Enhanced Corn Leaf Disease Classification Using DenseNet121 and Hybrid Machine Learning Models

The advent of recent advancements in computer vision and deep learning has led to a notable enhancement in agricultural practices, particularly in the domains of disease detection and classification. This study focuses on developing an automated system for corn leaf disease classification using a hybrid approach. We employed DenseNet121 as a feature extractor, yielding 1024 feature vectors per image. These features were used to train and tune four classifiers: The Boosted Decision Trees (DT), Random Forest (RF), Logistic Regression (LR), and Support Vector Machine (SVM) models were evaluated. The Boosted DT model achieved the highest overall accuracy of 96.18%, precision of 96.38%, and recall of 93.55%, outperforming the other models including the end-to-end DenseNet121. The results indicate the potential of hybrid deep learning models for improving the accuracy and efficiency of corn leaf disease classification.