ENHANCED CROSS-CROP PLANT DISEASE RECOGNITION MODEL VIA DOMAIN-SPECIFIC ALIGNMENT NETWORK AND ELASTIC-MIXUP

Authors

  • Sani Saleh Saminu Department of Computer Engineering, Ahmadu Bello University, Zaria, Nigeria
    • Yusuf Ibrahim Department of Computer Engineering, Ahmadu Bello University, Zaria, Nigeria https://orcid.org/0000-0003-1942-4826 (unauthenticated)
      • Zaharuddeen Haruna Department of Computer Engineering, Ahmadu Bello University, Zaria, Nigeria
        • Abubakar Umar Department of Computer Engineering, Ahmadu Bello University, Zaria, Nigeria

          DOI:

          https://doi.org/10.15282/

          Keywords:

          Domain Shift, Plant Disease Recognition, DsAN, Elastic Mixup, Semantic Embedding, Zero-Shot Learning

          Abstract

          The reliable identification of crop diseases across different species is often limited by variations in imaging conditions and feature distribution differences between crops. This study specifically targets cross-crop and zero-shot recognition settings, where target domain labels are unavailable, and introduces a domain-invariant plant disease recognition framework that combines elastic mixup and a domain-specific alignment network (DsAN). A pretrained MobileNetV2 acts as the visual encoder, producing 1280-dimensional features projected into a 300-dimensional semantic space derived from Word2Vec representations of disease names. The Elastic-Mixup module jointly interpolates visual and semantic embeddings under a shared Beta (0.4, 0.4) distribution, promoting smoother transitions and richer feature diversity across the domains. Meanwhile, the DsAN aligns class-conditional subdomains through a localised maximum mean discrepancy criterion and a weak gradient-reversal mechanism, enabling balanced adaptation without margin collapse. Comprehensive experiments on the PlantVillage dataset (tomato to potato transfer) demonstrated the effectiveness of the proposed approach, achieving 97.58 % accuracy, 96.60 % precision, 96.59 % recall, and 96.59 % F1-score, representing an 88.9 % reduction in classification error compared with existing zero-shot transfer methods. These results highlight the principal contribution of this study, demonstrating that the cooperative integration of Elastic-Mixup and DsAN yields a scalable, semantically consistent, and domain-invariant solution for intelligent agricultural diagnostics.

          Downloads

          Download data is not yet available.

          Author Biographies

          • Sani Saleh Saminu, Department of Computer Engineering, Ahmadu Bello University, Zaria, Nigeria

            Sani Saleh SAMINU is currently pursuing an MSc degree in Artificial Intelligence in the Department of Computer Engineering, Ahmadu Bello University (ABU) Zaria, Nigeria. He received his BSc(Ed) in Computer Science from the same institution. His research interests include plant disease recognition, computer vision, deep learning, zero-shot learning, and intelligent systems for agricultural and medical applications. He can be contacted via:

            email: sanisaminusaleh1994@gmail.com

          • Yusuf Ibrahim, Department of Computer Engineering, Ahmadu Bello University, Zaria, Nigeria

            Dr. Yusuf Ibrahim is a Senior Lecturer in the Department of Computer Engineering at Ahmadu Bello University, Zaria, Nigeria, with over a decade of teaching and research experience. He holds a B.Eng. in Electrical Engineering (First Class Honours), an MSc and a Ph.D. in Computer Engineering. He is also a Huawei Certified ICT Associate (AI), Huawei Certified Academy Instructor, and a COREN registered engineer. His research interests span Artificial Intelligence, Natural Language Processing, Computer Vision, and Computing.

            Email: yibrahim@abu.edu.ng

          • Zaharuddeen Haruna, Department of Computer Engineering, Ahmadu Bello University, Zaria, Nigeria

            Dr. Zaharuddeen Haruna is a lecturer in the Department of Computer Engineering at Ahmadu Bello University, Zaria, Nigeria, with over seven years of teaching and research experience. He holds a B.Eng. in Electrical Engineering, an MSc and a Ph.D. in Control Engineering. He is also a COREN registered engineer. His research interests include Control Systems Design, System Modeling, Intelligent Robotics, Autonomous Systems, and Embedded Systems.

