All Issue

2025 Vol.27, Issue 4 Preview Page

Research Paper

Performance verification of deep-learning based tunnel crack detection using no-reference image quality assessment

무참조 이미지 품질 평가를 활용한 딥러닝 기반 터널 균열 검출 성능 검증

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
31 July 2025. pp. 287-304
PDF XML Citation
Abstract

The safety assessment of tunnel structures heavily relies on the accuracy of crack detection on concrete linings. While deep learning (DL) based automation is being actively adopted, image quality degradation due to motion blur (MB) inherent to mobile tunnel scanning systems (MTSSs) that capture images at high speeds remains a primary factor undermining crack detection performance. To address this issue, this study proposed a method to improve the performance of a convolutional neural network (CNN)-based crack detection model by using no-reference image quality assessment (NR-IQA) to pre-filter low-quality images. An experimental dataset was constructed by applying directional MB of varying intensities to actual MTSS images. The quality of these images was then scored using four NR-IQA metrics (BRISQUE, NIQE, PIQE, and CPBD). Subsequently, the correlation and sensitivity between the NR-IQA scores and the crack detection performance (F1-score) of a Dynamic U-Net model were analyzed. The results confirmed the critical impact of quality degradation on performance, with the F1-score dropping sharply from 89.43% to 4.45% as MB intensity increased. Notably, the CPBD, PIQE, and NIQE metrics demonstrated a very high statistical correlation with the F1-score, proving their viability as reliable indicators for predicting model performance. This study experimentally verified that NR-IQA is an effective tool for quantitatively assessing the quality of MTSS images and, based on this assessment, systematically managing input data to enhance the stability and reliability of DL models.

Keywords
No-reference image quality assessment (NR-IQA)
Crack detection
Motion blur
Deep learning
Mobile tunnel scanning system

터널 구조물의 안전성 평가는 콘크리트 라이닝의 균열 검출 정확도에 크게 좌우된다. 최근 딥러닝(deep-learning, DL) 기반 자동화 기술이 활발히 도입되고 있으나, 고속으로 이동하며 촬영하는 이동식 터널 스캐닝 장비(mobile tunnel scanning systems, MTSSs)의 특성상 모션 블러(motion blur, MB)로 인한 이미지 품질 저하는 균열 검출 성능을 저해하는 주요 원인으로 작용한다. 본 연구는 이러한 문제를 해결하기 위해 무참조 이미지 품질 평가(no-reference image quality assessment, NR-IQA) 기법을 활용하여 저품질 이미지를 사전에 필터링함으로써 convolutional neural network (CNN) 기반 균열 검출 모델의 성능을 향상시키는 방안을 제시하였다. 실제 MTSS 이미지에 방향성을 갖는 MB를 단계적으로 적용한 데이터셋을 구축하고, 4종의 NR-IQA 지표(BRISQUE, NIQE, PIQE, CPBD)로 품질 점수를 산출하였다. 이후, Dynamic U-Net 모델의 균열 검출 성능(F1-score)과 NR-IQA 점수 간의 상관관계를 분석하였다. 실험 결과, MB 강도가 증가함에 따라 F1-score는 89.43%에서 4.45%로 급감하여 품질 저하가 성능에 미치는 치명적 영향을 확인하였다. 특히, CPBD, PIQE와 NIQE 지표는 F1-score와 매우 높은 통계적 상관관계 결과를 보여, 모델 성능을 예측하는 신뢰성 있는 지표로 활용될 수 있음을 입증하였다. 본 연구는 NR-IQA가 MTSS의 이미지 품질을 정량적으로 평가하고, 이를 기반으로 입력 데이터를 체계적으로 관리하여 CNN 모델의 안정성과 신뢰도를 향상시킬 수 있는 효과적인 도구임을 실험적으로 검증하였다.

키워드
무참조 이미지 품질 평가
균열 검출
모션블러
딥러닝
이동식 터널 스캐닝 장비
References
1

Abdullah-Al-Mamun, M., Tyagi, V., Zhao, H. (2021), “A new full-reference image quality metric for motion blur profile characterization”, IEEE Access, Vol. 9, pp. 156361-156371.

10.1109/ACCESS.2021.3130177
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, Cham, 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.2016.2644615
4

Bae, H., Jang, K., An, Y.K. (2020), “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

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 15th European Conference on Computer Vision (ECCV 2018), Munich, Germany, pp. 801-818.

