Skip to main content
Article | Green Intelligent Systems and Applications | Volume 6 - Issue 1 - 2026 | 120−132 | https://doi.org/10.53623/gisa.v6i1.1000
© 2026 by the author(s), and is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) License Creative Commons Attribution 4.0 International (CC BY 4.0) License

Lightweight Rice Leaf Disease Classification Using MobileNetV2: A Comprehensive Performance Evaluation

Author(s): Melinda Melinda 1 ORCID https://orcid.org/0000-0001-9082-6639 , Rahmat Maulana 1 , Yunidar yunidar 1 , Muhammad Irhamsyah 1 , Muhammad Saifullah Nur 1 , Nurlida Basir 2 , Elizar Elizar 1 , Muhammad Syafrudin 3 ORCID https://orcid.org/0000-0002-5640-4413
Author(s) information:
1 Department of Electrical Engineering, Faculty of Engineering, Universitas Syiah Kuala, Banda Aceh, 23111, Indonesia
2 Universiti Sains Islam Malaysia (USIM), Malaysia
3 Department of Artificial Intelligence and Data Science, Sejong University, Seoul 05006, Republic of Korea

Corresponding author

Green Intelligent Systems and Applications

Volume 6 - Issue 1 - 2026

 About the journal
Submit your article

Rice leaf diseases pose a significant threat to agricultural productivity, and accurate automated detection is essential for timely intervention. This study presents a comparative evaluation of lightweight convolutional neural network architectures for the classification of six rice leaf disease categories: Bacterial Leaf Blight, Brown Spot, Leaf Blast, Leaf Scald, Narrow Brown Leaf Spot, and Rice Healthy. MobileNetV2 is proposed as the primary model and benchmarked against EfficientNetB0 and NASNetMobile. All three architectures were trained under an identical experimental setup comprising a two-stage transfer learning strategy, a unified custom classification head consisting of Global Average Pooling, Batch Normalization, two dense layers with dropout and L2 regularization, and a Softmax output layer. The dataset comprised 1,920 images across six classes obtained from Roboflow Universe, with no pre-augmentation applied by the original source. Training-time augmentation including rotation, shifting, shearing, zooming, and horizontal flipping was applied exclusively to the training subset. Experiments were conducted on a stratified split of 1,536 training, 192 validation, and 192 test images with a fixed random seed of 42 to ensure reproducibility. MobileNetV2 achieved the highest test accuracy of 96.35% and macro F1-score of 96.35%, outperforming EfficientNetB0 at 94.27% and NASNetMobile at 89.06%. In terms of computational efficiency, MobileNetV2 also demonstrated the most favorable deployment profile with a TensorFlow Lite model size of 2.75 MB and inference latency of 3.22 ms per image, indicating potential suitability for resource-constrained deployment scenarios. These results suggest that MobileNetV2 offers a competitive balance between classification accuracy and computational efficiency for rice leaf disease identification.

Read Full-Text
Previous article
Next article

Kim, Y.S.; Choi, S.H.; Park, J.H. (2025). Changes in the occurrence patterns of rice fungal diseases due to climate change. Research in Plant Disease, 31, 17‒33. https://doi.org/10.5423/RPD.2025.31.1.17.

Shen, M.; Cai, C.; Song, L.; Qiu, J.; Ma, C.; Wang, D.; et al. (2023). Elevated CO₂ and temperature under future climate change increase severity of rice sheath blight. Frontiers in Plant Science, 14, 1115614. https://doi.org/10.3389/fpls.2023.1115614.

Singh, P.K.; Ray, S.; Thakur, S.; Rathour, R.; Sharma, V.; Sharma, T.R. (2018). Co-evolutionary interactions between host resistance and pathogen avirulence genes in rice-Magnaporthe oryzae pathosystem. Fungal Genetics and Biology, 115, 9‒19. https://doi.org/10.1016/j.fgb.2018.04.005.

Kusuma, B.; Hermanto, T.I.; Lestari, C.D. (2025). Classification of disease types in rice plants using convolutional neural network algorithm. Jurnal Informatika dan Komputer, 9, 40. https://doi.org/10.26798/jiko.v9i1.1395.

Akter, S.; Sumon, R.I.; Ali, H.; Kim, H.C. (2024). Utilizing convolutional neural networks for the effective classification of rice leaf diseases through a deep learning approach. Electronics, 13, 4095. https://doi.org/10.3390/electronics13204095.

Ayyappan, A.B.; Gobinath, T.; Kumar, M.; Sivaramakrishnan, A. (2025). Rice plant disease detection using convolutional neural networks. Discover Artificial Intelligence, 5, 277. https://doi.org/10.1007/S44163-025-00277-X.

Dutta, M.; Saikia, M.J.; Bora, K.; Barman, D.; Sarma, K.K.; Paul, R.; et al. (2024). Rice leaf disease classification: a comparative approach using CNN, CAAR-U-Net, and MobileNetV2. Technologies, 12, 214. https://doi.org/10.3390/technologies12110214.

