Skip to main content
SpringerLink
Log in

DcsNet: a real-time deep network for crack segmentation

  • Original Paper
  • Published:
Signal, Image and Video Processing Aims and scope Submit manuscript

Abstract

Detecting cracks are a great significance for the maintenance of the man-made buildings, and deep learning methods such as semantic segmentation have greatly boosted this process in recent years. However, the existing crack segmentation methods often sacrifice feature resolution to achieve real-time inference speed which leads to poor performance, or use complex network module to improve the accuracy which leads to lower inference speed. In this paper, we propose a novel Deep Crack Segmentation Network (DcsNet) that incorporates two feature extraction branches to achieve the balance of speed and accuracy. We first design a morphology branch (MB) to preserve the morphology information of scale invariance that consists of a lightweight convolution network, a pyramid pooling module (PPM), and an attention module (CSA). Meanwhile, a shallow detail branch (DB) with a small stride is constructed to supplement detailed information. Extensive experiments are conducted on five challenging datasets (Crack500, Deepcrack, Gaps384, Structure, and Damcrack), and the results demonstrated that the proposed network achieves a good trade-off between accuracy and inference speed and outperforms state-of-the-art methods.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9

Similar content being viewed by others

References

  1. Zheng, M.J., Lei, Z.J., Zhang, K.: Intelligent detection of building cracks based on deep learning. Image Vis. Comput. 103(11), 103987 (2020)

    Article  Google Scholar 

  2. Wu, C.F., Sun, K.K., Xu, Y.M., Zhang, S., Huang, X., Zeng, S.Q.: Concrete crack detection method based on optical fiber sensing network and microbending principle. Saf. Sci. 117(9), 299–304 (2019)

    Article  Google Scholar 

  3. Kim, B., Yuvaraj, N., Preethaa, K., Pandian, R.: Surface crack detection using deep learning with shallow CNN architecture for enhanced computation. Neural Comput. Appl. (2021). https://doi.org/10.1007/s00521-021-05690-8

    Article  Google Scholar 

  4. Fang, F., Li, L.Y., Gu, Y., Zhu, H.Y., Lim, J.H.: A novel hybrid approach for crack detection. Pattern Recognit. 107(11), 107474 (2021)

    Google Scholar 

  5. Guilherme, F.G., Yohan, A.D.M., Patrícia, D.S.L.A., Sebastião, S.D.C.J., Antonio, C.A.J.: The use of intelligent computational tools for damage detection and identification with an emphasis on composites–a review. Compos. Struct. 196(7), 44–54 (2018)

    Google Scholar 

  6. Juan, J.R., Takahiro, K., Teera, L., Wenlong, D., Kohei, N., Sergio, E., Kotaro, N., Yutaka, M., Helmut, P.: Multi-class structural damage segmentation using fully convolutional networks. Comput. Ind. 112(11), 103121 (2019)

    Google Scholar 

  7. Amir, R., Radhakrishna, A., Michele, G., Katrin, B.: Comparison of crack segmentation using digital image correlation measurements and deep learning. Constr. Build. Mater. 261(11), 120474 (2020)

    Google Scholar 

  8. Uche, A.N.: Fully adaptive segmentation of cracks on concrete surfaces. Comput. Electr. Eng. 83(5), 106561 (2020)

    Google Scholar 

  9. Zhou, S.L., Song, W.: Concrete roadway crack segmentation using encoder-decoder networks with range images. Automat. Constr. 120(12), 103403 (2020)

    Article  Google Scholar 

  10. Mohan, R., Abhinav, V.: EfficientPS: Efficient panoptic segmentation. (2020)

  11. Lin, D.Y., Li, Y.Q., Tin, L.N., Dong, S., Zaw, M.O.: RefineU-Net: improved U-Net with progressive global feedbacks and residual attention guided local refinement for medical image segmentation. Pattern Recogn. Lett. 138(10), 267–275 (2020)

    Article  Google Scholar 

  12. Sang, H.W., Zhou, Q.H., Zhao, Y.: PCANet: Pyramid convolutional attention network for semantic segmentation. Image Vis. Comput. 103(11), 103997 (2020)

    Article  Google Scholar 

  13. Adam, P., Abhishek, C., Sangpil, K., Eugenio, C.: ENet: A deep neural network architecture for real-time semantic segmentation. (2016)

  14. Si, H.Y., Zhang, Z.Q., Lv, F.F., Yu, G., Lu, F.: Real-time semantic segmentation via multiply spatial fusion network. (2019)

  15. Yang, F., Zhang, L., Yu, S.J., Danil, P., Mei, X., Ling, H.B.: Feature pyramid and hierarchical boosting network for pavement crack detection. IEEE Trans. Intell. Transp. Syst. 4(4), 1525–1535 (2020)

    Article  Google Scholar 

  16. Liu, Y.H., Yao, J., Lu, X.H., Xie, R.P., Li, L.: DeepCrack: a deep hierarchical feature learning architecture for crack segmentation. Neurocomputing 338, 139–153 (2019)

    Article  Google Scholar 

  17. Chen, F.C., Mohammad, R.J.: ARF-Crack: rotation invariant deep fully convolutional network for pixel-level crack detection. Mach. Vis. Appl. (2020). https://doi.org/10.1007/s00138-020-01098-x

    Article  Google Scholar 

  18. Bai, Y.S., Zha, B., Halil, S., Alper, Y.: Deep cascaded neural networks for automatic detection of structural damage and cracks from images. In: ISPRS2020, pp. 411–417 (2020)

