Cracks are the most common evaluation indicators in structures such as road surfaces and concrete. However, cracks often exhibit complex interference, fuzzy boundaries, and discontinuous topological structures, making accurate crack segmentation tasks extremely difficult. Therefore, a Triple-Component Dynamic adjacency Graph attention network (TCDGNet) was constructed for automatic crack detection in structures such as concrete and pavement. To achieve adaptive modeling of channel dependence, a three component dynamic adjacency graph channel attention (TDA-GCA) mechanism is proposed. In addition, the model integrates two-stage dilated wavelet decomposition with gated feature modulation to enhance the structural edge response while suppressing background interference. Our model was tested on four public benchmarks (DeepCrack, CrackTree260, CrackForest and Crack500), it got mIoU scores of 87.50%, 85.15%, 80.85% and 76.60% on these benchmarks. Respectively, surpassing the ten existing methods. This improvement enhances the structural consistency of cracks and provides a more reliable assessment of the condition of infrastructure and other structures.
ChaYJChoiWBüyüköztürkO (2017) Deep learning-based crack damage detection using convolutional neural networks. Computer-Aided Civil and Infrastructure Engineering32(5): 361–378. https://doi.org/10.1111/mice.12263
2.
ChenLCPapandreouGKokkinosI, et al. (2017) Deeplab: semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected crfs. IEEE Transactions on Pattern Analysis and Machine Intelligence40(4): 834–848. https://doi.org/10.1109/TPAMI.2017.2699184
3.
ChenLCZhuYPapandreouG, et al. (2018) Encoder-decoder with atrous separable convolution for semantic image segmentation In: Proceedings of the European Conference on Computer Vision (ECCV). Springer, pp. 801–818.
GuoDShaoYCuiY, et al. (2021) Graph attention tracking In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. IEEE Computer Society, pp. 9543–9552.
6.
GuoBLiXLiD (2025) Crackwave r-convolutional neural network: a discrete wavelet transform and deep learning fusion model for underwater dam crack detection. Structural Health Monitoring25: 14759217241308132.
7.
HanKWangYChenH, et al. (2022) A survey on vision transformer. IEEE Transactions on Pattern Analysis and Machine Intelligence45(1): 87–110. https://doi.org/10.1109/TPAMI.2022.3152247
8.
HouQZhouDFengJ (2021) Coordinate attention for efficient Mobile network design In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. IEEE Computer Society, pp. 13713–13722.
9.
HuJShenLSunG (2018) Squeeze-and-excitation networksProceedings of the IEEE Conference on Computer Vision and Pattern Recognition. IEEE Computer Society, pp. 7132–7141.
10.
HuZDangCWangD, et al. (2025) Error-informed parallel adaptive kriging method for time-dependent reliability analysis. Reliability Engineering & System Safety262: 111194. https://doi.org/10.1016/j.ress.2025.111194
11.
KangDHChaYJ (2022) Efficient attention-based deep encoder and decoder for automatic crack segmentation. Structural Health Monitoring21(5): 2190–2205. https://doi.org/10.1177/14759217211053776
12.
KangDBenipalSSGopalDL, et al. (2020) Hybrid pixel-level concrete crack segmentation and quantification across complex backgrounds using deep learning. Automation in Construction118: 103291. https://doi.org/10.1016/j.autcon.2020.103291
LiHZhangHZhuH, et al. (2024) Automatic crack detection on concrete and asphalt surfaces using semantic segmentation network with hierarchical transformer. Engineering Structures307: 117903.
15.
LiYWuSQiaoJ, et al. (2026) Bending performance analysis and crack width calculation of ultra-high-performance concrete beams. Advances in Structural Engineering29(2): 279–293. https://doi.org/10.1177/13694332251353616
16.
LiuMDanJLuZ, et al. (2024) Cm-unet: hybrid cnn-mamba unet for remote sensing image semantic segmentation. arXiv preprint arXiv:2405.10530.
17.
LiuBKangJGuanH, et al. (2025a) Rtcnet: a novel real-time triple branch network for pavement crack semantic segmentation. International Journal of Applied Earth Observation and Geoinformation136: 104347. https://doi.org/10.1016/j.jag.2024.104347
18.
LiuKRenTLanZ, et al. (2025b) Cgv-net: tunnel lining crack segmentation method based on graph convolution guided transformer. Buildings15(2): 197. https://doi.org/10.3390/buildings15020197
MaYZhangBHuangK, et al. (2025) Probabilistic prediction and early warning for bridge bearing displacement using sparse variational gaussian process regression. Structural Safety114: 102564. https://doi.org/10.1016/j.strusafe.2024.102564
21.
O’sheaKNashR (2015) An Introduction to Convolutional Neural Networks. arXiv preprint arXiv:1511.08458.
22.
RonnebergerOFischerPBroxT (2015) U-net: convolutional networks for biomedical image segmentation In: Medical Image Computing and computer-assisted intervention–MICCAI 2015: 18Th International Conference, Munich, Germany, October 5-9, 2015, Proceedings, Part III 18. Springer, pp. 234–241.
23.
SalehinejadHSankarSBarfettJ, et al. (2017) Recent advances in recurrent neural networks. arXiv preprint arXiv:1801.01078.
24.
