A Deep Learning Framework for Paddy and Maize Leaf Disease Detection

This study presents an innovative deep learning model designed for detecting diseases in paddy and maize leaves by integrating Convolutional Neural Networks (CNNs) and Residual Networks (ResNet). This hybrid approach enhances the accuracy of disease identification in these crucial crops. The model is trained using an extensive dataset from the Plant Village database, which includes high-resolution images of leaves affected by diseases like blast, hispa, and bacterial leaf streak. Rigorous preprocessing steps, such as normalization, resizing, and augmentation, are applied to the images. Achieving an impressive accuracy of 97.0%, the model surpasses traditional machine learning and current deep learning techniques in terms of precision, recall, and F1-score. This research highlights the model’s effectiveness in accurately detecting diseases, leveraging a specialized dataset for this purpose. The findings suggest significant potential for improving agricultural productivity and effective disease management. Future research will aim to expand the model’s application to a broader range of crops and incorporate it with IoT systems for real-time monitoring, enhancing its utility in agricultural practices, and optimizing crop yields.