ABSTRACT
Timely and accurate detection of the initiation and expansion of crack is of great significance for improving safe operation of civil infrastructures. Image-based visual surface inspection has been an indispensable way for long-time infrastructure monitoring. However, existing crack detection methods generally suffer from the interference of complex background, leading to obvious performance drops. To tackle this, an improved encoder-decoder architecture based on SegNet is proposed in this paper, namely crack-SegNet. The encoder network hierarchically learns visual features from the original image, and the decoder network gradually up-samples and maps the encoded features to the input size for the pixel-level classification. In order to enhance the feature capacity of cracks in complex background, a channel attention mechanism is integrated into the encoder, as well as a spatial attention module in the decoder to improve the feature representation of cracks. Meanwhile, a spatial pyramid pooling is also attached to the last convolutional layer of the encoder to capture crack with different scales. To better validate the proposed method, a challenging metal surface crack dataset with much more complex background is collected. Experimental results on the datasets show that the proposed crack-SegNet outperforms other state-of-the-art crack detection methods, especially in complex background.
- Liu, K., Wu, S. J., Gao, X. Y. and Yang, L. X. Simultaneous Measurement of In-Plane and Out-of-Plane Deformations Using Dual-Beam Spatial-Carrier Digital Speckle Pattern Interferometry. Trans Tech Publications Ltd, City, 2015.Google ScholarCross Ref
- Michalcová, L. and Růžek, R. Fatigue test of an integrally stiffened panel: Prediction and crack growth monitoring using acoustic emission. Procedia Structural Integrity, 2 (1/1/2016 2016), 3049-3056.Google ScholarCross Ref
- Priewald, R. H., Magele, C., Ledger, P. D., Pearson, N. R. and Mason, J. S. D. Fast Magnetic Flux Leakage Signal Inversion for the Reconstruction of Arbitrary Defect Profiles in Steel Using Finite Elements. IEEE Transactions on Magnetics, 49, 1 (Jan 2012), 506-516.Google Scholar
- Peng, L., Chao, W., Shuangmiao, L. and Baocai, F. Research on Crack Detection Method of Airport Runway Based on Twice-Threshold Segmentation. City, 2015.Google Scholar
- Xu, H., Tian, Y., Lin, S. and Wang, S. Research of image segmentation algorithm applied to concrete bridge cracks. IEEE, City, 2013.Google ScholarCross Ref
- Luo, Q. and He, Y. A cost-effective and automatic surface defect inspection system for hot-rolled flat steel. Robotics and Computer-Integrated Manufacturing, 38 (2016), 16-30.Google ScholarDigital Library
- Cortes, C. and Vapnik, V. Support-vector networks. Machine learning, 20, 3 (1995/09/01 1995), 273-297.Google Scholar
- Kaseko, M. S. and Ritchie, S. G. A neural network-based methodology for pavement crack detection and classification. Transportation Research Part C: Emerging Technologies, 1, 4 (1993), 275–291.Google ScholarCross Ref
- Huyan, J., Li, W., Tighe, S., Deng, R. and Yan, S. Illumination Compensation Model with k -Means Algorithm for Detection of Pavement Surface Cracks with Shadow. Journal of Computing in Civil Engineering, 34, 1 (01/01 2020), 04019049.Google ScholarCross Ref
- Guan, H., Li, J., Yu, Y., Chapman, M., Wang, H., Wang, C. and Zhai, R. Iterative tensor voting for pavement crack extraction using mobile laser scanning data. IEEE Transactions on Geoscience and Remote Sensing, 53, 3 (2014), 1527-1537.Google Scholar
- Shi, Y., Cui, L., Qi, Z., Meng, F. and Chen, Z. Automatic road crack detection using random structured forests. IEEE Transactions on Intelligent Transportation Systems, 17, 12 (2016), 3434-3445.Google ScholarDigital Library
- Li, S. and Zhao, X. Convolutional neural networks-based crack detection for real concrete surface. International Society for Optics and Photonics, City, 2018.Google Scholar
- Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V. and Rabinovich, A. Going deeper with convolutions. City, 2015.Google Scholar
