An Explainable Rice Leaf Disease Recognition Using ResNet50 Transfer Learning and SHAP-Based Top-K Feature Selection
DOI:
https://doi.org/10.24113/mhz86w18Keywords:
: Rice leaf disease detection, Transfer learning, ResNet50, SHAP (Shapley Additive Explanations), Feature selection,Abstract
Rice is a major staple crop, but its yield and grain quality are often reduced by leaf diseases. In many farming regions, disease identification is still done by eye, which can be slow, inconsistent, and difficult when symptoms change with lighting, plant age, or background conditions. To support early and reliable diagnosis, this paper proposes a rice leaf disease classification approach that combines a strong deep-learning model with an explainability-based feature selection method. We fine-tune a pretrained ResNet50 model to learn disease-related visual patterns from leaf images resized to 240×240. The network produces a compact 512-dimensional embedding that summarizes the most important disease characteristics. To make the system more interpretable and to remove less useful features, we train a Random Forest model on these embeddings and use SHapley Additive exPlanations (SHAP) to measure the importance of each embedding dimension. Using a per-class union Top-K strategy that selects the most informative features while ensuring that minority classes are not ignored. A lightweight classifier is then trained on the selected feature subset using a two-stage training process. Experiments show that the proposed SHAP-guided ResNet50 model achieves 98% accuracy. Compared to an existing approach, the proposed method improves accuracy with less training time per epoch.
Downloads
References
[1] S. M. Hassan and A. K. Maji, “Plant Disease Identification Using a Novel Convolutional Neural Network,” IEEE Access, vol. 10, pp. 5390–5401, 2022, doi: 10.1109/ACCESS.2022.3141371.
[2] G. Latif, S. E. Abdelhamid, R. E. Mallouhy, J. Alghazo, and Z. A. Kazimi, “Deep Learning Utilization in Agriculture: Detection of Rice Plant Diseases Using an Improved CNN Model,” Plants, vol. 11, no. 17, 2022, doi: 10.3390/plants11172230.
[3] R. Dogra, S. Rani, A. Singh, M. A. Albahar, A. E. Barrera, and A. Alkhayyat, “Deep learning model for detection of brown spot rice leaf disease with smart agriculture,” Comput. Electr. Eng., vol. 109, p. 108659, 2023, doi: https://doi.org/10.1016/j.compeleceng.2023.108659.
[4] N. T. Su, P. D. Hung, B. T. Vinh, and V. T. Diep, “Rice Leaf Disease Classification Using Deep Learning and Target for Mobile Devices,” in Proceedings of International Conference on Emerging Technologies and Intelligent Systems, 2022, pp. 136–148.
[5] S. Jain et al., “Automatic Rice Disease Detection and Assistance Framework Using Deep Learning and a Chatbot,” Electronics, vol. 11, no. 14, 2022, doi: 10.3390/electronics11142110.
[6] Muhammad Juman Jhatial et al., “Deep Learning-Based Rice Leaf Diseases Detection Using Yolov5,” Sukkur IBA J. Comput. Math. Sci., vol. 6, no. 1, pp. 49–61, 2022, doi: 10.30537/sjcms.v6i1.1009.
[7] V. Rajpoot, A. Tiwari, and A. S. Jalal, “Automatic early detection of rice leaf diseases using hybrid deep learning and machine learning methods,” Multimed. Tools Appl., vol. 82, no. 23, pp. 36091–36117, 2023, doi: 10.1007/s11042-023-14969-y.
[8] V. Vinoth Kumar, K. M. Karthick Raghunath, N. Rajesh, M. Venkatesan, R. B. Joseph, and N. Thillaiarasu, “Paddy Plant Disease Recognition, Risk Analysis, and Classification Using Deep Convolution Neuro-Fuzzy Network,” J. Mob. Multimed., vol. 18, no. 2, pp. 325–348, 2022, doi: 10.13052/JMM1550-4646.1829.
[9] M. S. H. Shovon, S. J. Mozumder, O. K. Pal, M. F. Mridha, N. Asai, and J. Shin, “PlantDet: A Robust Multi-Model Ensemble Method Based on Deep Learning For Plant Disease Detection,” IEEE Access, vol. 11, pp. 34846–34859, 2023, doi: 10.1109/ACCESS.2023.3264835.
[10] E. Moupojou et al., “FieldPlant: A Dataset of Field Plant Images for Plant Disease Detection and Classification With Deep Learning,” IEEE Access, vol. 11, pp. 35398–35410, 2023, doi: 10.1109/ACCESS.2023.3263042.
[11] C. Sherin Shibi, R. Babitha Lincy, J. Jency Rubia, and P. T. Sheeba, “Deep learning based Rice plant Disease Identification with Transfer learning,” in 2022 International Conference on Computing, Communication, and Intelligent Systems (ICCCIS), 2022, pp. 697–702. doi: 10.1109/ICCCIS56430.2022.10037638.
[12] Haridasan, J. Thomas, and E. D. Raj, “Deep learning system for paddy plant disease detection and classification,” Environ. Monit. Assess., vol. 195, no. 1, p. 120, 2022, doi: 10.1007/s10661-022-10656-x.
[13] S. P. Singh, K. Pritamdas, K. J. Devi, and S. D. Devi, “Custom Convolutional Neural Network for Detection and Classification of Rice Plant Diseases,” Procedia Comput. Sci., vol. 218, pp. 2026–2040, 2023, doi: https://doi.org/10.1016/j.procs.2023.01.179.
[14] H. Liu, Y. Cui, J. Wang, and H. Yu, “Analysis and Research on Rice Disease Identification Method Based on Deep Learning,” Sustainability, vol. 15, no. 12, 2023, doi: 10.3390/su15129321.
[15] J. A. Wani, S. Sharma, M. Muzamil, S. Ahmed, S. Sharma, and S. Singh, “Machine Learning and Deep Learning Based Computational Techniques in Automatic Agricultural Diseases Detection: Methodologies, Applications, and Challenges,” Arch. Comput. Methods Eng., vol. 29, no. 1, pp. 641–677, 2022, doi: 10.1007/s11831-021-09588-5.
[16] C. R. Rahman et al., “Identification and recognition of rice diseases and pests using convolutional neural networks,” Biosyst. Eng., vol. 194, pp. 112–120, Jun. 2020, doi: 10.1016/J.BIOSYSTEMSENG.2020.03.020.
[17] P. K. Sethy, N. K. Barpanda, A. K. Rath, and S. K. Behera, “Deep feature based rice leaf disease identification using support vector machine,” Comput. Electron. Agric., vol. 175, p. 105527, Aug. 2020, doi: 10.1016/J.COMPAG.2020.105527.
[18] https://www.kaggle.com/datasets/minhhuy2810/rice-diseases-image-dataset/code
[19] Ahad, Md Taimur, et al. "Comparison of CNN-based deep learning architectures for rice diseases classification." Artificial Intelligence in Agriculture 9 (2023): 22-35
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Prabhat Kumar Suman, Rakesh Shivhare
This work is licensed under a Creative Commons Attribution 4.0 International License.
IJOSCIENCE follows an Open Journal Access policy. Authors retain the copyright of the original work and grant the rights of publication to the publisher with the work simultaneously licensed under a Creative Commons CC BY License that allows others to distribute, remix, adapt, and build upon your work, even commercially, as long as they credit you for the original creation. Authors are permitted to post their work in institutional repositories, social media or other platforms.
Under the following terms:
-
Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.
- No additional restrictions — You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.
