Integrating CNN and Random Forest for Accurate Classification of Mango Leaf Diseases

This study introduces a newly developed deep-learning approach to classify and identify different diseases that affect mango leaves. This method improves the efficiency of a convolutional neural network. This method uses a random forest to classify different classes, pooling layers, and additional components that contribute to the model's functionality. This paper utilized a total of 3633 images. In this paper, the images were obtained from both primary and secondary resources during the calculation process. Primary resources are calculated from the Punjab area. To evaluate the model's performance, various metrics are used such as precision, recall, F1-score, support, accuracy, and average. To check the effectiveness of the model use these metrics, and it was found to have an average weighted F1-score of 80.85% and an overall accuracy of 80.89%. The model achieved the highest precision score of 90.28% and the lowest precision score of 74.16%. The model achieved the highest recall of 94.33% and the lowest recall of 74.33%. These findings highlight the importance of early disease detection and effective management practices, such as cultural methods and chemical treatments. These findings are beneficial for farmers, helping them save valuable time.