All Issue

2024 Vol.26, Issue 6 Preview Page

Research Paper

Crack detection performance of CNN models on motion blur images

모션블러 이미지에 대한 CNN 모델의 균열 검출 성능

Chulhee Lee1
,
Dongku Kim1
,
Donggyou Kim2*
이 철희1
,
김 동구1
,
김 동규2*
1Senior Researcher, Dept. of Geotechnical Engineering Research, Korea Institute of Civil Engineering and Building Technology
2Senior Research Fellow, Dept. of Geotechnical Engineering Research, Korea Institute of Civil Engineering and Building Technology
1정회원, 한국건설기술연구원 지반연구본부 수석연구원
2정회원, 한국건설기술연구원 지반연구본부 선임연구위원
*Corresponding Author
30 November 2024. pp. 777-792
PDF XML Citation
Abstract

In this study, we analyzed the effect of motion blur on images used for detecting cracks in concrete tunnel linings on the performance of CNN models. Motion-blurred images with intensities ranging from 10 to 50 were generated on the Kaggle and KICT datasets. A semantic segmentation model with ResNet 18, ResNet 34, VGG 11, and AlexNet as backbones for feature extraction was employed, all pre-trained on the U-Net architecture. The performance of these models in crack detection was then assessed. It was observed that detection accuracy decreased across all models as the intensity of motion blur increased for each dataset. Within the same model, the F1-score on the KICT dataset showed over 20% higher performance than on the Kaggle dataset. This study demonstrates that CNN-based crack detection performance is affected by the quality of the image data and that the crack detection accuracy of CNN models can vary depending on the quality of the dataset used in training.

Keywords
CNN (convolutional neural network)
Motion blur
Dataset
Crack detection
F1 score

본 연구에서는 터널 콘크리트 라이닝에 발생하는 균열을 검출하는데 활용되는 이미지에 발생하는 모션블러가 CNN 모델들의 성능에 미치는 영향을 분석하였다. Kaggle과 KICT 데이터셋에 10~50 범위의 강도를 갖는 모션블러 이미지를 생성하고, U-Net 아키텍처에 사전 학습된 ResNet 18, ResNet 34, VGG 11 및 AlexNet를 특징추출을 위한 백본으로 사용한 의미론적 분할 모델을 활용하여 균열 검출 성능을 분석하였다. 각각의 데이터셋에 모션블러가 증가함에 따라 모든 모델에서 검출 정확도가 감소하는 것으로 나타났다. 동일한 학습 모델에서 KICT 데이터셋의 F1-score가 Kaggle 데이터셋보다 20% 이상 높은 성능을 갖는 것으로 평가되었다. 본 연구의 결과로, CNN기반 균열 검출 성능은 이미지 데이터의 품질에 따라 차이가 발생하며, 기 구축된 학습 데이터셋의 품질에 따라서도 CNN 모델의 균열 검출 성능이 상이할 수 있는 것으로 판단된다.

키워드
CNN (convolutional neural network)
모션블러
데이터셋
균열 검출
F1 score
References
1

Ali, L., Alnajjar, F., Jassmi, H.A., Gocho, M., Khan, W., Serhani, M.A. (2021), "Performance evaluation of deep CNN-based crack detection and localization techniques for concrete structures", Sensors, Vol. 21, No. 5, 1688.

10.3390/s2105168833804490PMC7957757
2

Alidoost, F., Austen, G., Hahn, M. (2022), "A multi-camera mobile system for tunnel inspection", iCity. Transformative Research for the Livable, Intelligent, and Sustainable City, Springer, pp. 211-224.

10.1007/978-3-030-92096-8_13
3

Badrinarayanan, V., Kendall, A., Cipolla, R. (2017), "SegNet: a deep convolutional encoder-decoder architecture for image segmentation", IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 39, No. 12, pp. 2481-2495.

10.1109/TPAMI.2016.264461528060704
4

Bae, H., Jang, K., An, Y.K. (2021), "Deep super resolution crack network (SrcNet) for improving computer vision-based automated crack detectability in in situ bridges", Structural Health Monitoring, Vol. 20, No. 4, pp. 1428-1442.

