Predicting Diverse Diseases in Rice Leaves with Convolutional Neural Networks-VGG16

In the context of rice, a staple crop in India, the occurrence of diverse diseases throughout its growth stages poses a significant challenge for farmers, especially those with limited expertise in disease identification. To address this issue, automated systems employing convolutional neural network (CNN) models have emerged as promising solutions in contemporary deep learning research. Given the scarcity of comprehensive image datasets for rice leaf diseases, our study employs transfer learning on a modest dataset to construct an effective deep learning model. VGG-16 serves as the foundational architecture for training and evaluating our proposed CNN model, utilizing datasets sourced from both rice fields and the internet. Despite the challenges associated with dataset availability, our model achieves an impressive accuracy rate of 95%.