Deep Learning-Based Rice Disease Classification Using EfficientNetB0

This study proposed a deep learning–based rice disease classification system using the EfficientNetB0 architecture combined with Explainable Artificial Intelligence (XAI) techniques. The proposed model successfully classified multiple rice diseases with an accuracy of more than 80%, demonstrating the effectiveness of transfer learning in agricultural image analysis. In addition to classification, the integration of the LIME explanation technique provides visual insights into the model’s decision-making process. This improves the transparency and interpretability of the system, making it more reliable for practical agricultural applications. Future research can focus on the following areas: • Expanding the dataset by including more real-world agricultural images • Deploying the system on mobile devices to assist farmers in real-time disease detection • Integrating Internet of Things (IoT)–based crop monitoring systems for continuous field analysis • Exploring other explainability techniques such as SHAP and Grad-CAM to further improve model interpretability Overall, the proposed approach demonstrates the potential of deep learning and explainable AI in improving automated plant disease detection and supporting smart agriculture systems.

1. "Using deep learning for image-based plant disease identification," Frontiers in Plant Science, vol. 7, 2016, Sharada P. Mohanty, David P. Hughes, and Marcel Salathé. 2. Konstantinos P. Ferentinos, "Deep learning models for plant disease diagnosis and detection," Computers and Electronics in Agriculture, vol. 145, pp. 311–318, 2018. 3. "EfficientNet: Rethinking model scaling for convolutional neural networks," by Marcus Tan and Quoc V. Le, Proceedings of the International Conference on Machine Learning, 2019. 4. Carlos Guestrin, Sameer Singh, and Marco Tulio Ribeiro, "Why should I believe you? "Explaining any classifier's predictions," in the 2016 ACM SIGKDD Conference on Knowledge Discovery and Data Mining Proceedings.   5. "Grad-CAM: Visual explanations from deep networks using gradient-based localization," Ramprasaath R. Selvaraju et al., IEEE International Conference on Computer Vision, 2017. 6. "Very deep convolutional networks for large-scale image recognition," Karen Simonyan and Andrew Zisserman, International Conference on Learning Representations, 2015. 7. "Deep residual learning for image recognition," Kaiming He et al., IEEE Conference on Computer Vision and Pattern Recognition, 2016. 8. "MobileNets: Efficient convolutional neural networks for mobile vision applications," Andrew G. Howard et al., 2017. 9. Too, E. C., Yujian, L., Njuki, S., and Yingchun, L., "A comparative study of fine-tuning deep learning models for plant disease identification," Computers and Electronics in Agriculture, 2019.   10. "Rice disease recognition using deep convolutional neural network," Computers and Electronics in Agriculture, 2018; Zhang, S., Huang, W., and Zhang, C.