CornNet: Implementation of Transfer Learning for Disease Classification in Corn Plants Based on Web Application

This study addresses the classification of corn leaf diseases caused by infections such as Blight, Common Rust, and Grey Leaf Spot, which significantly affect corn production. Early and accurate classification is crucial for effective disease management and yield improvement. To solve this problem, this research implements Transfer Learning and Explainable AI (XAI) to classify corn leaf disease images and integrates the solution into a web-based system. The contribution of this research lies in combining modern deep learning models with XAI to enhance transparency in plant disease classification systems, specifically through the integration of these models into a web-based platform and a comprehensive performance comparison across different optimizers to evaluate robustness and efficiency. Five pre-trained deep learning architectures—ResNet101, VGG16, EfficientNetB1, DenseNet201, and InceptionV3—are utilized as Transfer Learning models. Grad-CAM (Gradient-weighted Class Activation Mapping) is used to visualize the most influential regions in disease image classification. The dataset used is “Corn or Maize Leaf Disease” containing 4,188 images across four classes: Blight, Common Rust, Grey Leaf Spot, and Healthy. The results demonstrate that Transfer Learning and Explainable AI can be effectively applied to corn leaf disease classification and web deployment. Among the models, EfficientNetB1 achieved the highest accuracy of 95%, along with clear Grad-CAM visualizations that enhance interpretability. This study contributes to the development of intelligent agricultural systems and supports decision-making in crop disease management using transparent AI solutions.