Diseases: CNN-SVM Integration

To distinguish between six different classes of illnesses associated with walnut leaves, this research evaluates the effectiveness of a classification model: Healthy, Thousand Cankers Disease, Anthracnose, Powdery Mildew, Phytophthora Root Rot, and Walnut Wilt. The model's performance indicators, which include recall, F1-score, accuracy, support, precision, and support, remained carefully examined to assess how well it could predict various outcomes. Across all classes, precision values, which indicate the percentage of accurate positive predictions among all projected positives, vary from 95.29% to 95.40%. This demonstrates how well the model can distinguish between healthy leaves and each illness group. Recall measures, which show how well the model can identify true positives, also nearly match accuracy; they fall between 95.29% and 95.40%. The model's remarkable total accuracy rating of 95.33% indicates that it can correctly predict each of the six classes together. Furthermore, the total, weighed, and micro averages confirm the model's reliable performance in a variety of assessment techniques, emphasizing its ability to differentiate between diseased and healthy walnut leaves. These results highlight the model's dependability in real-world scenarios and provide a useful instrument for disease control in walnut trees as well as agricultural settings. To sum up, the model performs well in accurately detecting a broad variety of illnesses that impact walnut leaves, hence providing significant potential for accurate disease detection in agricultural environments. These strong performance indicators highlight the model's importance as a useful instrument for supporting disease control procedures in the cultivation of walnut trees.