KSCE Journal of Civil and Environmental Engineering Research (대한토목학회논문집)

Korean Society of Civil Engeneers (대한토목학회)
권호 표지
DOI QR코드

DOI QR Code

Image-Based Automatic Bridge Component Classification Using Deep Learning

딥러닝을 활용한 이미지 기반 교량 구성요소 자동분류 네트워크 개발

  • 조문원 (충북대학교 토목공학부) ;
  • 이재혁 (충북대학교 토목공학부) ;
  • 유영무 (한국철도기술연구원 첨단궤도토목본부 철도구조연구팀) ;
  • 박정준 (한국철도기술연구원 첨단궤도토목본부 철도구조연구팀) ;
  • 윤형철 (충북대학교 토목공학부)
  • Received : 2021.06.03
  • Accepted : 2021.09.23
  • Published : 2021.12.01
Download PDF
⟨ Previous Next ⟩

Abstract

Most bridges in Korea are over 20 years old, and many problems linked to their deterioration are being reported. The current practice for bridge inspection mainly depends on expert evaluation, which can be subjective. Recent studies have introduced data-driven methods using building information modeling, which can be more efficient and objective, but these methods require manual procedures that consume time and money. To overcome this, this study developed an image-based automaticbridge component classification network to reduce the time and cost required for converting the visual information of bridges to a digital model. The proposed method comprises two convolutional neural networks. The first network estimates the type of the bridge based on the superstructure, and the second network classifies the bridge components. In avalidation test, the proposed system automatically classified the components of 461 bridge images with 96.6 % of accuracy. The proposed approach is expected to contribute toward current bridge maintenance practice.

우리나라의 교량은 대부분이 건설된 지 20년 이상이 지나 현재 노후화로 인하여 많은 문제점이 제기되고 있으며, 교량의 안전점검은 대부분 전문 인력의 주관적인 평가로 이루어지고 있다. 최근 교량 안전점검의 데이터의 체계적인 관리를 위해 BIM 등을 활용한 데이터 기반의 유지관리 기술들이 개발되고 있지만, BIM과 구조물의 유지관리 데이터를 연동을 위해서 영상정보를 직접 라벨링하는 수작업을 필요로한다. 따라서 본 논문에서는 이미지 기반의 자동 교량 구성요소 분류 네트워크를 개발하고자 한다. 본 연구에서 제안한 방법은 두 개의 CNN 네트워크로 구성되었다. 첫 번째 네트워크에서 특정 교량 이미지에 대하여 교량의 형식을 자동으로 분류한 뒤, 두 번째 네트워크에서 교량의 형식별로 구성요소를 분류함으로써 정확도와 효율성을 향상시키고자 한다. 본 연구에서 개발한 시스템을 검증한 결과, 847개의 교량 이미지에 대해서 98.1 %의 정확도로 교량의 구성요소를 자동으로 분류 할 수 있었다. 본 연구에서 개발한 교량의 구성요소 자동분류 기술은 향후 교량의 유지관리에 기여를 할 수 있을 것으로 기대된다.

Keywords

Acknowledgement

이 논문은 충북대학교 국립대학육성사업(2020)지원을 받아 작성되었습니다.

References

  1. Calvi, G. M., Moratti, M., O'Reilly, G. J., Scattarreggia. N., Monteiro, R., Malomo, D., Calvi, P. M. and Pinho, R. (2019). "Once upon a time in Italy: The tale of the Morandi Bridge." Structural Engineering International, Vol. 29, No. 2, pp. 198-217. https://doi.org/10.1080/10168664.2018.1558033
  2. He, K., Zhang, X., Ren, S. and Sun, J. (2016). "Deep residual learning for image recognition." In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, CVPR, San Juan, Puerto Rico, USA, pp. 770-778.
  3. Jeong, Y. S., Kim, W. S., Lee, I. K. and Lee, J. H. (2016). "Development of bridge inspection reliability and improvement strategy." Journal of the Korea Institute for Structural Maintenance and Inspection, Vol. 20, No. 5, pp. 50-57 (in Korean). https://doi.org/10.11112/JKSMI.2016.20.5.050
  4. Jung, H. S., Lee, M. J., Yoo, M. T. and Lee, I. W. (2020). "Response dominant frequency analysis for scour safety evaluation of railroad piers." Journal of the Korean Geotechnical Society, Vol. 36, No. 11, pp. 83-95 (in Korean). https://doi.org/10.7843/KGS.2020.36.11.83
  5. Kang, J. O. and Lee, Y. C. (2016). "Preliminary research for drone based visual-safety inspection of bridge." Proceedings of Korean Society for Geospatial Information Science, pp. 207-210 (in Korean).
  6. Krizhevsky, A., Sutskever, I. and Hinton, G. E. (2012). "Imagenet classification with deep convolutional neural networks." In Advances in Neural Information Processing Systems, Lake Tahoe, Nevada, USA, pp. 1097-1105.
  7. LeCun, Y., Bottou, L., Bengio, Y. and Haffner, P. (1998). "Gradient-based learning applied to document recognition." Proceedings of the IEEE, Vol. 86, No. 11, pp. 2278-2324. https://doi.org/10.1109/5.726791
  8. Lee, G. (2016). "10 Years of BIM application in Korea in interior design." Korean Institute of Interior Design, pp. 17-20 (in Korean).
  9. Lee, J. H., Park, J. J. and Yoon, H. C. (2020). "Automatic classification of bridge component based on deep learning." Journal of the Korean Society of Civil Engineers, KSCE, Vol. 40, No. 2, pp. 239-245 (in Korean). https://doi.org/10.12652/Ksce.2020.40.2.0239
  10. Lee, T. S., Lee, J. S., Koo, J. K. and Hwang I. H. (2006). "A fundamental study on the feasibility analysis of the introduction of automated bridge inspection equipment." Proceedings of the Journal of the Korean Society of Civil Engineers, KSCE, pp. 3677-3680 (in Korean).
  11. Lim, S. C., Lee, J. H., Han, S. W. and Byun, G. S. (2017). "Study on the Method of Inspection of railway bridge of Busan urban railway using drone." Proceedings of The Korean Institute of Electrical Engineers, pp. 139-140 (in Korean).
  12. Meng, X., Nguyen, D. T., Owen, J. S., Xie, Y., Psimoulis, P. and Ye, G. (2019). "Application of GeoSHM system in monitoring extreme wind events at the forth road bridge." Remote Sensing, Vol. 11, No. 23, pp. 2799. https://doi.org/10.3390/rs11232799
  13. National Assembly Reseach Service (NARS) (2019). Road maintenance status and project, No. r2019-37 (in Korean).
  14. Oh, D. G. (2011). A study on the bim utilization of building maintenance management through case study, Ph.D. Thesis, Hanyang University (in Korean).
  15. Park, T. S. and Park, H. S. (2018). "A study on the bim status and activation of civil engineering works." Journal of the Korean Society of Civil Engineers, KSCE, Vol. 38, No. 1, pp. 133-140 (in Korean). https://doi.org/10.12652/Ksce.2018.38.1.0133
  16. Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V. and Rabinovich, A. (2015). "Going deeper with convolutions." In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, CVPR, SanJuan, Puerto Rico, USA, pp. 1-9.
  17. Yoon, H. C., Elanwar, H., Choi, H. J., Golparvar-Fard, M. and Spencer Jr, B. F. (2016). "Target-free approach for vision-based structural system identification using consumer-grade cameras." Structural Control and Health Monitoring, Vol. 23, No. 12, pp. 1405-1416. https://doi.org/10.1002/stc.1850