Rice Disease Detection

This study presented a deep learning-based system for rice disease detection designed to address the challenges associated with traditional manual inspection methods. The primary contribution of this work lies in combining a Convolutional Neural Network (CNN) for accurate disease classification with Grad-CAM for visual interpretability. This integrated approach not only improves detection performance but also enhances the transparency of the model’s decision- making process. The results obtained from experimental evaluation indicate that the proposed system is capable of accurately identifying various rice leaf diseases while maintaining consistent performance across different test samples. The inclusion of Grad-CAM visualizations allows users to observe the specific regions of the leaf that influenced the prediction, making the system more reliable and easier to understand. This is particularly important in agricultural applications, where users often require clear justification before taking corrective actions. Furthermore, the system demonstrates the practical benefit of combining prediction outputs with visual explanations. Instead of relying solely on numerical results, users are provided with both classification labels and heatmap-based insights, which support better decision-making in real-world scenarios. This dual capability reduces dependency on expert knowledge and enables faster identification of plant diseases. However, the effectiveness of the system depends on the quality and diversity of the training dataset. In cases where certain diseases are underrepresented, the model may show reduced accuracy. Additionally, environmental factors such as lighting conditions and image quality can influence prediction results, indicating the need for controlled input conditions. Overall, the proposed approach highlights the importance of integrating explainable AI techniques with deep learning models in agricultural systems. By ensuring both accuracy and interpretability, the system provides a practical and scalable solution for early detection of rice diseases, contributing to improved crop management and reduced agricultural losses.

  [1]       S. Mohanty, D. Hughes, and M. Salathé, “Using deep learning for image-based plant disease detection,” Frontiers in Plant Science, 2016. [2]       R. Ferentinos, “Deep learning models for plant disease detection and diagnosis,” Computers and Electronics in Agriculture, 2018.I. S. Jacobs and C. P. Bean, “Fine particles, thin films and exchange anisotropy,” in Magnetism, vol. III, G. T. Rado and H. Suhl, Eds. New York: Academic, 1963, pp. 271–350. [3]       K. P. Ferentinos, “Deep learning models for plant disease classification,” Agricultural Systems, 2018.R. Nicole, “Title of paper with only first word capitalized,” J. Name Stand. Abbrev., in press. [4]       R. Selvaraju et al., “Grad-CAM: Visual explanations from deep networks via gradient-based localization,” Proc. IEEE International Conference on Computer Vision (ICCV), 2017.M. Young, The Technical Writer’s Handbook. Mill Valley, CA: University Science, 1989. [5]        K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016. [6]        K. Simonyan and A. Zisserman, “Very deep convolutional networks for large-scale image recognition,” arXiv preprint arXiv:1409.1556, 2014.. [7]       M. Tan and Q. Le, “EfficientNet: Rethinking model scaling for convolutional neural networks,” Proc. International Conference on Machine Learning (ICML), 2019. [8]       A. Kamilaris and F. X. Prenafeta-Boldú, “Deep learning in agriculture: A survey,” Computers and Electronics in Agriculture, 2018. [9]       J. Too, S. Yujian, N. Njuki, and L. Yingchun, “A comparative study of fine-tuned deep learning models for plant disease identification,” Computers and Electronics in Agriculture, 2019. [10]    H. Hughes and M. Salathé, “An open access repository of plant disease images to enable the development of mobile disease diagnostics,” arXiv preprint arXiv:1511.08060, 2015.