Top.Mail.Ru
искать
TRANSLATED ARTICLES

The application of artificial intelligence in geotechnical investigation

Авторы
МЭЙ Х.Компания «Институт транспортного проектирования и планирования провинции Шэньси», г. Сиань, провинция Шэньси, Китай
ЧЖАН В.Институт искусственного интеллекта Сианьского университета электронных технологий, г. Сиань, провинция Шэньси, Китай
ГУ Цз.Институт искусственного интеллекта Сианьского университета электронных технологий, г. Сиань, провинция Шэньси, Китай

Abstract: We present to the readers an adapted translation of the article “The application of artificial intelligence in geotechnical investigation” by Chinese researchers. This work was published electronically in the collection of scientific papers “Advances in Artificial Intelligence, Big Data and Algorithms” by the international publisher “IOS Press”. It is available in open access under the CC BY-NC 4.0 license, which allows copying and redistribution of the article, adaptation, modification, and creation of new works based on it, but not for commercial purposes, provided that the license type, changes made are indicated and the original source is referenced. In our case, the full reference to the original source is provided at the end of the translation.

The standard penetration test (SPT) and dynamic probing test (DPT) are commonly used exploration methods in geotechnical investigation. However, errors can occur during data collection, often attributed to factors such as human error. To mitigate this issue, this paper proposes the utilization of an improved YOLOv5 object detection algorithm, a form of artificial intelligence technology, to automatically count the number of hammer strikes during geotechnical investigations. The proposed approach incorporates several enhancements to the YOLOv5 network architecture. Firstly, a focal loss function is introduced to address sample imbalance, ensuring better handling of different classes of hammer strikes. Additionally, online hard example mining technology is employed to improve model accuracy by focusing on challenging samples that are most informative for training. The improved YOLOv5 model is then applied to detect hammer strikes in SPT and DPT tests. To facilitate training and evaluation, a hammer detection dataset is created, tailored to the specific requirements of geotechnical investigation. Experimental results demonstrate the superior performance of the proposed improved YOLOv5 object detection model on the hammer detection dataset.

Keywords: geotechnical investigation; field dynamic tests of soils; SPT method; DPT method; artificial intelligence; neural network; YOLOv5 algorithm; hammer detection; hammer strikes detection; model training; online hard example mining; number of hammer strikes; automated counting

DOI: 10.58339/2949-0677-2025-7-3-68-78

UDC: 624.131.385; 004

For citation: Mei H., Zhang W., Gu J. Primenenie iskusstvennogo intellekta pri geotekhnicheskikh izyskaniyakh (adapt. per. s angl.) [The application of artificial intelligence in geotechnical investigation (adapted translation from English into Russian)] // Geoinfo. 2025. T. 7. № 3. S. 68–78. DOI:10.58339/2949-0677-2025-7-3-68-78 (in Rus.)

REFERENCES:

  1. Girshick R., Donahue J., Darrell T., Malik J. Rich feature hierarchies for accurate object detection and semantic segmentation // Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2014). 2014. P. 580–587.
  2. Girshick R. Fast R-CNN // Proceedings of the 2015 IEEE International Conference on Computer Vision (ICCV). 2015. P. 1440–1448.
  3. Ren S., He K., Girshick R., Sun J. Faster R-CNN: Towards real-time object detection with region proposal networks // Proceedings of the 2015 IEEE International Conference on Computer Vision (ICCV 2015). 2015 . P. 91–99.
  4. Redmon J., Divvala S., Girshick R., Farhadi A. You only look once: Unified, real-time object detection // Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2016). 2016. P. 779–788.
  5. Liu W., Anguelov D., Erhan D., Szegedy C., Reed S., Fu C.Y., Berg A.C. SSD: Single shot multibox detector // Proceedings of the 2016 European Conference on Computer Vision (ECCV 2016), 2016. P. 21–37.
  6. Redmon J., Farhadi A. YOLO9000: Better, faster, stronger // Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2017), 2017. P. 6517–6525.
  7. Redmon J., Farhadi A. YOLOv3: An incremental improvement // Proceedings of the 2018 IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2018). 2018. P. 6517–6525.
  8. Bochkovskiy A., Wang C.Y., Liao H.Y.M. YOLOv4: Optimal speed and accuracy of object detection // Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2020). 2020. P. 10934–10944.
  9. Wong B., AbdSalam R., Wong S.H. YOLOv5: A better, faster, stronger object detector // Proceedings of the 2020 International Conference on Neural Information Processing (ICONIP 2020). 2020. P. 1078–1090.
  10. Lin T.Y., Maire M., Belongie S., Hays J., Perona P., Ramanan D., Dollar P., Zitnick C.L. Microsoft COCO: Common objects in context. Lecture Notes in Computer Science // Proceedings of the 2014 European Conference on Computer Vision (ECCV 2014). Part V. Lecture Notes in Computer Science. Switzerland: Springer International Publishing, 2014. Vol. 8693. P. 740–755.
  11. He K., Gkioxari G., Dollar P., Girshick R. Mask R-CNN // Proceedings of the 2017 IEEE International Conference on Computer Vision (ICCV 2017). 2017. P. 2980–2988.
  12. Liu X., Wang Z., He Y., Liu Q. Research on small target detection based on deep learning // Tactical Missile Technology. 2019. Vol. 1. P. 100–107.
  13. He K., Zhang X., Ren S., Sun J. Deep residual learning for image recognition // Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2016). 2016. P. 770–778.
  14. Liu S., Qi L., Qin H., Shi J., Jia J. Path aggregation network for instance segmentation // Proceedings of the 2018 IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2018). 2018. P. 8759–8768.
  15. Lisi W., Yu Z. Glass bottle mouth defect detection based on YOLOv5 // Yangtze River Information Communication. 2023. Vol. 36. № 1. P. 9–11 (in Chinese).
  16. Jiang L., Cui Y. Small object detection based on YOLOv5 // Computer Knowledge and Technology. 2021. Vol. 17. № 26. P. 131–133.
  17. Song Y.X., Zhao Y., Zhang J.Y., Zhu W.P., Yang Z.H., Zhang Q. Design of sitting posture monitoring system based on YOLOv5 // FEMT (Frontiers of Electronic Materials). 2023. Vol. 19. № 8. P. 22–25.
  18. Lin T.Y., Dollar P., Girshick R., He K., Hariharan B., Belongie S. Feature pyramid networks for object detection // Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2017). 2017. P. 936–944.
  19. Shrivastava A., Gupta A., Girshick R. Training region-based object detectors with Online Hard Example Mining // Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2016). 2016. P. 761–769.
  20. Shi F., Qiu Z., Han Q., Li J., Qian H., Xiang W. Improved faster R-CNN algorithm based on variable weight loss function and hard example mining module // Computer and Modernization, 2020. Vol. 8. P. 56–62 (in Chinese).
  21. Lin T.Y., Goyal P., Girshick R., He K., Dollar P. Focal loss for dense object detection // Proceedings of the 2018 IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2018). 2018. P. 2980–2988.
  22. Huang J., Zhang G. A Review of object detection algorithms based on deep convolutional neural networks // Computer Engineering and Applications. 2020. Vol. 56. № 17. P. 12–23.

Article in RSCI: https://www.elibrary.ru/item.asp?id=85275400