            Email: hzaharuddeen@abu.edu.ng

          • Abubakar Umar, Department of Computer Engineering, Ahmadu Bello University, Zaria, Nigeria

            Abubakar Umar     is a lecturer in the Department of Computer Engineering at Ahmadu Bello University, Zaria, Nigeria. He earned his BEng Degree from Electrical Engineering Department Ahmadu Bello University, Zaria, Nigeria, in 2011, MSc, and Ph.D. degrees from Computer Engineering Department, Ahmadu Bello University, Zaria, Nigeria, in 2017 and 2024. He specializes in various aspects of computer engineering. His primary research focus is in Control Engineering, where he explores the development and optimization of control systems for different applications. He is dedicated to advancing his research and contributing to academic knowledge in this field. He can be contacted via email at abubakaru061010@gmail.com, abuumar@abu.edu.ng

          References

          [1] M. I. Hossen, M. Awrangjeb, S. Pan, and A. A. Mamun, “Transfer learning in agriculture: A review,” Artificial Intelligence Review, vol. 58, Art. no. 97, 2025, https://doi.org/10.1007/s10462-024-11081-x

          [2] X. Fan, P. Luo, Y. Mu, R. Zhou, T. Tjahjadi, and Y. Ren, “Leaf image based plant disease identification using transfer learning and feature fusion,” Computers and Electronics in Agriculture, vol. 196, Art. no. 106892, 2022, https://doi.org/10.1016/j.compag.2022.106892

          [3] V. Tiwari, R. C. Joshi, and M. K. Dutta, “Dense convolutional neural networks based multiclass plant disease detection and classification using leaf images,” Ecological Informatics, vol. 63, Art. no. 101289, 2021, https://doi.org/10.1016/j.ecoinf.2021.101289

          [4] S. Farahani, S. Voghoei, K. Rasheed, and H. R. Arabnia, “A brief review of domain adaptation,” in Advances in Data Science and Information Engineering, 2021, pp. 877–894.

          [5] H. A. Cruz Ayala and T. Pérez González, “Social image classification from multimodal analysis,” Voces Universitarias: Perspectivas desde la Universidad, vol. 2, no. 1, pp. 67–76, 2025, https://doi.org/10.47731/vubeta.v2i1.34153

          [6] S. M. Hassan and A. K. Maji, “Deep feature-based plant disease identification using machine learning classifier,” Innovations in Systems and Software Engineering, vol. 20, no. 4, pp. 789–799, 2024, https://doi.org/10.1007/s11334-022-00513-y

          [7] A. T. Le, M. Shakiba, and I. Ardekani, “Tomato disease detection with lightweight recurrent and convolutional deep learning models for sustainable and smart agriculture,” Frontiers in Sustainability, vol. 5, Art. no. 1383182, 2024, https://doi.org/10.3389/frsus.2024.1383182

          [8] J. Jackulin and S. Murugavalli, “A comprehensive review on detection of plant disease using machine learning and deep learning approaches,” Measurement: Sensors, vol. 24, Art. no. 100441, 2022.

          [9] X. Wu, X. Fan, P. Luo, S. Das Choudhury, T. Tjahjadi, and C. Hu, “From laboratory to field: Unsupervised domain adaptation for plant disease recognition in the wild,” Plant Phenomics, vol. 5, Art. no. 0038, 2023, https://doi.org/10.34133/plantphenomics.0038

          [10] R. Sharma, A. Singh, Kavita, et al., “Plant disease diagnosis and image classification using deep learning,” Computers, Materials & Continua, vol. 71, no. 2, pp. 2125–2140, 2022, https://doi.org/10.32604/cmc.2022.020017

          [11] N. Ullah, J. A. Khan, S. Almakdi, et al., “An effective approach for plant leaf diseases classification based on a novel DeepPlantNet deep learning model,” Frontiers in Plant Science, vol. 14, Art. no. 1212747, 2023, https://doi.org/10.3389/fpls.2023.1212747

          [12] P. Kumar, J. Mathew, R. K. Sanodiya, T. Setty, and B. P. Bhaskarla, “Zero shot plant disease classification with semantic attributes,” Artificial Intelligence Review, vol. 57, no. 11, Art. no. 305, 2024. https://doi.org/10.1007/s10462-024-10950-9.

