Accurate Anthracnose Disease Classification using a Hybrid CNN-Random Forest Architecture

Using a strong classification model, this research work examines mango powdered leaf disease evaluation and mitigation. The research findings should summarised in the table below, which also shows how well the model performed in each of the ten categories of disease severity. Precision measures, which range from 94.70% to an astonishing 98.04%, highlight the model's accuracy in forecasting intensity levels. These metrics demonstrate its capacity to minimize false positives and match expected positives to real occurrences. Recall metrics, with recall percentages ranging from 95.43% to 98.28%, show the model's capacity to recognize true positive cases across intensity categories. The stronger recall is demonstrated at higher intensity levels, demonstrating the model's capacity to recognize complicated cases. The model's ability to strike a balance between precise predictions and thorough coverage is shown by the F1-Score, which combines precision and recall. The performance ranges from 95.06% and 98.16% and is steady. While support proportions provide a relative distribution, support values give precise occurrence counts. These revelations shed light on the class distribution and confirm metric importance over the entire dataset. Different stages of the mango powdered leaf disease have been accurately identified by the model. The thorough analysis, as shown in the table, provides information on how to fight sickness. This study contributes to data-driven disease detection methods and plant pathology.