Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Skip to main content

Advertisement

Springer Nature Link
Log in

Unsupervised deep metric learning algorithm for crop disease images based on knowledge distillation networks

  • Regular Paper
  • Published:
Multimedia Systems Aims and scope Submit manuscript

Abstract

Current crop disease image classification models usually require a large number of pre-labeled training images for training, however, obtaining crop disease image datasets with accurate labels is a challenging task. Many studies have addressed this problem based on unsupervised deep metric learning (UDML), which exploits the similarity of unlabeled crop disease images in the embedding space. However, the successful implementation of UDML methods relies on accurate pseudo labels to guide model training, which may prevent the model from converging correctly during the learning process if the pseudo labels has large errors. To overcome these limitations, we propose a UDML-based algorithm for classifying crop disease images called an unsupervised deep metric learning algorithm based on a knowledge distillation network (KDUM). The KDUM algorithm can extract accurate pseudo labels from unlabeled crop disease images to guide the training process of the model. Specifically, we map the crop disease images to the metric space through the mapper of the teacher network. A clustering algorithm is used in the metric space to generate hard pseudo labels with noise for computing the loss in the metric space. At the same time, the classifiers of the teacher network are used to generate robust soft pseudo labels as auxiliary training information. The information is used with the output of the classifiers of the student network to calculate the soft cross-entropy loss, which is used to assist in guiding the training of the student network. When updating the model parameters, an exponentially weighted moving average is used to smoothly update the parameters of the teacher network. Mitigating the problem of error amplification caused by noisy pseudo labels generated by clustering. We conducted comparative experiments with seven previous years’ state-of-the-art (SOTA) methods on three datasets. The results show that our algorithm improves the NMI metrics by about 3.61% on the Plant Village dataset, 3.09% on the AFL-33 dataset, and 7.76% on the Kaggle Tomato dataset compared to the SOTA methods.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Subscribe and save

Springer+ Basic
$34.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9

Similar content being viewed by others

Data availibility

The datasets analyzed during the current study are available from the corresponding author upon reasonable request.

References

  1. Savary, S., Willocquet, L.: Modeling the impact of crop diseases on global food security. Annu. Rev. Phytopathol. 58, 313–341 (2020)

    Article  Google Scholar 

  2. Chaube, H., Pundhir, V.: Crop diseases and their management (2005)

  3. Wang, A., Zhang, W., Wei, X.: A review on weed detection using ground-based machine vision and image processing techniques. Comput. Electron. Agric. 158, 226–240 (2019)

    Article  Google Scholar 

  4. Agarwal, M., Gupta, S.K., Biswas, K.: Development of efficient cnn model for tomato crop disease identification. Sustainable Computing: Informatics and Systems 28, 100407 (2020)

    Google Scholar 

  5. Sharma, R., Das, S., Gourisaria, M.K., Rautaray, S.S., Pandey, M.: A model for prediction of paddy crop disease using cnn, 533–543 (2020)

  6. Kurmi, Y., Saxena, P., Kirar, B.S., Gangwar, S., Chaurasia, V., Goel, A.: Deep cnn model for crops’ diseases detection using leaf images. Multidimension. Syst. Signal Process. 33(3), 981–1000 (2022)

    Article  Google Scholar 

  7. Fenu, G., Malloci, F.M.: An application of machine learning technique in forecasting crop disease. In: Proceedings of the 3rd International Conference on Big Data Research, pp. 76–82 (2019)

  8. Coulibaly, S., Kamsu-Foguem, B., Kamissoko, D., Traore, D.: Deep neural networks with transfer learning in millet crop images. Comput. Ind. 108, 115–120 (2019)

    Article  Google Scholar 

  9. Miao, L., Jingxian, W., Hualong, L., Zelin, H., XuanJiang, Y., Xiaoping, H., Weihui, Z., Jian, Z., Sisi, F.: Method for identifying crop disease based on cnn and transfer learning. Smart Agriculture 1(3), 46 (2019)

    Google Scholar 

  10. Rangarajan Aravind, K., Raja, P.: Automated disease classification in (selected) agricultural crops using transfer learning. Automatika: časopis za automatiku, mjerenje, elektroniku, računarstvo i komunikacije 61(2), 260–272 (2020)