10.1177/1475921720917227
5

Balaguer, C., Montero, R., Victores, J.G., Martínez, S., Jardón, A. (2014), "Towards fully automated tunnel inspection: a survey and future trends", Proceedings of the 31st ISARC, Sydney, Australia, pp. 19-33.

10.22260/ISARC2014/0005
6

Chen, L.C., Zhu, Y., Papandreou, G., Schroff, F., Adam, H. (2018), "Encoder-decoder with atrous separable convolution for semantic image segmentation", Proceedings of the European Conference on Computer Vision (ECCV), Munich, Germany, pp. 801-818.

10.1007/978-3-030-01234-2_49
7

Dynamic UNet, Unet model using PixelShuffle ICNR upsampling that can be built on top of any pretrained architecture, https://docs.fast.ai/vision.models.unet.html (November 1, 2024).

8

Fastai, https://docs.fast.ai/ (September 1, 2024).

9

Guo, J., Liu, P., Xiao, B., Deng, L., Wang, Q. (2024), "Surface defect detection of civil structures using images: Review from data perspective", Automation in Construction, Vol. 158, 105186.

10.1016/j.autcon.2023.105186
10

Guo, J., Wang, Q., Li, Y., Liu, P. (2020), "Façade defects classification from imbalanced dataset using meta learning-based convolutional neural network", Computer-Aided Civil and Infrastructure Engineering, Vol. 35, No. 12, pp. 1403-1418.

10.1111/mice.12578
11

He, K., Zhang, X., Ren, S., Sun, J. (2016), "Deep residual learning for image recognition", Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA, pp. 770-778.

10.1109/CVPR.2016.90
12

Kaggle, Crack segmentation dataset, https://www.kaggle.com/datasets/lakshaymiddha/crack-segmentation-dataset (September 1, 2024).

13

Krizhevsky, A., Sutskever, I., Hinton, G.E. (2012), "Imagenet classification with deep convolutional neural networks", Proceedings of the Advances in Neural Information Processing Systems 25, NIPS, Stateline, pp. 1097-105.

14

Kumar, S.S., Wang, M., Abraham, D.M., Jahanshahi, M.R., Iseley, T., Cheng, J.C.P. (2020), "Deep learning-based automated detection of sewer defects in CCTV videos", Journal of Computing in Civil Engineering, Vol. 34, No. 1, 04019047.

10.1061/(ASCE)CP.1943-5487.0000866
15

Lee, C., Kim, D., Kim, D. (2024), "Analysis of the application of image quality assessment method for mobile tunnel scanning system", Journal of Korean Tunnelling and Underground Space Association, Vol. 26, No. 4, pp. 365-384.

10.9711/KTAJ.2024.26.4.365
16

Li, D., Xie, Q., Gong, X., Yu, Z., Xu, J., Sun, Y., Wang, J. (2021), "Automatic defect detection of metro tunnel surfaces using a vision-based inspection system", Advanced Engineering Informatics, Vol. 47, 101206.

10.1016/j.aei.2020.101206
17

Liu, Y., Yeoh, J.K.W., Chua, D.K.H. (2020), "Deep learning-based enhancement of motion blurred UAV concrete crack images", Journal of Computing in Civil Engineering, Vol. 34, No. 5, 04020028.

10.1061/(ASCE)CP.1943-5487.0000907
18

Liu, Z., Cao, Y., Wang, Y., Wang, W. (2019), "Computer vision-based concrete crack detection using U-Net fully convolutional networks", Automation in Construction, Vol. 104, pp. 129-139.

10.1016/j.autcon.2019.04.005
19

Long, J., Shelhamer, E., Darrell, T. (2015), "Fully convolutional networks for semantic segmentation", Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Boston, MA, USA, pp. 3431-3440.

10.1109/CVPR.2015.7298965
20

Mei, Q., Gül, M., Azim, M.R. (2020), "Densely connected deep neural network considering connectivity of pixels for automatic crack detection", Automation in Construction, Vol. 110, 103018.

10.1016/j.autcon.2019.103018
21

Miao, P., Srimahachota, T. (2021), "Cost-effective system for detection and quantification of concrete surface cracks by combination of convolutional neural network and image processing techniques", Construction and Building Materials, Vol. 293, 123549.

