Cross-Domain Adaptation Techniques for Robust Plant Disease Detection: A DANN-CORAL Hybrid Approach

Pulicherla Siva Prasad; Senthilrajan Agniraj

doi:10.52756/ijerr.2024.v42.007

Authors

Pulicherla Siva Prasad Department of Computer Science, Alagappa University, Karaikudi 630003, India https://orcid.org/0000-0002-7424-2200
Senthilrajan Agniraj Department of Computational Logistics, Alagappa University, Karaikudi 630003, India https://orcid.org/0000-0003-4265-7097

DOI:

https://doi.org/10.52756/ijerr.2024.v42.007

Keywords:

Plant Disease Detection, Cross-Domain Adaptation, Domain-Adversarial Neural Network, Correlation Alignment and Deep Learning

Abstract

Plant disease detection with deep learning models has shown promising results, but these models often struggle with generalizing across diverse agricultural environments due to domain shifts in imaging conditions. This paper presents a novel hybrid approach focusing on cross-domain adaptation techniques to address the challenge of domain shift. Our proposed method combines the Domain-Adversarial Neural Network (DANN) with Correlation Alignment (CORAL) to mitigate domain shifts between datasets. The DANN framework enforces domain-invariant feature learning through adversarial training. Using the PlantVillage Dataset, with controlled environment images, and the New Plant Village Dataset, with varied conditions, the model is first trained on PlantVillage and then adapted to New Plant Village using the CORAL loss to support the second-order statistics. In case of domain shift experiments with various datasets, DANN-CORAL achieved accuracies 91.39%, precision 93.36%, recall 88.9% and F1-scores 91.05% indicating the robustness and generalizability of our model is better than the other baseline models. This approach enhances model robustness and adaptability, providing insights into combining adversarial and statistical alignment for cross-domain adaptation in agricultural imaging.

References

Chen, C., Weiping, X., Yi, W., Yue, H., & Xinghao, D. (2020). Multiple-source domain adaptation with generative adversarial nets. Knowledge-Based Systems, 199, 105962. https://doi.org/10.1016/j.knosys.2020.105962

Cheng, Z., Chen, C., & Chen, Z. (2021). Robust and high-order correlation alignment for unsupervised domain adaptation. Neural Comput & Applic., 33, 6891–6903. https://doi.org/10.1007/s00521-020-05465-7.

Chlap, P., Hang, M., Nym, V., Jason, D., Lois, H., & Annette, H. (2021). A review of medical image data augmentation techniques for deep learning applications. Journal of Medical Imaging and Radiation Oncology, 65(5), 545-563. https://doi.org/10.1111/1754-9485.13261.

Chulif, S., Lee, S.H., & Chang, Y.L. (2023). A machine learning approach for cross-domain plant identification using herbarium specimens. Neural Comput & Applic., 35, 5963–5985.

https://doi.org/10.1007/s00521-022-07951-6.

Farahani, A., Sahar, V., Khaled, R., & Hamid, R. A. (2020). A brief review of domain adaptation. Advances in data science and information engineering: proceedings from ICDATA 2020 and IKE 2020, 2021, 877-894.

https://doi.org/10.48550/arXiv.2010.03978.

Hongsong, W., Liao, S., & Shao, L. (2021). Afan: Augmented feature alignment network for cross-domain object detection. IEEE Transactions on Image Processing, 30, 4046-4056.

https://doi.org/10.48550/arXiv.2106.05499

Hsu, H.K., Chun-Han, Y., Yi-Hsuan, T., Wei-Chih, H., Hung-Yu, T., Maneesh, S., & Ming-Hsuan, Y. (2020). Progressive domain adaptation for object detection. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pp. 749-757. https://doi.org/10.48550/arXiv.1910.11319

Huang, J., Chen, K., Ren, Y., Sun, J., Wang, Y., Tao, T., & Pu, X. (2023). CDDnet: Cross-domain denoising network for low-dose CT image via local and global information alignment. Comput Biol Med., 163, 107219. https://doi.org/10.1016/j.compbiomed.2023.107219.

Hughes, D. P., & Salathe, M. (2015). An open access repository of images on plant health to enable the development of mobile disease diagnostics. arXiv preprint. https://doi.org/10.48550/arXiv.1511.08060.

Ida, A., Niels, C. O., Agnieszka, K. (2020). Domain-adversarial neural network for improved generalization performance of Gleason grade classification. Proc. SPIE 11320, Medical Imaging 2020: Digital Pathology, 1132016. https://doi.org/10.1117/12.2549011.

Laura, F., Lorenzo, M., Denis, D. L., Danilo, P., Valeria, T., & Claudio, T. (2022). A CNN-based image detector for plant leaf diseases classification. HardwareX, 12, e00363. https://doi.org/10.1016/j.ohx.2022.e00363.

Li, L., Zhang, S., & Wang, B. (2021). Plant disease detection and classification by deep learning—a review. IEEE Access, 9, 56683-56698. https://doi.org/10.1109/ACCESS.2021.3069646.

Mahmud, M., Kaiser, M.S., McGinnity, T.M. (2021). Deep Learning in Mining Biological Data. Cogn Comput., 13, 1–33. https://doi.org/10.1007/s12559-020-09773-x.

Manh-Ha, B., Tran, T., Tran, A., & Phung, D. (2021). Exploiting domain-specific features to enhance domain generalization. Advances in Neural Information Processing Systems, 34, 21189-21201, https://doi.org/10.48550/arXiv.2110.09410.

