Journal of the Korean Geotechnical Society (한국지반공학회논문집)

Korean Geotechnical Society (한국지반공학회)
권호 표지
DOI QR코드

DOI QR Code

Application of Mask R-CNN Algorithm to Detect Cracks in Concrete Structure

콘크리트 구조체 균열 탐지에 대한 Mask R-CNN 알고리즘 적용성 평가

  • Bae, Byongkyu (Dept. of Civil and Environmental Engrg., Pusan National Univ.) ;
  • Choi, Yongjin (School of Civil and Environmental Engrg., Georgia Institute of Technology) ;
  • Yun, Kangho (Dept. of Civil Engrg., Pusan National Univ.) ;
  • Ahn, Jaehun (Dept. of Civil and Environmental Engrg., Pusan National Univ.)
  • 배병규 (부산대학교 사회환경시스템공학과 ) ;
  • 최용진 (조지아 공과대학 토목환경공학부 ) ;
  • 윤강호 (부산대학교 토목공학과 ) ;
  • 안재훈 (부산대학교 사회환경시스템공학과)
  • Received : 2024.04.09
  • Accepted : 2024.05.28
  • Published : 2024.06.30
Download PDF
⟨ Previous Next ⟩

Abstract

Inspecting cracks to determine a structure's condition is crucial for accurate safety diagnosis. However, visual crack inspection methods can be subjective and are dependent on field conditions, thereby resulting in low reliability. To address this issue, this study automates the detection of concrete cracks in image data using ResNet, FPN, and the Mask R-CNN components as the backbone, neck, and head of a convolutional neural network. The performance of the proposed model is analyzed using the intersection over the union (IoU). The experimental dataset contained 1,203 images divided into training (70%), validation (20%), and testing (10%) sets. The model achieved an IoU value of 95.83% for testing, and there were no cases where the crack was not detected. These findings demonstrate that the proposed model realized highly accurate detection of concrete cracks in image data.

구조물의 상태를 파악하기 위한 균열조사는 정밀안전 진단에 필수적인 검사 항목이다. 그러나 육안으로 이루어지는 균열조사 방식은 현장 상황의 변화에 따라 주관적으로 수행될 수 있다. 이러한 육안검사의 한계를 극복하기 위해 본 연구에서는, ResNet, FPN, Mask R-CNN을 백본(Backbone), 넥(Neck), 헤드(head)로 구성한 합성곱 신경망을 바탕으로, 이미지 데이터에서의 콘크리트 균열 탐지를 자동화하고. 그 성능을 IoU 값을 바탕으로 분석하였다. 해석에 사용된 데이터는 총 1,203개의 이미지 데이터로 구성하였으며, 이 중 70%를 훈련(Training)에, 20%를 검증(Validation)에, 그리고 10%의 데이터를 시험(Testing)에 사용하였다. 시험 결과의 평균 IoU값은 95.83%로 산정되었고, 또한 이미지 내 균열이 전혀 탐지되지 않는 경우는 존재하지 않아, 본 연구에 가정한 모델이 콘크리트의 균열 탐지를 성공적으로 수행하는 것을 확인하였다.

Keywords

Acknowledgement

이 과제는 부산대학교 기본연구지원사업(2년)에 의하여 연구되었음.

References

  1. Abdulla, W. (2018), Mask R-CNN for object detection and instance segmentation on Keras and TensorFlow (v2.1) [Computer software]. GitHub. https://github.com/matterport/Mask_RCNN/releases 
  2. Bouraya, S. and Belangour, A. (2021), "Deep Learning based Neck Models for Object Detection: A Review and a Benchmarking Study", International Journal of Advanced Computer Science and Applications, Vol.12, No.11, pp.161-167.  https://doi.org/10.14569/IJACSA.2021.0121119
  3. Cheng, B., Girshick, R., Dollar, P., Berg, A. C., and Kirillov, A. (2021), "Boundary IoU: Improving Object-centric Image Segmentation Evaluation", In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp.15334-15342). 
  4. Dung, C. V. (2019), "Autonomous Concrete Crack Detection Using Deep Fully Convolutional Neural Network", Automation in Construction, Vol.99, pp.52-58.  https://doi.org/10.1016/j.autcon.2018.11.028
  5. Everingham, M., Van Gool, L., Williams, C. K., Wi nn, J., and Zisserman, A. (2010), "The Pascal Visual Object Classes (voc) Challenge", International Journal of Computer Vision, Vol.88, pp. 303-338.  https://doi.org/10.1007/s11263-009-0275-4
  6. Golding, V. P., Gharineiat, Z., Munawar, H. S., and Ullah, F. (2022), "Crack Detection in Concrete Structures Using Deep Learning", Sustainability, Vol.14, No.13, p.8117. 
  7. He, K., Gkioxari, G., Dollar, P., and Girshick, R. (2017), "Mask r-cnn", In Proceedings of the IEEE International Conference on Computer Vision (pp. 2961-2969). 
  8. 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 (pp.770-778). 
  9. Jung, S. Y., Lee, S. K., Park, C. I., Cho, S. Y., and Yu, J. H. (2019), A method for detecting concrete cracks using deep-learning and image processing. Journal of the Architectural Institute of Korea Structure & Construction, Vol.35, No.11, pp.163-170. 
  10. KALIS (2022), Detailed Guidelines for Implementing Safety and Maintenance of Facilities (Safety Inspection and Diagnosis) Report, Korea. 
  11. Kim, A. R., Kim, D., Byun, Y. S., and Lee, S. W. (2018), "Crack Detection of Concrete Structure Using Deep Learning and Image Processing Method in Geotechnical Engineering", Journal of the Korean Geotechnical Society, Vol.34, No.12, pp.145-154.  https://doi.org/10.7843/KGS.2018.34.12.145
  12. Kim, A. R., Kim, D., Byun, Y. S., and Lee, S. W. (2018), "Crack Detection of Concrete Structure Using Deep Learning and Image Processing Method in Geotechnical Engineering", Journal of the Korean Geotechnical Society, Vol.34, No.12, pp.145-154.  https://doi.org/10.7843/KGS.2018.34.12.145
  13. Kim, J., Jang, A., Park, M. J., and Ju, Y. K. (2021), "Comparison Analysis of Machine Learning for Concrete Crack Depths Prediction Using Thermal Image and Environmental Parameters", Journal of Korean Association for Spatial Structures, Vol.21, No.2, pp.99-110.  https://doi.org/10.9712/KASS.2021.21.2.99
  14. 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
  15. Lin, T. Y., Dollar, P., Girshick, R., He, K., Hariharan, B., and Belongie, S. (2017), "Feature Pyramid Networks for Object Detection", In Proceedings of the IEEE conference on computer vision and pattern recognition (pp.2117-2125). 
  16. NARS (2015), Basic Research for the Development of Comprehensive Fiscal Statistics System for Infrastructure Report, Korea. 
  17. Ozgenel, C. F. (2018), "Concrete Crack Images for Classification", Mendeley Data, v1 http://dx.doi.org/10.17632/5y9wdsg2zt.1 
  18. Park, J., Kim, K. Y., and Yun, T. S. (2023), "Morphological Analysis of Hydraulically Stimulated Fractures by Deep-Learning Segmentation Method", Journal of the Korean Geotechnical Society, Vol.39, No.8, pp.17-28.  https://doi.org/10.7843/KGS.2023.39.8.17
  19. Simonyan, K. and Zisserman, A. (2014), Very deep convolutional networks for large-scale image recognition, arXiv preprint arXiv:1409.1556.