Quantifying Cherry Leaf Spot Disease Severity: Using a mix of CNN and Gradient Boosting in Deep Learning Model

Measuring cherry leaf spot disease properly is very important for managing orchards well. This study is the first to use a mix of Convolutional Neural Networks (CNN) and Gradient Boosting (GB) algorithms. It does this to help computers automatically measure how severe diseases are in an improved way. A collection of 10,500 high-quality pictures showing different disease levels was carefully put together. The mixed model had an excellent total correctness of 98.9%. It proved that it could accurately place disease levels in their right groupings. Measures like accuracy, precision, recall and F1 score show if the model works well. These numbers prove that it can reduce mistakes while finding real disease cases properly. Comparing it to top methods shows that the mixed model is best for measuring disease seriousness. It sets new goals in farming with precision, making everything more accurate than before. This plan's everyday use goes beyond cherry gardens, and has hope for more general uses in plant sickness checking. Its job in making quick actions and helping people make smart choices helps stop money loss and keeps crop yield safe. This study creates a new way for machines to measure how severe diseases are. This will help farmers in the future and make food safer all around the world, making it more sustainable.