Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{195519,
        author = {Mrs.D.N.B.T.Sundari and Ch. Meghana and K. Abhisree Reddy and Sara Syed and G. Bindhu},
        title = {Interpretable Lightweight Model for Rice Leaf Disease Detection Using Knowledge Distillation},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {11},
        pages = {560-565},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=195519},
        abstract = {Rice is one of the most important food crops in the world, and rice production is greatly influenced by diseases like Bacterial Leaf Blight, Brown Spot, Blast, Sheath Blight, and Tungro. It is often difficult to detect diseases in remote areas due to the unavailability of experts. In this paper, a novel rice leaf disease detection system with minimal resources is suggested by using knowledge distillation. In this paper, a teacher model with a ResNet50 backbone is utilized to train a lightweight CNN student model with 91.8K parameters and a model size of 0.35 MB, which achieved an accuracy of 99.29%. The model is robust to brightness changes and has fewer training samples, ensuring the reliability of the model in real-world scenarios. Moreover, the model is enhanced by the addition of Grad-CAM, which helps to increase the transparency of the model.},
        keywords = {},
        month = {April},
        }