Skip to main content
SpringerLink
Log in

Multi-source domain adaptation of GPR data for IED detection

  • Original Paper
  • Published:
Signal, Image and Video Processing Aims and scope Submit manuscript

Abstract

Learning an object category from only a few samples is generally inadequate for the correct classification of large-scale problems. One needs many training samples to obtain a classifier that generalizes well and has reasonable success rates for an operator. However, collecting many annotated data is not always possible in several applications, including target detection from ground penetrating radar (GPR) data. Furthermore, GPR images of target or clutter objects show a nonlinear dependence on soil permeability and permittivity. Therefore, even if enough training data were available to train a good classifier for one soil type (such as dry sand), the success of this classifier does not translate well if the soil type is changed (say, to wet sand). To decrease this domain gap, in this work, we propose to do a multi-model knowledge transfer (KT) for improvised explosive device detection from GPR data and investigate how effective it is to pass the models learned from known environments to models trained for unknown environments. We show that (1) knowledge transfer from multiple sources (i.e., multiple types of sand) generates better results than single-source transfer, and (2) as little as three training data from the unknown source increases the detection rates by 10% for single KT and 4% for multiple KT on simulated data. Further, we show that adapting the models for each type of source is a better approach than just combining all the training data in a single model. These results demonstrate that a multi-source domain adaptation approach significantly reduces data collection and manual annotation efforts and increases detection rates in unknown environments.

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

Access this article

Log in via an institution

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

Similar content being viewed by others

Data availability

All the simulation data and their input/output files are available at https://github.com/parsslab/domain_adaptation_of_gpr_data.

References

  1. Iraq coalition casualty count, deaths by IED (2009). https://web.archive.org/web/20090113201909/. http://icasualties.org/oif/IED.aspx

  2. Tommasi, T., Orabona, F., Caputo, B.: Learning categories from few examples with multi model knowledge transfer. IEEE Trans. Pattern Anal. Mach. Intell 36(5), 928–941 (2014)

    Article  Google Scholar 

  3. Biederman, I.: Recognition-by-components: a theory of human image understanding. Psychol. Rev. 94(2), 115 (1987)

    Article  Google Scholar 

  4. Wang, J., Lan, C., Liu, C., Ouyang, Y., Qin, T., et al.: Generalizing to unseen domains: a survey on domain generalization. IEEE Trans. Knowl. Data Eng. (2022). https://doi.org/10.1109/TKDE.2022.3178128

  5. Zhou, L., Ye, M., Zhang, D., et al.: Prototype-based multisource domain adaptation. IEEE Trans. Neural Netw. Learn. Syst. 33(10), 5308–5320 (2022)

    Article  Google Scholar 

  6. Liu, F., Zhang, F., Lu, J.: Heterogeneous domain adaptation: an unsupervised approach. IEEE Trans. Neural Netw. Learn. Syst. 31(12), 5588–5602 (2020)

    Article  MathSciNet  Google Scholar 

  7. Liu, Y., Tu, W., Du, B., et al.: Homologous component analysis for domain adaptation. IEEE Trans. Image Process. 29, 1074–1089 (2020)

    Article  MathSciNet  MATH  Google Scholar 

  8. Wu, H., Yan, Y., Ng, M.K., Zhang, L., Wu, Q.: Domain-attention conditional wasserstein distance for multi-source domain adaptation. ACM Trans. Intell. Syst. Technol. 11(4), 1–19 (2020)

    Article  Google Scholar 

  9. Peng, X., Bai, Q., Xia, X., Huang, Z., et al.: Moment matching for multi-source domain adaptation. In: IEEE ICCV, pp. 1406–1415 (2019)

  10. Li, Y., Yuan, L., Chen, Y.: Dynamic transfer for multi-source domain adaptation. In: IEEE CVPR, pp. 10998–11007 (2021)

  11. Ren, C.X., Liu, Y., Zhang, X., Huang, K.: Multi-source unsupervised domain adaptation via pseudo target domain. IEEE Trans. Image Process. 31, 2122–2135 (2022)

    Article  Google Scholar 

  12. Wu, H., Yan, Y., Lin, G., Yang, M., Ng, M.K., Wu, Q.: Iterative refinement for multi-source visual domain adaptation. IEEE Trans. Knowl. Data Eng. 34(6), 2810–2823 (2022)

    Google Scholar 

  13. Peng X., Saenko, K.: Synthetic to real adaptation with generative correlation alignment networks. In: IEEE WACV, pp. 1982–1991 (2018)

  14. Pinheiro, P. O.: Unsupervised domain adaptation with similarity learning. In: IEEE CVPR, pp. 8004–8013 (2018)

  15. Pan, S.J., Yang, Q.: A survey on transfer learning. IEEE Trans. Knowl. Data Eng. 22(10), 1345–1359 (2009)

    Article  Google Scholar 

  16. Weiss, K., Khoshgoftaar, T.M., Wang, D.: A survey of transfer learning. J. Big Data 3(1), 1–40 (2016)

    Article  Google Scholar 

  17. Giannopoulos, A.: Modelling ground penetrating radar by GprMax. Constr. Build. Mater. 19(10), 755–762 (2005)

    Article  Google Scholar 

  18. Warren, C., Giannopoulos, A., Giannakis, I.: gprMax: open source software to simulate electromagnetic wave propagation for ground penetrating radar. Comput. Phys. Commun. 209, 163–170 (2016)