10.1016/j.conbuildmat.2021.123549
22

Ni, F., Zhang, J., Chen, Z. (2019), "Zernike-moment measurement of thin-crack width in images enabled by dual-scale deep learning", Computer-Aided Civil and Infrastructure Engineering, Vol. 34, No. 5, pp. 367-384.

10.1111/mice.12421
23

Pan, S.J., Yang, Q. (2010), "A survey on transfer learning", IEEE Transactions on Knowledge and Data Engineering, Vol. 22, No. 10, pp. 1345-1359.

10.1109/TKDE.2009.191
24

Redmon, J., Divvala, S., Girshick, R., Farhadi, A. (2016), "You only look once: unified, real-time object detection", Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA, pp. 779-788.

10.1109/CVPR.2016.91
25

Ren, S., He, K., Girshick, R., Sun, J. (2017), "Faster R-CNN: towards real-time object detection with region proposal networks", IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 39, No. 6, pp. 1137-1149.

10.1109/TPAMI.2016.257703127295650
26

Shin, H., Ahn, Y., Tae, S., Gil, H., Song, M., Lee, S. (2021), "Enhancement of multi-class structural defect recognition using generative adversarial network", Sustainability, Vol 13, No. 22, 12682.

10.3390/su132212682
27

Simonyan, K., Zisserman, A. (2015), "Very deep convolutional networks for large-scale image recognition", ArXiv, Vol. 1409, 1556.

10.48550/arXiv.1409.1556
28

Spencer Jr, B.F., Hoskere, V., Narazaki, Y. (2019), "Advances in computer vision-based civil infrastructure inspection and monitoring", Engineering, Vol. 5, No. 2, pp. 199-222.

10.1016/j.eng.2018.11.030
29

Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., Rabinovich, A. (2015), "Going deeper with convolutions", Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Boston, MA, USA, pp. 1-9.

10.1109/CVPR.2015.7298594
30

Szegedy, C., Vanhoucke, V., Ioffe, S., Shlens, J., Wojna, Z. (2016), "Rethinking the inception architecture for computer vision", Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA, pp. 2818-2826.

10.1109/CVPR.2016.308
31

Wang, X., Kong, T., Shen, C., Jiang, Y., Li, L. (2020), "SOLO: segmenting objects by locations", Proceedings of the Computer Vision - ECCV 2020: Part XVIII 16, Glasgow, UK, pp. 649-665.

10.1007/978-3-030-58523-5_38
32

Wei, F., Yao, G., Yang, Y., Sun, Y. (2019), "Instance-level recognition and quantification for concrete surface bughole based on deep learning", Automation in Construction, Vol 107, 102920.

10.1016/j.autcon.2019.102920
33

Xue, Y., Li, Y. (2018), "A fast detection method via region-based fully convolutional neural networks for shield tunnel lining defects", Computer-Aided Civil and Infrastructure Engineering, Vol. 33, No. 8, pp. 638-654.

10.1111/mice.12367
34

Yu, Y., Samali, B., Rashidi, M., Mohammadi, M., Nguyen, T.N., Zhang, G. (2022), "Vision-based concrete crack detection using a hybrid framework considering noise effect", Journal of Building Engineering, Vol. 61, 105246.

10.1016/j.jobe.2022.105246
Information
  • Publisher :Korean Tunneling and Underground Space Association
  • Publisher(Ko) :한국터널지하공간학회
  • Journal Title :Journal of Korean Tunnelling and Underground Space Association
  • Journal Title(Ko) :한국터널지하공간학회 논문집
  • Volume : 26
  • No :6
  • Pages :777-792
  • Received Date : 2024-11-01
  • Revised Date : 2024-11-20
  • Accepted Date : 2024-11-20
  • DOI :https://doi.org/10.9711/KTAJ.2024.26.6.777
Journal Informaiton Journal of Korean Tunnelling and Underground Space Association Journal of Korean Tunnelling and Underground Space Association
Online Submission & Review System
  • NRF
  • KOFST
  • KISTI Current Status
  • KISTI Cited-by
  • crosscheck
  • orcid
  • open access
  • ccl
Journal Informaiton Journal Informaiton - close