10.1007/978-3-030-01234-2_49
6

De, K., Masilamani, V. (2013), “Image sharpness measure for blurred images in frequency domain”, Procedia Engineering, Vol. 64, pp. 149-158.

10.1016/j.proeng.2013.09.086
7

Dong, C., Loy, C.C., He, K., Tang, X. (2014), “Learning a deep convolutional network for image super-resolution”, Proceedings of the 13th European Conference on Computer Vision (ECCV 2014), Zurich, Switzerland, pp. 184-199.

10.1007/978-3-319-10593-2_13
8

Dong, Y., Wang, J., Wang, Z., Zhang, X., Gao, Y., Sui, Q., Jiang, P. (2019), “A deep-learning-based multiple defect detection method for tunnel lining damages”, IEEE Access, Vol. 7, pp. 182643-182657.

10.1109/ACCESS.2019.2931074
9

Ferwerda, J.A. (2003), “Three varieties of realism in computer graphics”, Proceedings of the Human Vision and Electronic Imaging VIII, Vol. 5007, Santa Clara, CA, USA, pp. 290-297.

10.1117/12.473899
10

Giniatullin, A., Dmitrii, T., Elshin, L. (2024), “On the efficiency of the BRISQUE metric for assessing linearly blurred images when deconvoluted with 3x3 convolution matrices”, Proceedings of the Third International Conference on Optics, Computer Applications, and Materials Science (CMSD-III 2023), Vol. 13065, Dushanbe, Tajikistan, 130650U-1.

10.1117/12.3025069
11

Golestaneh, S., Karam, L.J. (2016), “Reduced-reference quality assessment based on the entropy of DWT coefficients of locally weighted gradient magnitudes”, IEEE Transactions on Image Processing, Vol. 25, No. 11, pp. 5293-5303.

10.1109/TIP.2016.2601821
12

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
13

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, Las Vegas, NV, USA, pp. 770-778.

10.1109/CVPR.2016.90
14

Huang, H., Sun, Y., Xue, Y., Wang, F. (2017), “Inspection equipment study for subway tunnel defects by grey-scale image processing”, Advanced Engineering Informatics, Vol. 32, pp. 188-201.

10.1016/j.aei.2017.03.003
15

Krizhevsky, A., Sutskever, I., Hinton, G.E. (2017), “ImageNet classification with deep convolutional neural networks”, Communications of the ACM, Vol. 60, No. 6, pp. 84-90.

10.1145/3065386
16

Ledig, C., Theis, L., Huszár, F., Caballero, J., Cunningham, A., Acosta, A., Aitken, A., Tejani, A., Totz, J., Wang, Z., Shi, W. (2017), “Photo-realistic single image super-resolution using a generative adversarial network”, Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA, pp. 4681-4690.

10.1109/CVPR.2017.19
17

Lee, C.H., 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
18

Lee, G.P., Lim, H.J., Kim, J.H. (2020), “Availability evaluation of automatic inspection equipment using line scan camera for concrete lining”, Journal of Korean Tunnelling and Underground Space Association, Vol. 22, No. 6, pp. 643-653.

10.9711/KTAJ.2020.22.6.643
19

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
20

Lim, B., Son, S., Kim, H., Nah, S., Lee, K.M. (2017), “Enhanced deep residual networks for single image super-resolution”, Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), Honolulu, HI, USA, pp. 136-144.

10.1109/CVPRW.2017.151
21

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
22

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
23

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
24

Lu, Y., Xie, F., Liu, T., Jiang, Z., Tao, D. (2015), “No reference quality assessment for multiply-distorted images based on an improved bag-of-words model”, IEEE Signal Processing Letters, Vol. 22, No. 10, pp. 1811-1815.

10.1109/LSP.2015.2436908
25

Mittal, A., Moorthy, A.K., Bovik, A.C. (2012), “No-reference image quality assessment in the spatial domain”, IEEE Transactions on Image Processing, Vol. 21, No. 12, pp. 4695-4708.

10.1109/TIP.2012.2214050
26

Mittal, A., Soundararajan, R., Bovik, A.C. (2013), “Making a completely blind image quality analyzer”, IEEE Signal Processing Letters, Vol. 20, No. 3, pp. 209-212.