  19. Christian, K., Kristina, D., Varun, K., Burcu, A.: A review on computer vision based defect detection and condition assessment of concrete and asphalt civil infrastructure. Adv. Eng. Inform. 29(2), 196–210 (2015)

    Article  Google Scholar 

  20. Cao, V.D., Le, D.A.: Autonomous concrete crack detection using deep fully convolutional neural network. Autom. Constr. 99(3), 52–58 (2018)

    Google Scholar 

  21. Wang, S., Wu, X., Zhang, Y.H., Liu, X.Q., Zhao, L.: A neural network ensemble method for effective crack segmentation using fully convolutional networks and multi-scale structured forests. Mach. Vis. Appl. (2020). https://doi.org/10.1007/s00138-020-01114-0

    Article  Google Scholar 

  22. Jacob, K., Mark, D.J., Mike, M., Peter, B., Gordon, M.: Optimized deep encoder-decoder methods for crack segmentation. Digit. Signal Process. 108, 102907 (2020)

    Google Scholar 

  23. Mei, Q.P., Mustafa, G., Md, R.A.: Densely connected deep neural network considering connectivity of pixels for automatic crack detection. Autom. Constr. 110(2), 10301 (2020)

    Google Scholar 

  24. Wooram, C., Young, J.C.: SDDNet: real-time crack segmentation. IEEE Trans. Industr. Electron. 67(9), 8016–8025 (2019)

    Google Scholar 

  25. Zhao, H.S., Qi, X.J., Shen, X.Y., Shi, J.P., Jia, J.Y.: ICNet for real-time semantic segmentation on high-resolution images. In: ECCV2018, pp. 418–434 (2018)

  26. Yu, C.Q., Wang, J.B., Peng, C., Gao, C.X., Yu, G., Sang, N.: BiSeNet: bilateral segmentation network for real-time semantic segmentation. In: ECCV2018, pp. 334–349 (2018)

  27. Peng, C., Zhang, X.Y., Yu, G., Luo, G.M., Sun, J.: Large kernel matters-improve semantic segmentation by global convolutional network. In: CVPR2017 (2017)

  28. Alexander, K., Ross, G., He, K.M., Piotr, D.: Panoptic feature pyramid networks. In: CVPR2019 (2019)

  29. Szegedy, C., Liu, W., Jia, Y.Q., Pierre, S., Scott, R., Dragomir, A., Dumitru, E., Vincent, V., Andrew, R.: Going deeper with convolutions. In: CVPR2015 (2015)

  30. He, K.M., Zhang, X.Y., Ren, S.Q., Sun, J.: Spatial pyramid pooling in deep convolutional networks for visual recognition. IEEE Trans. Pattern Anal. Mach. Intell. 37(9), 1904–1916 (2015)

    Article  Google Scholar 

  31. Xu, H.J., Gao, Y., Li, J., Gao, X.B.: CBFNet: constraint balance factor for semantic segmentation. Neurocomputing 397(15), 39–47 (2020)

    Article  Google Scholar 

  32. Sanghyun, W., Jongchan, P., Lee, J.Y., In, S.K.: CBAM: Convolutional block attention module. In: ECCV2018, pp. 3–19 (2018)

  33. Mo, J., Zhang, L.: Multi-level deep supervised networks for retinal vessel segmentation. Int. J. Comput. Assist. Radiol. Surg. 12(12), 2181–2193 (2017)

    Article  Google Scholar 

  34. Tsungyi, L., Priya, G., Ross, G., He, K.M., Piotr, D.: Focal loss for dense object detection. IEEE Trans. Pattern Anal. Mach. Intell. 42(2), 318–327 (2020)

    Article  Google Scholar 

  35. Li, X.Y., Sun, X.F., Meng, Y.X., Liang, J.J., Wu, F., Li, J.W.: Dice loss for data-imbalanced NLP tasks. ArXiv: 1911.02855 (2019)

  36. Pang, J., Zhang, H., Feng, C.C., Li, L.J.: Research on crack segmentation method of hydro-junction project based on target detection network. KSCE J. Civ. Eng. 24(7), 2731–2741 (2020)

    Article  Google Scholar 

  37. Vijay, B., Alex, K., Roberto, C.: SegNet: A deep convolutional encoder-decoder architecture for image segmentation. IEEE Trans. Pattern Anal. Mach. Intell. 39(12), 2481–2495 (2017)

    Article  Google Scholar 

  38. Jonathan, L., Evan, S., Trevor, D.: Fully convolutional networks for semantic segmentation. IEEE Trans. Pattern Anal. Mach. Intell. 39(4), 640–2495 (2017)

    Article  Google Scholar 

  39. Olaf, R., Philipp, F., Thomas, B.: U-Net: convolutional networks for biomedical image segmentation. In: MICCAI2015, pp. 234–241 (2015)

Download references

Author information

Authors and Affiliations

  1. School of Information Engineering, Southwest University of Science and Technology, Mianyang, 621000, China

    Jie Pang, Hua Zhang, Hao Zhao & Linjing Li

  2. Special Environment Robot Technology Key Laboratory of Sichuan Province, Mianyang, 621000, China

    Jie Pang, Hua Zhang, Hao Zhao & Linjing Li

  3. Department of Automation, University of Science and Technology of China, Hefei, 230000, China

    Hao Zhao

Authors
  1. Jie Pang
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Hua Zhang
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Hao Zhao
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Linjing Li
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Hao Zhao.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Pang, J., Zhang, H., Zhao, H. et al. DcsNet: a real-time deep network for crack segmentation. SIViP 16, 911–919 (2022). https://doi.org/10.1007/s11760-021-02034-w

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11760-021-02034-w

Keywords

Search

Navigation