Geomechanics and Engineering

Techno-Press (테크노프레스)
권호 표지
DOI QR코드

DOI QR Code

Automated ground penetrating radar B-scan detection enhanced by data augmentation techniques

  • Donghwi Kim (Department of Civil and Environmental Engineering, Hongik University) ;
  • Jihoon Kim (Department of Civil and Environmental Engineering, Hongik University) ;
  • Heejung Youn (Department of Civil and Environmental Engineering, Hongik University)
  • Received : 2023.12.31
  • Accepted : 2024.06.21
  • Published : 2024.07.10
⟨ Previous Next ⟩

Abstract

This research investigates the effectiveness of data augmentation techniques in the automated analysis of B-scan images from ground-penetrating radar (GPR) using deep learning. In spite of the growing interest in automating GPR data analysis and advancements in deep learning for image classification and object detection, many deep learning-based GPR data analysis studies have been limited by the availability of large, diverse GPR datasets. Data augmentation techniques are widely used in deep learning to improve model performance. In this study, we applied four data augmentation techniques (geometric transformation, color-space transformation, noise injection, and applying kernel filter) to the GPR datasets obtained from a testbed. A deep learning model for GPR data analysis was developed using three models (Faster R-CNN ResNet, SSD ResNet, and EfficientDet) based on transfer learning. It was found that data augmentation significantly enhances model performance across all cases, with the mAP and AR for the Faster R-CNN ResNet model increasing by approximately 4%, achieving a maximum mAP (Intersection over Union = 0.5:1.0) of 87.5% and maximum AR of 90.5%. These results highlight the importance of data augmentation in improving the robustness and accuracy of deep learning models for GPR B-scan analysis. The enhanced detection capabilities achieved through these techniques contribute to more reliable subsurface investigations in geotechnical engineering.

Keywords

Acknowledgement

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (No. 2023R1A2C1007635), partly supported by 2024 Hongik University Innovation Support Program Fund.