10.1109/LSP.2012.2227726
27

Narvekar, N.D., Karam, L.J. (2011), “A no-reference image blur metric based on the cumulative probability of blur detection (CPBD)”, IEEE Transactions on Image Processing, Vol. 20, No. 9, pp. 2678-2683.

10.1109/TIP.2011.2131660
28

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
29

Pennada, S., Perry, M., McAlorum, J., Dow, H., Dobie, G. (2023), “Threshold-based Brisque-assisted deep learning for enhancing crack detection in concrete structures”, Journal of Imaging, Vol. 9, No. 10, 218.

10.3390/jimaging910021837888325PMC10607118
30

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
31

Ren, S., He, K., Girshick, R., Sun, J. (2015), “Faster R-CNN: Towards real-time object detection with region proposal networks”, Proceedings of the Advances in Neural Information Processing Systems, Vol. 28, Montreal, Quebec, Canada, pp. 91-99.

32

Sheikh, H.R., Sabir, M.F., Bovik, A.C. (2006), “A statistical evaluation of recent full reference image quality assessment algorithms”, IEEE Transactions on Image Processing, Vol. 15, No. 11, pp. 3440-3451.

10.1109/TIP.2006.881959
33

Simonyan, K., Zisserman, A. (2014), “Very deep convolutional networks for large-scale image recognition”, arXiv preprint arXiv:1409.1556.

34

Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Dumitru, E., Vincent, V., Andrew, R., 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
35

Venkatanath, N., Praneeth, D., Bh, M.C., Channappayya, S.S., Medasani, S.S. (2015), “Blind image quality evaluation using perception based features”, Proceedings of the Twenty First National Conference on Communications, Mumbai, India, pp. 1-6.

10.1109/NCC.2015.7084843
36

Wang, Z., Bovik, A.C., Sheikh, H.R., Simoncelli, E.P. (2004), ‘Image quality assessment: From error visibility to structural similarity”, IEEE Transactions on Image Processing, Vol. 13, No. 4, pp. 600-612.

10.1109/TIP.2003.819861
37

Whang, S.E., Roh, Y., Song, H., Lee, J.G. (2023), “Data collection and quality challenges in deep learning: A data-centric ai perspective”, The VLDB Journal, Vol. 32, No. 4, pp. 791-813.

10.1007/s00778-022-00775-9
38

Wu, Q., Li, H., Meng, F., Ngan, K.N., Zhu, S. (2015), “No reference image quality assessment metric via multi-domain structural information and piecewise regression”, Journal of Visual Communication and Image Representation, Vol. 32, pp. 205-216.

10.1016/j.jvcir.2015.08.009
39

Yang, J., Marcus, D.S., Sotiras, A. (2025), “Dynamic u-net: Adaptively calibrate features for abdominal multi-organ segmentation”, Medical Imaging 2025: Computer-Aided Diagnosis, SPIE, 134071D.

10.1117/12.3046359
40

Yang, W., Zhang, X., Tian, Y., Wang, W., Xue, J.H., Liao, Q. (2019), “Deep learning for single image super-resolution: A brief review”, IEEE Transactions on Multimedia, Vol. 21, No. 12, pp. 3106-3121.

10.1109/TMM.2019.2919431
41

Yang, X., Li, H., Yu, Y., Luo, X., Huang, T., Yang, X. (2018), “Automatic pixel-level crack detection and measurement using fully convolutional network”, Computer-Aided Civil and Infrastructure Engineering, Vol. 33, No. 12, pp. 1090-1109.

10.1111/mice.12412
42

Zhang, Y., Li, K., Li, K., Wang, L., Zhong, B., Fu, Y. (2018), “Image super-resolution using very deep residual channel attention networks”, Proceedings of the 15th European Conference on Computer Vision (ECCV 2018), Munich, Germany, pp. 286-301.

10.1007/978-3-030-01234-2_18
43

Zhu, H., Li, L., Wu, J., Dong, W., Shi, G. (2020), “MetaIQA: Deep meta-learning for no-reference image quality assessment”, Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA, pp. 14143-14152.

10.1109/CVPR42600.2020.01415
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 : 27
  • No :4
  • Pages :287-304
  • Received Date : 2025-06-24
  • Revised Date : 2025-07-21
  • Accepted Date : 2025-07-22
  • DOI :https://doi.org/10.9711/KTAJ.2025.27.4.287
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