          [13] S. R. Singh and R. K. Sanodiya, “Zero-shot transfer learning framework for plant leaf disease classification,” IEEE Access, vol. 11, pp. 143861–143880, 2023, https://doi.org/10.1109/ACCESS.2023.3343759

          [14] M. Ahmed and A. Ahmed, “Palm tree disease detection and classification using residual network and transfer learning of inception ResNet,” PLOS ONE, vol. 18, no. 3, Art. no. e0282250, 2023. https://doi.org/10.1371/journal.pone.0282250.

          [15] M. Tanveer, T. Rajani, R. Rastogi, Y.-H. Shao, and M. A. Ganaie, “Comprehensive review on twin support vector machines,” Annals of Operations Research, 2022, https://doi.org/10.1007/s10479-022-04575-w

          [16] Y. Gulzar, “Fruit image classification model based on MobileNetV2 with deep transfer learning technique,” Sustainability, vol. 15, no. 3, Art. no. 1906, 2023, https://doi.org/10.3390/su15031906

          [17] S. U. M. Rao, K. Sreekala, P. S. Rao, N. Shirisha, G. Srinivas, and E. Sreedevi, “Plant disease classification using novel integration of deep learning CNN and graph convolutional networks,” Indonesian Journal of Electrical Engineering and Computer Science, vol. 36, no. 3, pp. 1721–1730, 2024. https://doi.org/10.11591/ijeecs.v36.i3.pp1721-1730.

          [18] S. Lengyel, S. Garg, M. Milford, and J. C. van Gemert, “Zero-shot day–night domain adaptation with a physics prior,” in Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2021, pp. 4399–4409.

          [19] X. Li, M. Fang, and B. Chen, “Generalized zero-shot domain adaptation with target unseen class prototype learning,” Neural Computing and Applications, vol. 34, no. 20, pp. 17793–17807, 2022.

          [20] Y. Ma, Z. Zhang, H. L. Yang, and Z. Yang, “An adaptive adversarial domain adaptation approach for corn yield prediction,” Computers and Electronics in Agriculture, vol. 187, Art. no. 106314, 2021, https://doi.org/10.1016/j.compag.2021.106314

          [21] B. Yang, M. Li, F. Li, et al., “A novel plant type, leaf disease and severity identification framework using CNN and transformer with multi-label method,” Scientific Reports, vol. 14, Art. no. 11664, 2024, https://doi.org/10.1038/s41598-024-62452-x

          [22] P. Wang, K. Han, X.-S. Wei, L. Zhang, and L. Wang, “Contrastive learning based hybrid networks for long-tailed image classification,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021, pp. 943–952.

          [23] R. Katafuchi and T. Tokunaga, “LEA-Net: Layer-wise external attention network for efficient color anomaly detection,” arXiv, arXiv:2109.05493, 2021. [Online]. Available: https://arxiv.org/abs/2109.05493.

          [24] S. Vallabhajosyula, V. Sistla, and V. K. K. Kolli, “Transfer learning-based deep ensemble neural network for plant leaf disease detection,” Journal of Plant Diseases and Protection, vol. 129, no. 3, pp. 545–558, 2022.

          [25] A. Prasad, N. Mehta, M. Horak, and W. D. Bae, “A two-step machine learning approach for crop disease detection using GAN and UAV technology,” Remote Sensing, vol. 14, no. 19, Art. no. 4765, 2022. https://doi.org/10.3390/rs14194765.

          [26] Y. F. Zhang, J. Wang, J. Liang, et al., “Domain-specific risk minimization for domain generalization,” in Proceedings of the ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2023, pp. 3409–3421.

          Downloads

          Published

          2026-06-09

          Issue

          Vol. 12 No. 1 (2026): International Journal of Software Engineering and Computer Systems

          Section

          Full Length Article

          License

          Copyright (c) 2026 The Author(s)

          Creative Commons License

          This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

          How to Cite

          [1]
          Sani Saleh Saminu, Yusuf Ibrahim, Zaharuddeen Haruna, and Abubakar Umar, “ENHANCED CROSS-CROP PLANT DISEASE RECOGNITION MODEL VIA DOMAIN-SPECIFIC ALIGNMENT NETWORK AND ELASTIC-MIXUP”, IJSECS, vol. 12, no. 1, pp. 39–49, Jun. 2026, doi: 10.15282/.

          Share