    Article  Google Scholar 

  19. Borgioli, G., Capineri, L., Falorni, P., et al.: The detection of buried pipes from time-of-flight radar data. IEEE Trans. Geosci. Remote Sens. 46(8), 2254–2266 (2008)

    Article  Google Scholar 

  20. Genc, A., Akar, G.B.: Combination of physics-based and image-based features for landmine identification in ground penetrating radar data. J. Appl. Remote Sens. 13(2), 026503 (2019)

    Article  Google Scholar 

  21. Yuksel, S.E., Bolton, J., Gader, P.: Multiple-instance hidden Markov models with applications to landmine detection. IEEE Trans. Geosci. Remote Sens. 53(12), 6766–6775 (2015)

    Article  Google Scholar 

  22. Torrione, P.A., Morton, K.D., Sakaguchi, R., et al.: Histograms of oriented gradients for landmine detection in ground-penetrating radar data. Trans. Geosci. Remote Sens. 52(3), 1539–1550 (2013)

  23. Baydar, B., Akar, G.B., Yuksel, S.E., et al.: Fusion of KLMS and blob based pre-screener for buried landmine detection using ground penetrating radar. In: Detection and Sensing of Mines, Explosive Objects, and Obscured Targets XXI, 9823, 98231D (2016)

  24. Aydin E. and Yüksel, S.E.: Buried target detection with ground penetrating radar using deep learning method. In: IEEE SIU, pp. 1–4 (2017)

  25. Pham, M. T., Lefèvre, S.: Buried object detection from B-scan ground penetrating radar data using faster-RCNN. In: IEEE IGARSS, pp. 6804–6807 (2018)

  26. Ozkaya, U., Seyfi, L.: Deep dictionary learning application in GPR B-scan images. SIVIP 12, 1567–1575 (2018)

    Article  Google Scholar 

  27. Malof, J.M., Reichman, D., Karem, A., et al.: A large-scale multi-institutional evaluation of advanced discrimination algorithms for buried threat detection in ground penetrating radar. IEEE Trans. Geosci. Remote Sens. 57(9), 6929–6945 (2019)

    Article  Google Scholar 

  28. Wilson, J.N., et al.: A large-scale systematic evaluation of algorithms using ground-penetrating radar for landmine detection and discrimination. IEEE Trans. Geosci. Remote Sens 45(8), 2560–2572 (2007)

    Article  Google Scholar 

  29. Chen, Y.: A transfer learning model with multi-source domains for biomedical event trigger extraction. BMC Genomics 22(1), 1–18 (2021)

    Google Scholar 

  30. Guo, J., Shah, D., Barzilay, R.: Multi-source domain adaptation with mixture of experts. In: EMNLP, pp. 4694–4703 (2018)

  31. Rakshit, S., Tamboli, D., Meshram, P.S., et al.: Multi-source open-set deep adversarial domain adaptation. In: ECCV, pp. 735–750 (2020)

  32. Lowe, D.G.: Distinctive image features from scale-invariant keypoints. Int. J. Comput. Vis 60(2), 91–110 (2004)

    Article  Google Scholar 

  33. Suykens, J.A., Vandewalle, J.: Least squares support vector machine classifiers. Neural Process. Lett. 9(3), 293–300 (1999)

    Article  Google Scholar 

  34. Cawley, G.C.: Leave-one-out cross-validation based model selection criteria for weighted LS-SVMs. In: IEEE IJCNN, pp. 1661–1668 (2006)

Download references

Funding

This research did not receive any specific grant.

Author information

Authors and Affiliations

  1. Department of Electrical and Electronics Engineering, Hacettepe University, Beytepe, 06800, Ankara, Turkey

    Mehmet Oturak, Seniha Esen Yuksel & Sefa Kucuk

  2. Department of Electrical and Electronics Engineering, Erzurum Technical University, 25050, Erzurum, Turkey

    Sefa Kucuk

Authors
  1. Mehmet Oturak
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Seniha Esen Yuksel
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Sefa Kucuk
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

MO generated the data, conducted the experiments, and drafted the work. SEY conceptualized the idea, supervised the experiments, and reviewed the work. SK analyzed the results and contributed to the final manuscript.

Corresponding author

Correspondence to Seniha Esen Yuksel.

Ethics declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Ethical approval

Not applicable.

Additional information

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

Oturak, M., Yuksel, S.E. & Kucuk, S. Multi-source domain adaptation of GPR data for IED detection. SIViP 17, 1831–1839 (2023). https://doi.org/10.1007/s11760-022-02394-x

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11760-022-02394-x

Keywords

Search

Navigation