Res. Agr. Eng., 2025, 71(1):1-9 | DOI: 10.17221/67/2024-RAE

Towards interpretability: Assessment of residual networks for tomato leaf disease classificationOriginal Paper

Raphael Berdin1, Rob Christian Caduyac ORCID...1
1 Department of Electrical Engineering, University of the Philippines Los Baños, Laguna, Philippines

The tomato occupies a prominent place in the Philippines’ agricultural economy. However, tomato leaf diseases are challenges in tomato crop production leading to economic losses. Among the tomato leaf diseases, early blight and Septoria leaf spot are prevalent in the Philippines due to the climate. Thus, the accurate identification of diseases affecting tomato leaves is essential. Currently, a visual inspection is the primary method for diagnosing tomato leaf diseases which is time-consuming and inefficient. This study aims to develop a quantized Residual Network with convolutional 50 layer (ResNet-50) based model to classify tomato leaves as healthy or affected by Septoria leaf spot or early blight. Furthermore, to enhance the reliability of the models’ classification, gradient-weighted class activation mapping (Grad-CAM) was implemented. In contrast with the visual inspection, a programmed system does not get tired and can provide consistent performance results. As a result, the original 32-bit floating point model attained an accuracy rate of 91.22%. The quantized 16-bit floating point model demonstrated comparable performance with 90.10% accuracy with a 50% reduction in the model size and inference time of 0.3942 seconds. The minimal accuracy loss of the 16-bit model relative to the 32-bit model is due to the post-training quantization. The reduction to 16-bit precision is significant for the future deployment of edge devices where resources are limited.

Keywords: tomato leaf disease classification; Grad-CAM; quantization; ResNet50

Received: July 29, 2024; Accepted: November 26, 2024; Prepublished online: March 11, 2025; Published: March 20, 2025  Show citation

ACS AIP APA ASA Harvard Chicago Chicago Notes IEEE ISO690 MLA NLM Turabian Vancouver
Berdin R, Caduyac RC. Towards interpretability: Assessment of residual networks for tomato leaf disease classification. Res. Agr. Eng. 2025;71(1):1-9. doi: 10.17221/67/2024-RAE.
Share...
Download citation
  • BibTeX (.bib)
  • Bookends (.ris)
  • EasyBib (.ris)
  • EndNote (.enw)
  • EndNote 8 (.xml)
  • ISI WoS (.isi)
  • Medlars (.medlars)
  • Mendeley (.ris)
  • MODS (.xml)
  • Papers (.ris)
  • RefWorks (.txt)
  • RefManager (.ris)
  • RIS (.ris)
  • MS Word (.xml)
  • Zotero (.ris)
    • Open full article