Marvin, Z., Marklund, H., Dhawan, N., Gupta, A., Levine, S., & Finn, C. (2021). Adaptive Risk Minimization: Learning to Adapt to Domain Shift. NeurIPS, arXiv, 2007.02931. https://doi.org/10.48550/arXiv.2007.02931

Minghao, C., Zhao, S., Liu, H., & Cai, D. (2020). Adversarial-learned loss for domain adaptation. In Proceedings of the AAAI Conference on Artificial Intelligenc, 34(04), 3521-3528. 2020.

https://doi.org/10.48550/arXiv.2001.01046.

Nam, M.G., & Suh-Yeon, D. (2023). Classification of companion animals’ ocular diseases: Domain adversarial learning for imbalanced data. IEEE Access. https://doi.org/10.1109/ACCESS.2023.3344579.

Preciado-Grijalva, A., & Venkata Santosh, S.R.M. (2021). Evaluation of Deep Neural Network Domain Adaptation Techniques for Image Recognition. arXiv preprint arXiv, 2021, 2109.13420. https://doi.org/10.48550/arXiv.2109.13420.

Samir-Bhattarai (2019). New Plant Disease Dataset. Available at https://www.kaggle.com/datasets/vipoooool/new-plant-diseases-dataset.

Singh, V., Sharma, N., & Singh, S. (2020). A review of imaging techniques for plant disease detection. Artificial Intelligence in Agriculture, 4, pp. 229-242. https://doi.org/10.1016/j.aiia.2020.10.002.

Soto, P.J., Costa, G.A., Feitosa, R.Q., Ortega, M.X., Bermudez, J.D., & Turnes, J. (2022). Domain-Adversarial Neural Networks for Deforestation Detection in Tropical Forests. In IEEE Geoscience and Remote Sensing Letters, 19, 2504505. https://doi.org/10.1109/LGRS.2022.3163575

Storey, G., Qinggang, M., & Baihua, Li. (2022). Leaf disease segmentation and detection in apple orchards for precise smart spraying in sustainable agriculture. Sustainability, 14(3), 1458.

Talukder, M. S. H., Chowdhury, M. R., Sourav, M. S. U., Rakin, A. A., Shuvo, S. A., Sulaiman, R. B., Nipun, M. S., Islam, M., Islam, M. R., Islam, M. A., & Haque, Z. (2023). JutePestDetect: An intelligent approach for jute pest identification using fine-tuned transfer learning. Smart Agricultural Technology, 5, 100279. https://doi.org/10.1016/j.atech.2023.100279

Vijayalakshmi, D., Nath, M.K., & Acharya, O.P. (2020). A Comprehensive Survey on Image Contrast Enhancement Techniques in Spatial Domain. Sens Imaging., 21, 40. https://doi.org/10.1007/s11220-020-00305-3.

Wang, H., Cheng, Y., Chen, C.L.P., & Wang, X. (2022). Hyperspectral Image Classification Based on Domain Adversarial Broad Adaptation Network. In IEEE Transactions on Geoscience and Remote Sensing, 60, 5517813. https://doi.org/10.1109/TGRS.2021.3128162

Wang, W., Haojie, L., Zhengming, D., Feiping, N., Junyang, C., Xiao, D., & Zhihui, W. (2021). Rethinking maximum mean discrepancy for visual domain adaptation. IEEE Transactions on Neural Networks and Learning Systems, 34(1), 264-277. https://doi.org/10.1109/TNNLS.2021.3093468.

Wang, Y.Y., Gu, J.M., & Wang, C. (2020). Discrimination-Aware Domain Adversarial Neural Network. J. Comput. Sci. Technol., 35, 259–267. https://doi.org/10.1007/s11390-020-9969-4.

Wang, Z.Y., & Dae-Ki, K. (2021). P-norm attention deep coral: Extending correlation alignment using attention and the p-norm loss function. Applied Sciences, 11(11), 5267. https://doi.org/10.3390/app11115267.

Wei, W. (2023). Tea Leaf Disease Classification using Domain Adaptation Method." Frontiers in Computing and Intelligent Systems, 3(2), 48-50. https://doi.org/10.54097/fcis.v3i2.7187.

Wu, X., Fan, X., Luo, P., Choudhury, S.D., Tjahjadi, T., & Hu, C. (2023). From Laboratory to Field: Unsupervised Domain Adaptation for Plant Disease Recognition in the Wild. Plant Phenomics, 5, 0038. https://doi.org/10.34133/plantphenomics.0038.

Xiaoxu, L., Yang, X., Ma, Z., & Jing-Hao, X. (2023). Deep metric learning for few-shot image classification: A review of recent developments. Pattern Recognition, 138, 109381. https://doi.org/10.48550/arXiv.2105.08149.

Zhang, Y., Hu, D., Zhao, Q., Quan, G., Liu, J., Liu, Q., Zhang, Y., Coatrieux, G., Chen, Y., Yu, H. (2021). CLEAR: Comprehensive Learning Enabled Adversarial Reconstruction for Subtle Structure Enhanced Low-Dose CT Imaging. In IEEE Transactions on Medical Imaging, 40(11), 3089-3101. https://doi.org/10.1109/TMI.2021.3097808

Zhao, T., Zhencai, S., Hui, Z., Ping, Z., & Yingyi, C. (2022). Unsupervised adversarial domain adaptation based on interpolation image for fish detection in aquaculture. Computers and Electronics in Agriculture, 198, 107004. https://doi.org/10.1016/j.compag.2022.107004

Zhu, Y., Zhuang, F., Wang, J., Ke, G., Chen, J., Bian, J., Xiong, H., & He, Q. (2021). Deep Subdomain Adaptation Network for Image Classification. In IEEE Transactions on Neural Networks and Learning Systems, 32(4), 1713-1722. https://doi.org/10.1109/TNNLS.2020.2988928