Plum Pox Virus Impact Assessment Unveiling Through Convolutional Neural Network and Random Forest Model

Plum pox virus (PPV) is the greatest threat among all viral diseases associated with plant cultivation, especially since the majority of stone fruit crops become victims of this virus. The research seeks to evaluate the severity of PPV by deploying a hybrid deep learning approach that combines Convolutional Neural Networks (CNNs) and Random Forests. This dataset contains 9000 images of varying PPV severity, and the model was trained to appropriately classify them into seven major categories. This research provides a novel approach to distinguishing the levels of PPV severities by using an integrated deep-learning model with an overall accuracy of 96.84%. Performance metrics such as precision, recall, and F1-score are used to validate the model robustness that enables accurate identification and categorization of PPV severities at all levels. Some specificity severities demonstrate minor differences in the confusion matrix, which suggests potential model improvements. The hybrid model described exemplifies unparalleled skills in accurate measuring of PPV severity, allowing farmers a precise tool for tailored procedures. This study’s results indicate a major step towards addressing the effect of PPV and formulating management strategies in fruit growing that are focused on dealing with it. Future modifications of the mentioned model are promising in terms of providing improved assessment regarding the severity of PPV, thus strengthening resilience to fruit cultivation caused by this widespread viral danger.