    References

    1. Agarwal M., Singh A., Arjaria S., Sinha A., Gupta S. (2020): ToLeD: Tomato leaf disease detection using convolutional neural network. Procedia Computer Science, 167: 293-301. Go to original source...
    2. Alzubaidi L., Zhang J., Humaidi A.J., Al-Dujaili A., Duan Y., Al-Shamma O., Santamaria J., Fadhel M.A., Al-Amidie M., Farhan L. (2021): Review of deep learning: Concepts, CNN architectures, challenges, applications, future directions. Journal of Big Data, 8: 53. Go to original source...
    3. Arnal Barbedo J.G. (2013): Digital image processing techniques for detecting, quantifying, and classifying plant diseases. SpringerOpen Journal Barbedo, 2: 660. Go to original source...
    4. Attalah O. (2023): Tomato leaf disease classification via compact convolutional neural networks with transfer learning and feature selection. Horticulture, 9: 149. Go to original source...
    5. Baser P., Saini J., Kotecha K. (2023): TomConv: An improved CNN model for diagnosis of diseases in tomato plant leaves. Procedia Computer Science, 218: 1825-1833. Go to original source...
    6. Branthôme F.X. (2022): Worldwide (Total Fresh) Tomato Production Exceeds 187 Million Tonnes in 2020. Available at https://www.tomatonews.com/en/worldwide-total-fresh-tomato-production-exceeds-187-million-tonnes-in-2020_2_1565 (accessed July 10, 2024).
    7. Chopra G., Whig P. (2021): Analysis of tomato leaf disease identification techniques. Journal of Computer Science and Engineering, 2: 98-103. Go to original source...
    8. De Luna R.G., Dadios E.P., Bandala A.A. (2018): Automated image capturing system for deep learning-based tomato plant leaf disease. Detection and recognition. In: Proceedings of TENCON 2018 IEEE Region 10 Conference. Jeju, South Korea, Oct 28-31, 2018: 1414-1419. Go to original source...
    9. Department of Agriculture (2017): Tomato Production Guide. Available at https://cagayanvalley.da.gov.ph/wp-content/uploads/2018/02/Tomato.pdf (accessed July 10, 2024).
    10. Elhassouny A., Smarandache F. (2019): Smart mobile application to recognize tomato leaf diseases using convolutional neural networks. In: Proceedings of the 2019 International Conference of Computer Science and Renewable Energies. Agadir, Morocco, Jul 22-24, 2019: 1-4. Go to original source...
    11. Gorme A.L., Gonzaga Z., Capuno O., Rom J., McDougall S., Goldwater A., Rogers G. (2017): Growth and yield of tomato (Solanum lycopersicum) as influenced by different soil organic amendments and types of cultivation. Annals of Tropical Research, 39: 116-128. Go to original source...
    12. He K., Zhang X., Ren S., Sun J. (2015): Deep residual learning for image recognition. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, Nevada, June 27-30, 2016: 770-778.
    13. Liu H., Yao D., Yang J., Li X. (2019): Lightweight convolutional neural network and its application in rolling bearing fault diagnosis under variable working conditions. Sensors, 19: 4827. Go to original source...
    14. Maeda-Gutiérrez V., Galvan C.E., Zanella L.A., Celaya J.M., Galvan J.I., Gamboa H., Luna H., Magallanes R., Guerrero C.A., Olvera C.A. (2020): Comparison of convolutional neural network architectures for classification of tomato plant diseases. Applied Sciences, 10: 1245. Go to original source...
    15. Mahmud T., Barua K., Barua A., Basnin N., Das S., Hossain M., Andersson K. (2024): Explainable AI for tomato leaf disease detection: Insights into model interpretability. In: Proceedings of the 26th International Conference on Computer and Information Technology. Cox's Bazar, Bangladesh, Dec 13-15, 2023: 1-6. Go to original source...
    16. Mandal B., Okeukwu A., Theis Y. (2021): Masked Face Recognition using ResNet-50. Available at https://arxiv.org/abs/2104.08997 (accessed July 12, 2024).
    17. Molnar C. (2022): Interpretable Machine Learning: A Guide for Making Black-box Models Explainable. Vol. 2. Available at christophm.github.io/interpretable-ml-book/ (accessed July 12, 2024).
    18. O'Shea K., Nash R. (2015): An Introduction to Convolutional Neural Networks. Available at https://arxiv.org/abs/1511.08458 (accessed July 10, 2024).
    19. Plant Village (2024): The Plant Village Solution. Available at https://Plantvillage.Psu.Edu/ (accessed July 10, 2024).
    20. Rangarajan A., Purushotaman R., Ramesh A. (2018): Tomato crop leaf disease classification using pre-trained deep learning algorithm. Procedia Computer Science, 133: 1040-1047. Go to original source...
    21. Vinay P., Kumar G., Aravind K. (2020): Classification of tomato plant leaf disease using neural networks. In: Proceedings of the International Conference on Innovative Computing and Communication. New Delhi, India, Feb 20-22, 2020: 1-5. Go to original source...
    22. Yebasse M., Shimelis B., Warku H., Ko J., Cheoi K. J. (2021): Coffee disease visualization and classification. Plants, 10: 1257. Go to original source...
    23. Zhang K., Wu Q., Liu A., Meng X. (2018): Can deep learning identify tomato leaf disease? Advances in Multimedia, 2018: 710865. Go to original source...

    This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International (CC BY NC 4.0), which permits non-comercial use, distribution, and reproduction in any medium, provided the original publication is properly cited. No use, distribution or reproduction is permitted which does not comply with these terms.


    Return to the content