  11. Kaya, M., Bilge, H.Ş: Deep metric learning: A survey. Symmetry 11(9), 1066 (2019)

    Google Scholar 

  12. Ge, Y., Chen, D., Li, H.: Mutual mean-teaching: Pseudo label refinery for unsupervised domain adaptation on person re-identification. arXiv preprint arXiv:2001.01526 (2020)

  13. Wang, X., Qi, G.-J.: Contrastive learning with stronger augmentations. IEEE Trans. Pattern Anal. Mach. Intell. 45(5), 5549–5560 (2022)

    Google Scholar 

  14. Zimmermann, R.S., Sharma, Y., Schneider, S., Bethge, M., Brendel, W.: Contrastive learning inverts the data generating process. In: International Conference on Machine Learning, pp. 12979–12990 (2021). PMLR

  15. Talordphop, K., Sukparungsee, S., Areepong, Y.: New modified exponentially weighted moving average-moving average control chart for process monitoring. Connect. Sci. 34(1), 1981–1998 (2022)

    Article  Google Scholar 

  16. Celebi, M.E., Aydin, K.: Unsupervised learning algorithms 9 (2016)

  17. Vahdat, A., Kautz, J.: Nvae: A deep hierarchical variational autoencoder. Adv. Neural. Inf. Process. Syst. 33, 19667–19679 (2020)

    Google Scholar 

  18. Khattar, D., Goud, J.S., Gupta, M., Varma, V.: Mvae: Multimodal variational autoencoder for fake news detection. In: The World Wide Web Conference, pp. 2915–2921 (2019)

  19. Yamamoto, R., Song, E., Kim, J.-M.: Parallel wavegan: A fast waveform generation model based on generative adversarial networks with multi-resolution spectrogram. In: ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 6199–6203 (2020). IEEE

  20. Kumar, K., Kumar, R., De Boissiere, T., Gestin, L., Teoh, W.Z., Sotelo, J., De Brebisson, A., Bengio, Y., Courville, A.C.: Melgan: Generative adversarial networks for conditional waveform synthesis. Advances in neural information processing systems 32 (2019)

  21. Torfi, A., Fox, E.A.: Corgan: correlation-capturing convolutional generative adversarial networks for generating synthetic healthcare records. arXiv preprint arXiv:2001.09346 (2020)

  22. Li, R., Jiao, Q., Cao, W., Wong, H.-S., Wu, S.: Model adaptation: Unsupervised domain adaptation without source data. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 9641–9650 (2020)

  23. Pinheiro, P.O.: Unsupervised domain adaptation with similarity learning. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 8004–8013 (2018)

  24. Wu, X., Fan, X., Luo, P., Choudhury, S.D., Tjahjadi, T., Hu, C.: From laboratory to field: Unsupervised domain adaptation for plant disease recognition in the wild. Plant Phenomics 5, 0038 (2023)

    Article  Google Scholar 

  25. Yang, J., Shi, S., Wang, Z., Li, H., Qi, X.: St3d: Self-training for unsupervised domain adaptation on 3d object detection. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 10368–10378 (2021)

  26. Yao, X., She, D., Zhang, H., Yang, J., Cheng, M.-M., Wang, L.: Adaptive deep metric learning for affective image retrieval and classification. IEEE Trans. Multimedia 23, 1640–1653 (2020)

    Article  Google Scholar 

  27. Liu, Q., Li, W., Chen, Z., Hua, B.: Deep metric learning for image retrieval in smart city development. Sustain. Cities Soc. 73, 103067 (2021)

    Article  Google Scholar 

  28. Wojke, N., Bewley, A.: Deep cosine metric learning for person re-identification. In: 2018 IEEE Winter Conference on Applications of Computer Vision (WACV), pp. 748–756 (2018). IEEE

  29. Zou, G., Fu, G., Peng, X., Liu, Y., Gao, M., Liu, Z.: Person re-identification based on metric learning: a survey. multimedia tools and applications 80(17), 26855–26888 (2021)

  30. Deng, B., Jia, S., Shi, D.: Deep metric learning-based feature embedding for hyperspectral image classification. IEEE Trans. Geosci. Remote Sens. 58(2), 1422–1435 (2019)