References

  1. Abdelmawla, A., Ma, S., Yang, J.J. and Kim, S.S. (2023), "Subsurface anomaly detection utilizing synthetic GPR images and deep learning model", Geomech. Eng., 33(2), 203-209. https://doi.org/10.12989/gae.2023.33.2.203.
  2. Alani, A.M. and Tosti, F. (2018), "GPR applications in structural detailing of a major tunnel using different frequency antenna systems", Constr. Build. Mater., 158, 1111-1122. https://doi.org/10.1016/j.conbuildmat.2017.09.100.
  3. Alzubaidi, L., Bai, J., Al-Sabaawi, A., Santamaria, J., Albahri, A., Al-dabbagh, B.S.N., Fadhel, M.A., Manoufali, M., Zhang, J., and Al-Timemy, A.H. (2023), "A survey on deep learning tools dealing with data scarcity: definitions, challenges, solutions, tips, and applications", J. Big Data, 10(1), 1-82. https://doi.org/10.1186/s40537-023-00727-2.
  4. Baker, G.S., Jordan, T.E., Pardy, J., Baker, G.S. and Jol, H.M. (2007), An introduction to ground penetrating radar (GPR). Stratigraphic Analyses Using GPR, Geological Society of America.
  5. Bandara, K., Hewamalage, H., Liu, Y.H., Kang, Y. and Bergmeir, C. (2021), "Improving the accuracy of global forecasting models using time series data augmentation", Pattern Recognition, 120, 108148. https://doi.org/10.1016/j.patcog.2021.108148.
  6. Bansal, M.A., Sharma, D.R. and Kathuria, D.M. (2022), "A systematic review on data scarcity problem in deep learning: solution and applications", ACM Computing Surveys (CSUR), 54(10), 1-29. https://doi.org/10.1145/3502287.
  7. Benedetto, A. (2013), "A three dimensional approach for tracking cracks in bridges using GPR", J. Appl. Geophys., 97, 37-44. https://doi.org/10.1016/j.jappgeo.2012.12.010.
  8. Benedetto, A., Manacorda, G., Simi, A. and Tosti, F. (2012), "Novel perspectives in bridges inspection using GPR", Nondestruct. Test. Eval., 27(3), 239-251. https://doi.org/10.1080/10589759.2012.694883.
  9. Cap, Q.H., Uga, H., Kagiwada, S. and Iyatomi, H. (2020), "Leafgan: An effective data augmentation method for practical plant disease diagnosis", IEEE T. Automat. Sci. Eng., 19(2), 1258-1267. https://doi.org/10.1109/TASE.2020.3041499.
  10. Celik, E. (2015), "Investigation of the underground structure elements of GSM towers with GPR and GSM signal effects in GPR data", Proceedings of the 2015 8th International Workshop on Advanced Ground Penetrating Radar (IWAGPR), IEEE.
  11. Chaudhari, P., Agrawal, H. and Kotecha, K. (2020), "Data augmentation using MG-GAN for improved cancer classification on gene expression data", Soft Comput., 24, 11381-11391. https://doi.org/10.1007/s00500-019-04602-2.
  12. Chun, P.J., Suzuki, M. and Kato, Y. (2023), "Iterative application of generative adversarial networks for improved buried pipe detection from images obtained by ground-penetrating radar", Comput.-Aided Civil Infrastruct. Eng., 38(17), 2472-2490. https://doi.org/10.1111/mice.13070
  13. Esteva, A., Kuprel, B., Novoa, R.A., Ko, J., Swetter, S.M., Blau, H.M. and Thrun, S. (2017), "Dermatologist-level classification of skin cancer with deep neural networks", Nature, 542(7639), 115-118. https://doi.org/10.1038/nature21056.
  14. Everingham, M., Van Gool, L., Williams, C.K., Winn, J. and Zisserman, A. (2010), "The pascal visual object classes (voc) challenge", Int. J. Comput. Vision, 88, 303-338. https://doi.org/10.1007/s11263-009-0275-4
  15. Feng, D., Wang, X. and Zhang, B. (2018), "Specific evaluation of tunnel lining multi-defects by all-refined GPR simulation method using hybrid algorithm of FETD and FDTD", Constr. Build. Mater., 185, 220-229. https://doi.org/10.1016/j.conbuildmat.2018.07.039.
  16. Frid-Adar, M., Diamant, I., Klang, E., Amitai, M., Goldberger, J. and Greenspan, H. (2018), "GAN-based synthetic medical image augmentation for increased CNN performance in liver lesion classification", Neurocomputing, 321, 321-331. https://doi.org/10.1016/j.neucom.2018.09.013.
  17. Gao, Y., Pei, L., Wang, S. and Li, W. (2021), "Intelligent detection of urban road underground targets by using ground penetrating radar based on deep learning", J. Physi.: Conf. Ser., 1757, 012081. https://doi.org/10.1088/1742-6596/1757/1/012081.
  18. GSSI (2017), "RADAN 7 manual", GSSI Nashua, NH.
  19. Gu, S., Pednekar, M. and Slater, R. (2019), "Improve image classification using data augmentation and neural networks", SMU Data Science Review, 2(2), 1.
  20. Halevy, A., Norvig, P. and Pereira, F. (2009), "The unreasonable effectiveness of data", IEEE Intell. Syst., 24(2), 8-12. https://doi.org/10.1109/MIS.2009.36.
  21. He, K., Zhang, X., Ren, S. and Sun, J. (2016), "Deep residual learning for image recognition", Proceedings of the IEEE conference on computer vision and pattern recognition.
  22. Hou, F., Lei, W., Li, S. and Xi, J. (2021), "Deep learning-based subsurface target detection from GPR scans", IEEE Sensor. J., 21(6), 8161-8171. https://doi.org/10.1109/JSEN.2021.3050262.
  23. Kato, K., Nemoto, M., Kimura, Y., Kiyohara, Y., Koga, H., Yamazaki, N., Christensen, G., Ingvar, C., Nielsen, K. and Nakamura, A. (2020), "Performance improvement of automated melanoma diagnosis system by data augmentation", Adv. Biomed. Eng., 9, 62-70. https://doi.org/10.14326/abe.9.62.
  24. Kong, S.M., Kim, D.M., Lee, D.Y., Jung, H.S. and Lee, Y.J. (2018), "Field and laboratory assessment of ground subsidence induced by underground cavity under the sewer pipe", Geomech. Eng., 16(3), 285-293. https://doi.org/10.12989/gae.2018.16.3.285.
  25. Koprucu, F., Erer, I. and Kumlu, D. "Clutter aware deep detection for subsurface radar targets", Proceedings of the 2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS, IEEE.
  26. Kouros, G., Kotavelis, I., Skartados, E., Giakoumis, D., Tzovaras, D., Simi, A. and Manacorda, G. (2018), "3d underground mapping with a mobile robot and a gpr antenna", Proceedings of the 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), IEEE.
