 92.4%, Decision Tree (DT) at 93.9%, Logistic Regression (LR) at 89.4%, Random Forest (RF) at 95.5%, and Support Vector Machine (SVM) at 89.4%. The results indicate that VGG16 outperformed the other methods with an impressive accuracy of 99.87%. The results demonstrate a high accuracy of 99.87% in detecting coffee leaf diseases, showcasing the effectiveness of the VGG16 model in improving coffee disease management strategies. The study contributes to the broader agricultural field by providing a robust method for early detection and intervention in coffee leaf diseases, with potential applications for enhancing productivity and disease management across other coffee-growing regions