Enhanced Multiclassification of Avocado Leaf Diseases: CNN and Random Forest Integration

Through the use of advanced machine learning techniques, the primary objective of this research is to develop and evaluate a classification model to identify a variety of illnesses that may affect avocado leaves. Despite exhibiting remarkable accuracy, recall, and F1-score metrics, the model displays good performance across a wide range of illness categories. The accuracy levels of the model, which range from 92.61% to 95.32%, are indicative of the effectiveness of the model in reliably detecting different kinds of diseases. Furthermore, recall rates, ranging between 91.53% and 94.44%, demonstrate the model's efficacy in collecting important cases within each illness categorization. The model demonstrates resilience in the classification of avocado leaf diseases, as it maintains an overall accuracy of roughly 93.66% during its operation. These results suggest that the model has the potential to facilitate early disease diagnosis and accurate identification, which would give major contributions to the enhancement of disease control systems in avocado farming.