ShaoYLiLLiJ, et al. (2025) Advancing crack detection with generative ai for structural health monitoring. Structural Health Monitoring. Available at: https://doi.org/10.1177/14759217251369000
25.
SifuzzamanMIslamMRAliMZ (2009) Application of Wavelet Transform and its Advantages Compared to Fourier Transform. Journal of Physical Sciences13: 121–134. Available at: https://ir.vidyasagar.ac.in/handle/123456789/779.
26.
SunRLiXZhangL, et al. (2025) Wavelet-integrated deep neural network for deblurring and segmentation of crack images. Mechanical Systems and Signal Processing224: 112240. https://doi.org/10.1016/j.ymssp.2024.112240
27.
TakikawaTAcunaDJampaniV, et al. (2019) Gated-scnn: gated shape cnns for semantic segmentation In: Proceedings of the IEEE/CVF International Conference on Computer Vision. IEEE Computer Society, pp. 5229–5238.
TangXEshkevariSSChenH, et al. (2022) Golfer: Trajectory Prediction with Masked Goal Conditioning Mnm Network. arXiv preprint arXiv:2207.00738.
30.
VeličkovićPCucurullGCasanovaA, et al. (2017) Graph attention networks. arXiv preprint arXiv:1710.10903.
31.
WangHWangYZhangQ, et al. (2017) Gated convolutional neural network for semantic segmentation in high-resolution images. Remote Sensing9(5): 446. https://doi.org/10.3390/rs9050446
32.
WangYNiuMSongK, et al. (2023) Normal-knowledge-based pavement defect segmentation using relevance-aware and cross-reasoning mechanisms. IEEE Transactions on Intelligent Transportation Systems24(4): 4413–4427. https://doi.org/10.1109/tits.2023.3234330
33.
WangJZengZSharmaPK, et al. (2024a) Dual-path network combining cnn and transformer for pavement crack segmentation. Automation in Construction158: 105217. https://doi.org/10.1016/j.autcon.2023.105217
34.
WangRShaoYLiQ, et al. (2024b) A novel transformer-based semantic segmentation framework for structural condition assessment. Structural Health Monitoring23(2): 1170–1183. https://doi.org/10.1177/14759217231182303
35.
WangJWangJLiY, et al. (2026) Automatic crack segmentation network based on large kernel pooling transformer. Advances in Structural Engineering13694332261420437. Available at: https://doi.org/10.1177/13694332261420437
36.
WooSParkJLeeJY, et al. (2018) Cbam: convolutional block attention module In: Proceedings of the European Conference on Computer Vision (ECCV). Springer, pp. 3–19.
37.
XiangCGuoJCaoR, et al. (2023a) A crack-segmentation algorithm fusing transformers and convolutional neural networks for complex detection scenarios. Automation in Construction152: 104894. https://doi.org/10.1016/j.autcon.2023.104894
38.
XiangXWangZZhangJ, et al. (2023b) Agca: an adaptive graph channel attention module for steel surface defect detection. IEEE Transactions on Instrumentation and Measurement72: 1–12. https://doi.org/10.1109/tim.2023.3248111
39.
YangZFarsiuS (2023) Directional connectivity-based segmentation of medical images In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. IEEE, pp. 11525–11535.
40.
YuYSiXHuC, et al. (2019) A review of recurrent neural networks: lstm cells and network architectures. Neural Computation31(7): 1235–1270. https://doi.org/10.1162/neco_a_01199
41.
YuZHuangCLiH (2026) Cnn-transformer hybrid network for automated segmentation of multiple defects in concrete bridges. Advances in Structural Engineering29(1): 43–59. https://doi.org/10.1177/13694332251345935
42.
ZhangD (2019) Wavelet transform In: Fundamentals of Image Data Mining: Analysis, Features, Classification and Retrieval. Springer, pp. 35–44.
43.
ZhangLChenMArnabA, et al. (2022) Dynamic graph message passing networks for visual recognition. arXiv preprint arXiv:2209.09760.
44.
ZhangJZengZSharmaPK, et al. (2024) A dual encoder crack segmentation network with haar wavelet-based high–low frequency attention. Expert Systems with Applications256: 124950. https://doi.org/10.1016/j.eswa.2024.124950
45.
ZhangJLiDZengZ, et al. (2025a) Dual-branch crack segmentation network with multi-shape kernel based on convolutional neural network and mamba. Engineering Applications of Artificial Intelligence150: 110536. https://doi.org/10.1016/j.engappai.2025.110536
46.
ZhangJZhangSLiD, et al. (2025b) Crack segmentation network via difference convolution-based encoder and hybrid cnn-mamba multi-scale attention. Pattern Recognition167: 111723. https://doi.org/10.1016/j.patcog.2025.111723
47.
ZhouQQuZWangSY, et al. (2022) A method of potentially promising network for crack detection with enhanced convolution and dynamic feature fusion. IEEE Transactions on Intelligent Transportation Systems23(10): 18736–18745. https://doi.org/10.1109/tits.2022.3154746
48.
ZouQZhangZLiQ, et al. (2018) Deepcrack: learning hierarchical convolutional features for crack detection. IEEE Transactions on Image Processing28(3): 1498–1512. https://doi.org/10.1109/tip.2018.2878966