    Article  Google Scholar 

  31. Tang, J., Li, D., Tian, Y.: Image classification with multi-view multi-instance metric learning. Expert Syst. Appl. 189, 116117 (2022)

    Article  Google Scholar 

  32. Fang, Z., Ren, J., Marshall, S., Zhao, H., Wang, Z., Huang, K., Xiao, B.: Triple loss for hard face detection. Neurocomputing 398, 20–30 (2020)

    Article  Google Scholar 

  33. Sun, Y., Cheng, C., Zhang, Y., Zhang, C., Zheng, L., Wang, Z., Wei, Y.: Circle loss: A unified perspective of pair similarity optimization. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 6398–6407 (2020)

  34. Zhang, Z., Sabuncu, M.: Generalized cross entropy loss for training deep neural networks with noisy labels. Advances in neural information processing systems 31 (2018)

  35. Ding, S., Lin, L., Wang, G., Chao, H.: Deep feature learning with relative distance comparison for person re-identification. Pattern Recogn. 48(10), 2993–3003 (2015)

    Article  Google Scholar 

  36. Yu, B., Tao, D.: Deep metric learning with tuplet margin loss. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp. 6490–6499 (2019)

  37. Kim, S., Kim, D., Cho, M., Kwak, S.: Proxy anchor loss for deep metric learning. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 3238–3247 (2020)

  38. Teh, E.W., DeVries, T., Taylor, G.W.: Proxynca++: Revisiting and revitalizing proxy neighborhood component analysis. In: Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part XXIV 16, pp. 448–464 (2020). Springer

  39. Zhu, Y., Yang, M., Deng, C., Liu, W.: Fewer is more: A deep graph metric learning perspective using fewer proxies. Adv. Neural. Inf. Process. Syst. 33, 17792–17803 (2020)

    Google Scholar 

  40. Wang, F., Liu, H.: Understanding the behaviour of contrastive loss. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 2495–2504 (2021)

  41. Kim, S., Kim, D., Cho, M., Kwak, S.: Embedding transfer with label relaxation for improved metric learning. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 3967–3976 (2021)

  42. Yan, J., Luo, L., Deng, C., Huang, H.: Unsupervised hyperbolic metric learning. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 12465–12474 (2021)

  43. Liu, L., Huang, S., Zhuang, Z., Yang, R., Tan, M., Wang, Y.: Das: Densely-anchored sampling for deep metric learning. In: European Conference on Computer Vision, pp. 399–417 (2022). Springer

  44. Kirchhof, M., Roth, K., Akata, Z., Kasneci, E.: A non-isotropic probabilistic take on proxy-based deep metric learning. In: European Conference on Computer Vision, pp. 435–454 (2022). Springer

  45. Kim, S., Kim, D., Cho, M., Kwak, S.: Self-taught metric learning without labels. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 7431–7441 (2022)

  46. Wang, S., Zeng, Q., Zhang, X., Ni, W., Cheng, C.: Multi-modal pseudo-information guided unsupervised deep metric learning for agricultural pest images. Inf. Sci. 630, 443–462 (2023)

    Article  Google Scholar 

  47. Yu, L., Yazici, V.O., Liu, X., Weijer, J.v.d., Cheng, Y., Ramisa, A.: Learning metrics from teachers: Compact networks for image embedding. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 2907–2916 (2019)

  48. Van Assche, K., Beunen, R., Verweij, S.: Comparative planning research, learning, and governance: The benefits and limitations of learning policy by comparison. Urban Planning 5(1), 11–21 (2020)

    Article  Google Scholar 

  49. Ainam, J.-P., Qin, K., Liu, G., Luo, G.: View-invariant and similarity learning for robust person re-identification. IEEE Access 7, 185486–185495 (2019)

    Article  Google Scholar 

  50. Lian, S., Dong, X.-l., Li, P.-l., Wang, C.-x., Zhou, S.-y., Li, B.-h.: Effects of temperature and moisture on conidia germination, infection, and acervulus formation of the apple marssonina leaf blotch pathogen (diplocarpon mali) in china. Plant Disease 105(4), 1057–1064 (2021)

  51. Shrivastava, V.K., Pradhan, M.K.: Rice plant disease classification using color features: a machine learning paradigm. Journal of Plant Pathology 103, 17–26 (2021)