  27. Lai, W.W., Chang, R.K. and Sham, J.F. (2017), "Detection and imaging of city's underground void by GPR", Proceedings of the 2017 9th International Workshop on Advanced Ground Penetrating Radar (IWAGPR), IEEE.
  28. Lee, H.L., Song, K.I., Qi, C. and Kim, K.Y. (2022), "Sequential prediction of TBM penetration rate using a gradient boosted regression tree during tunneling", Geomech. Eng., 29(5), 523-533. https://doi.org/10.12989/gae.2022.29.5.523.
  29. Lei, W., Hou, F., Xi, J., Tan, Q., Xu, M., Jiang, X., Liu, G. and Gu, Q. (2019), "Automatic hyperbola detection and fitting in GPR B-scan image", Automati. Constr., 106, 102839. https://doi.org/10.1016/j.autcon.2019.102839.
  30. Lin, C., Wang, X., Li, Y., Zhang, F., Xu, Z. and Du, Y. (2020), "Forward modelling and GPR imaging in leakage detection and grouting evaluation in tunnel lining", KSCE J. Civil Eng., 24, 278-294. https://doi.org/10.1007/s12205-020-1530-z.
  31. Lin, T.Y., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Dollar, P. and Zitnick, C.L. (2014), "Microsoft coco: Common objects in context", Proceedings of the Computer Vision-ECCV 2014: 13th European Conference, Zurich, Switzerland, September 6-12, 2014, Proceedings, Part V 13, Springer.
  32. Liu, H., Lin, C., Cui, J., Fan, L., Xie, X. and Spencer, B.F. (2020), "Detection and localization of rebar in concrete by deep learning using ground penetrating radar", Automat. Constr., 118, 103279. https://doi.org/10.1016/j.autcon.2020.103279.
  33. Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C.Y. and Berg, A.C. (2016), "Ssd: Single shot multibox detector", Proceedings of the Computer Vision-ECCV 2016: 14th European Conference, Amsterdam, The Netherlands, October 11-14, 2016, Proceedings, Part I 14, Springer.
  34. Liu, Z., Han, K., Xue, K., Song, Y., Liu, L., Tang, Y. and Zhu, Y. (2022), "Improving CT-image universal lesion detection with comprehensive data and feature enhancements", Multimedia Syst., 28(5), 1741-1752. https://doi.org/10.1007/s00530-022-00943-5
  35. Luo, T.X. and Lai, W.W. (2020), "GPR pattern recognition of shallow subsurface air voids", Tunn. Undergr. Sp. Tech., 99, 103355. https://doi.org/10.1016/j.tust.2020.103355.
  36. Nanni, L., Maguolo, G. and Paci, M. (2020), "Data augmentation approaches for improving animal audio classification", Ecol. Inform., 57, 101084. https://doi.org/10.1016/j.ecoinf.2020.101084.
  37. Pasolli, E., Melgani, F. and Donelli, M. (2009), "Automatic analysis of GPR images: A pattern-recognition approach", IEEE T. Geosci. Remote Sens., 47(7), 2206-2217. https://doi.org/10.1109/TGRS.2009.2012701.
  38. Persico, R. (2014), Introduction to ground penetrating radar: inverse scattering and data processing, John Wiley & Sons.
  39. Pham, M.T. and Lefevre, S. (2018), "Buried object detection from B-scan ground penetrating radar data using Faster-RCNN", Proceedings of the IGARSS 2018-2018 IEEE international geoscience and remote sensing symposium, IEEE.
  40. Qin, H., Zhang, D., Tang, Y. and Wang, Y. (2021), "Automatic recognition of tunnel lining elements from GPR images using deep convolutional networks with data augmentation", Automat. Constr., 130, 103830. https://doi.org/10.1016/j.autcon.2021.103830.
  41. Ren, S., He, K., Girshick, R. and Sun, J. (2015), "Faster r-cnn: Towards real-time object detection with region proposal networks", Adv. Neural Inform. Process. Syst., 28.
  42. Sun, C., Shrivastava, A., Singh, S. and Gupta, A. (2017), "Revisiting unreasonable effectiveness of data in deep learning era", Proceedings of the IEEE international conference on computer vision.
  43. Tallini, M., Giamberardino, A., Ranalli, D. and Scozzafava, M. (2004), "GPR survey for investigation in building foundations", Proceedings of the 10th International Conference on Grounds Penetrating Radar, 2004. GPR 2004., IEEE.
  44. Tan, M. and Le, Q. (2019), "Efficientnet: Rethinking model scaling for convolutional neural networks", Proceedings of the International conference on machine learning, PMLR.
  45. Tanoli, W.A., Sharafat, A., Park, J. and Seo, J.W. (2019), "Damage Prevention for underground utilities using machine guidance", Automat. Constr., 107, 102893. https://doi.org/10.1016/j.autcon.2019.102893.
  46. Traore, A., Ly, A.O. and Akhloufi, M.A. (2020), "Evaluating deep learning algorithms in pulmonary nodule detection", Proceedings of the 2020 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC), IEEE.
  47. Utsi, E.C. (2017), Ground penetrating radar: theory and practice, Butterworth-Heinemann, United Kingdom, Oxford.
  48. Wu, Q., Chen, Y. and Meng, J. (2020), "DCGAN-based data augmentation for tomato leaf disease identification", IEEE Access, 8, 98716-98728. https://doi.org/10.1109/ACCESS.2020.2997001.
  49. Xiang, Z., Ou, G. and Rashidi, A. (2021), "Automated translation of rebar information from GPR data into as-built BIM: A deep learning-based approach", Comput. Civil Eng., 2021, 374-381.
  50. Yang, H., Liu, Y., Li, Y., Pan, R., Wang, H., Luo, F., Wang, H. and Cao, S. (2020), "Research on the support of larger broken gateway based on the combined arch theory", Geomech. Eng., 23(2), 93-102. https://doi.org/10.12989/gae.2020.23.2.093.
  51. Yang, Y., Lu, J., Li, R., Zhao, W. and Yan, D. (2020), "Small-scale void-size determination in reinforced concrete using GPR", Adv. Civil Eng., 2020, 1-11. https://doi.org/10.1155/2020/2740309.
  52. Yuan, C., Li, S., Cai, H. and Kamat, V. R. (2018), "GPR signature detection and decomposition for mapping buried utilities with complex spatial configuration", J. Comput. Civil Eng., 32(4), 04018026. https://doi.org/10.1061/(ASCE)CP.1943-5487.0000764.