Shankar, S.J.; Palanivel, S.; Venkateswarlu, S.C. (2025). Automating paddy crop disease classification with deep learning models. Journal of Computer Science, 21, 1772‒1784. https://doi.org/10.3844/jcssp.2025.1772.1784.

Putra, O.V.; Mustaqim, M.Z.; Muriatmoko, D. (2023). Transfer learning for classification of rice diseases and pests using MobileNetV2. Techno.Com, 22, 562‒575. https://doi.org/10.33633/tc.v22i3.8516.

Sandler, M.; Howard, A.; Zhu, M.; Zhmoginov, A.; Chen, L.C. (2018). MobileNetV2: inverted residuals and linear bottlenecks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 4510‒4520. https://doi.org/10.1109/CVPR.2018.00474.

Tan, M.; Le, Q.V. (2019). EfficientNet: rethinking model scaling for convolutional neural networks. Proceedings of the International Conference on Machine Learning, 97, 6105‒6114.

Zoph, B.; Vasudevan, V.; Shlens, J.; Le, Q.V. (2018). Learning transferable architectures for scalable image recognition. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 8697‒8710. https://doi.org/10.1109/CVPR.2018.00907.

Zhang, Z.; Gu, Y.; Hong, Q. (2021). Rice disease identification system using lightweight MobileNetV2. Proceedings of the International Conference on Image, Signal and Information Processing, 29‒36. https://doi.org/10.12792/icisip2021.007.

Tiwari, R.; Vora, N. (2024). Enhancing paddy leaf disease classification using CNN and MobileNetV2. Journal of Soft Computing Paradigm, 6, 324‒340. https://doi.org/10.36548/jscp.2024.3.008.

Pai, P.; Amutha, S.; Basthikodi, M.; Pai, B.H.; Pai, R.M.; Nayak, S.R. (2025). A twin CNN-based framework for optimized rice leaf disease classification with feature fusion. Journal of Big Data, 12, 89. https://doi.org/10.1186/s40537-025-01148-z.

Xu, Y.; Li, D.; Li, C.; Yuan, Z.; Dai, Z. (2025). LiSA-MobileNetV2: an extremely lightweight deep learning model for accurate rice disease classification. Frontiers in Plant Science, 16, 1‒14. https://doi.org/10.3389/fpls.2025.1619365.

Munda, K.K.; Patil, N. (2025). A deep learning framework for plant disease detection. Lecture Notes in Networks and Systems, 1239, 353‒366. https://doi.org/10.1007/978-981-96-1188-1_26.

Tanwar, V.; Lamba, S.; Sharma, B. (2023). Deep learning-based hybrid model for severity prediction of leaf smut sugarcane infection. Proceedings of the 2023 International Conference on Artificial Intelligence and Smart Energy, 1004‒1009. https://doi.org/10.1109/ICAIS56108.2023.10073880.

Jayasooriya, G.R.I.L.; Arachchi, S.M. (2022). Diagnosis of bacterial leaf blight, brown spots, and leaf smut rice plant diseases using Light GBM. International Journal of Computer Applications, 183, 53‒58. https://doi.org/10.5120/ijca2022921895.

Pal, C.; Karmakar, S.; Mukherjee, I.; Chakrabarti, P.P. (2025). A lightweight and explainable CNN model for empowering plant disease diagnosis. Scientific Reports, 15, 30720. https://doi.org/10.1038/s41598-025-94083-1.

Lu, J.; Liu, X.; Ma, X.; Tong, J.; Peng, J. (2023). Improved MobileNetV2 crop disease identification model for intelligent agriculture. PeerJ Computer Science, 9, e1595. https://doi.org/10.7717/peerj-cs.1595.

Rice Leaf - v2 2023-06-19. Roboflow Universe. (accessed on 12 May 2026) Available online: https://universe.roboflow.com/utopialab/rice-leaf-wfax3/dataset/2.

Hartono, A.C.; Muslikh, A.R. (2025). The application of MobileNetV2 transfer learning for image classification of fruit types. Journal of Information Systems and Application Development, 3, 102‒110. https://doi.org/10.26905/jisad.v3i2.16187

Pranav Dhole, D.S.; Nandwanshi, I. (2025). Crop disease classification for edge devices: a quantized MobileNetV2 approach. International Journal of Innovative Research in Electrical, Electronics, Instrumentation and Control Engineering, 13, 39‒45. https://doi.org/10.17148/ijireeice.2025.131006.

Read Full-Text
About this article

SUBMITTED: 19 January 2026
ACCEPTED: 12 May 2026
PUBLISHED: 4 June 2026
SUBMITTED to ACCEPTED: 113 days
DOI: https://doi.org/10.53623/gisa.v6i1.1000

Cite this article
Melinda, M., Maulana, R., yunidar, Y., Irhamsyah, M., Saifullah Nur, M., Basir, N., Elizar, E., & Syafrudin, M. . (2026). Lightweight Rice Leaf Disease Classification Using MobileNetV2: A Comprehensive Performance Evaluation. Green Intelligent Systems and Applications, 6(1), 120−132. https://doi.org/10.53623/gisa.v6i1.1000
Citations
0
Share this article