    Article  Google Scholar 

  52. Jadhav, S.B.: Convolutional neural networks for leaf image-based plant disease classification. IAES International Journal of Artificial Intelligence 8(4), 328 (2019)

    Google Scholar 

  53. Mukti, I.Z., Biswas, D.: Transfer learning based plant diseases detection using resnet50. In: 2019 4th International Conference on Electrical Information and Communication Technology (EICT), pp. 1–6 (2019). IEEE

  54. Pardede, H.F., Suryawati, E., Sustika, R., Zilvan, V.: Unsupervised convolutional autoencoder-based feature learning for automatic detection of plant diseases. In: 2018 International Conference on Computer, Control, Informatics and Its Applications (IC3INA), pp. 158–162 (2018). IEEE

  55. Jin, H., Li, Y., Qi, J., Feng, J., Tian, D., Mu, W.: Grapegan: Unsupervised image enhancement for improved grape leaf disease recognition. Comput. Electron. Agric. 198, 107055 (2022)

    Article  Google Scholar 

  56. Mohanty, S.P., Hughes, D.P., Salathé, M.: Using deep learning for image-based plant disease detection. Front. Plant Sci. 7, 1419 (2016)

    Article  Google Scholar 

  57. Agarwal, D., Chawla, M., Tiwari, N.: Plant leaf disease classification using deep learning: A survey. In: 2021 Third International Conference on Inventive Research in Computing Applications (ICIRCA), pp. 643–650 (2021). IEEE

  58. Wang, S., Zeng, Q., Ni, W., Cheng, C., Wang, Y.: Odp-transformer: Interpretation of pest classification results using image caption generation techniques. Comput. Electron. Agric. 209, 107863 (2023)

    Article  Google Scholar 

  59. He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 770–778 (2016)

  60. Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., Rabinovich, A.: Going deeper with convolutions. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 1–9 (2015)

  61. Radosavovic, I., Kosaraju, R.P., Girshick, R., He, K., Dollár, P.: Designing network design spaces. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 10428–10436 (2020)

  62. Movshovitz-Attias, Y., Toshev, A., Leung, T.K., Ioffe, S., Singh, S.: No fuss distance metric learning using proxies. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 360–368 (2017)

  63. Wang, Q., Cheng, J., Gao, Q., Zhao, G., Jiao, L.: Deep multi-view subspace clustering with unified and discriminative learning. IEEE Trans. Multimedia 23, 3483–3493 (2020)

    Article  Google Scholar 

  64. Sinaga, K.P., Yang, M.-S.: Unsupervised k-means clustering algorithm. IEEE access 8, 80716–80727 (2020)

    Google Scholar 

Download references

Acknowledgements

All authors have contributed equally. All authors reviewed the manuscript.

Funding

This work is supported by National Key R&D Program of China [2022ZD0119501]; NSFC [52374221]; Sci. & Tech. Development Fund of Shandong Province of China [ZR2022MF288, ZR2023MF097];the Taishan Scholar Program of Shandong Province[ts20190936].

Author information

Authors and Affiliations

  1. College of Computer Science and Engineering, Shandong University of Science and Technology, Qianwangang Road, Qingdao, 266590, Shandong, China

    Qingtian Zeng, Xinheng Li, Shansong Wang, Weijian Ni & Hua Duan

  2. Agricultural Information Institute of CAAS, Haidian District, Beijing, 100081, China

    Nengfu Xie

  3. National Climate Center, Zhongguancun District, Beijing, 100035, China

    Fengjin Xiao

Authors
  1. Qingtian Zeng
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Xinheng Li
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Shansong Wang
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Weijian Ni
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Hua Duan
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Nengfu Xie
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Fengjin Xiao
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding authors

Correspondence to Qingtian Zeng or Shansong Wang.

Ethics declarations

Conflict of interest

Conflict of interest The authors declare no Conflict of interest.

Additional information

Communicated by Bing-kun Bao.

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Zeng, Q., Li, X., Wang, S. et al. Unsupervised deep metric learning algorithm for crop disease images based on knowledge distillation networks. Multimedia Systems 30, 283 (2024). https://doi.org/10.1007/s00530-024-01491-w

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1007/s00530-024-01491-w

Keywords

Search

Navigation