- Qu, Z., Mei, J., Liu, L. and Zhou, D. Crack detection of concrete pavement with cross-entropy loss function and improved VGG16 network model. IEEE Access, 8 (2020), 54564-54573.Google ScholarCross Ref
- Simonyan, K. and Zisserman, A. Very Deep Convolutional Networks for Large-Scale Image Recognition. arXiv preprint arXiv: 1409.1556 (2014).Google Scholar
- Fan, Z., Wu, Y., Lu, J. and Li, W. Automatic pavement crack detection based on structured prediction with the convolutional neural network. arXiv preprint arXiv:1802.02208 (2018).Google Scholar
- Suh, G. and Cha, Y. J. Deep faster R-CNN-based automated detection and localization of multiple types of damage. International Society for Optics and Photonics, City, 2018.Google ScholarCross Ref
- Ren, S., He, K., Girshick, R. and Sun, J. Faster r-cnn: Towards real-time object detection with region proposal networks. IEEE Transactions on Pattern Analysis and Machine Intelligence, 39, 6 (2016), 1137-1149.Google Scholar
- Zeiler, M. D. and Fergus, R. Visualizing and understanding convolutional networks. Springer, City, 2014.Google ScholarCross Ref
- Cheng, J., Xiong, W., Chen, W., Gu, Y. and Li, Y. Pixel-level crack detection using U-Net. IEEE, City, 2018.Google ScholarCross Ref
- Escalona, U., Arce, F., Zamora, E. and Azuela, J. H. S. Fully convolutional networks for automatic pavement crack segmentation. Computación y Sistemas, 23, 2 (6/27/2019 2019), 451-460.Google Scholar
- Chen, T., Cai, Z., Zhao, X., Chen, C., Liang, X., Zou, T. and Wang, P. Pavement crack detection and recognition using the architecture of segNet. Journal of Industrial Information Integration, 18 (Google Scholar
- Jenkins, M. D., Carr, T. A., Iglesias, M. I., Buggy, T. and Morison, G. A deep convolutional neural network for semantic pixel-wise segmentation of road and pavement surface cracks. IEEE, City, 2018.Google ScholarCross Ref
- Ronneberger, O., Fischer, P. and Brox, T. U-net: convolutional networks for biomedical image segmentation. Springer, City, 2015.Google Scholar
- Badrinarayanan, V., Kendall, A. and Cipolla, R. SegNet: A deep convolutional encoder-decoder architecture for image segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 39, 12 (2017 2017), 2481-2495.Google ScholarCross Ref
- Long, J., Shelhamer, E. and Darrell, T. Fully convolutional networks for semantic segmentation. City, 2015.Google Scholar
- Chen, L.-C., Papandreou, G., Kokkinos, I., Murphy, K. and Yuille, A. L. Deeplab: Semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected crfs. IEEE transactions on pattern analysis and machine intelligence, 40, 4 (2017), 834-848.Google Scholar
- He, K., Zhang, X., Ren, S. and Sun, J. Spatial pyramid pooling in deep convolutional networks for visual recognition. IEEE transactions on pattern analysis and machine intelligence, 37, 9 (2015), 1904-1916.Google ScholarDigital Library
- Yu, F. and Koltun, V. Multi-scale context aggregation by dilated convolutions. arXiv preprint arXiv:1511.07122 (2015).Google Scholar
- Zhang, L., Yang, F., Zhang, Y. D. and Zhu, Y. J. Road crack detection using deep convolutional neural network. In Proceedings of the International Conference on Image Processing (8/19/2016, 2016).Google ScholarCross Ref
Recommendations
Numerical analysis of the creep crack constraint effects and the creep crack initiation for pressurized pipelines with circumferential surface cracks
The numerical analysis of the pressurized pipelines with circumferential surface cracks is proposed.The contour integral C* and load-independent constraint parameter Q* along the crack front are investigated.The creep damage accumulations and the CCI ...
Crack modelling: A novel technique for the prediction of fatigue failure in the presence of stress concentrations
Finite element (FE) analysis and other computational methods have developed rapidly in recent years, allowing accurate predictions of elastic stresses in components of complex geometry. However, the prediction of fatigue failure in these components is ...
Comments