Revolutionizing Blueberry Mosaic Disease Severity Classification with CNN-AdaBoost Hybrid Model

The proposed study introduces a new way of measuring how bad Blueberry Mosaic Disease (BMD) is using an advanced learning method that combines Convolutional Neural Networks (CNN) with AdaBoost. A carefully selected collection of 12,000 blueberry leaf pictures showing different levels of BMD damage is used as the base. Experts in the field carefully prepared and labeled data to help teach from examples. The mix of Deep Learning shows very good results, getting an all-around rightness of 98.20/0 in sorting out how risky BMD is. In Table I, performance measures such as precision rate, recall percentage, and F1 score tell how good the model is at telling different levels of seriousness apart. The Confusion Matrix helps to see what the model does well and where it needs improvement. This model's success in being right about the levels of BMD shows it might be a very important tool for treating diseases exactly in blueberry farms. It gives farmers and crop experts a quick, machine-driven way to find problems early. This lets them come up with special plans for dealing with these issues. The study's effects are not just about BMD. It shows how fancy technology can change the way of looking at diseases in farming too. This study builds a strong base for future improvements, aiming to improve the model's accuracy, size, and real-world use in making sure blueberry crops grow healthy.