Brown Rust Disease Detection Using CNN-Random Forest Fusion

The research provides a thorough analysis of the CNN-Random Rainforest method for distinguishing wheat leaf diseases, categorizing them into Disease Intensity Level1 towards Disease Intensity Level5 groups. The model's extraordinary ability to properly identify wheat leaf illnesses and reduce incorrect predictions is highlighted by the fact that each class exhibits excellent precision, recall, and the F1-s all above 97%. The actual examples for each class can be found in the "support" column and range from 1030 to 1220, indicating a sizable dataset representation. The "support proportion" emphasizes a balanced dataset essential for an objective assessment. The model demonstrates accurate categorization with a 0.99 accuracy consistently observed across all classes. The "macro average" presents an overall performance picture with an F1-score, recall, and precision total of 97.39%. Given the makeup of the dataset, the "weighted average" retains this excellent performance at 97.40%. Reiterating this constancy, the "micro average" achieved an exceptional 97.40%. These reliable evaluation metrics confirm the practical applicability of the CNN-Random Forest technique in agricultural settings and demonstrate its effectiveness in detecting a range of wheat leaf diseases. The model excels in identifying wheat leaf diseases as seen by its excellent recall, precision, and F1-score, balanced representation, and noteworthy accuracy.