Hybrid CNN & Random Forest Model for Effective Mint Green Leaf Disease Diagnosis

This investigation focuses on the automated identification the illnesses in minty green leaves utilizing a machine learning technique, specifically the CNN or Random Forest algorithms. The study's goal is to improve the accuracy and effectiveness of the disease detection in mint leaves, hence improving the general well-being and production of mint plants. The classification model was trained and tested on the large dataset containing the variety of illnesses such as the Mint Rust, Powdery Mildew, Mint Anthracnose, Verticillium Wilt, Green Spot illnesses, Downy Mildew, and Bacterial Leaf Spot. The analysis of the model's performance for each illness class, evaluation criteria such as Precision, Recall, F1-Score, Support, or Accuracy were used the results show that particular diseases have good precision, and recall, with F1-Score values ranging from 92.94% through 95.65%. The measures ‘Macro Average,’ ‘Weighted Average,’ or ‘Macro Average’ provide the additional aggregated insights into the overall model performance, producing consistently high scores of 94.27% throughout precision, recall, or FI-Score. The measure ‘Support’ represents the model's robustness, exhibiting significant cases for each illness class. The results of this study demonstrate the efficacy of the suggested machine-learning method for correctly categorizing illnesses in the mint leaves. The model's strong precision and recall scores validate its capacity to identify the particular illnesses, which aids in the early detection and tailored therapy.