Crack analysis of tall concrete wind towers using an ad-hoc deep multiscale encoder–decoder with depth separable convolutions under severely imbalanced data
Restricted accessResearch articleFirst published online November, 2025
Crack analysis of tall concrete wind towers using an ad-hoc deep multiscale encoder–decoder with depth separable convolutions under severely imbalanced data
An accurate and timely cracking assessment, including the presence, location and crack geometric feature measurement, is crucial for evaluating concrete wind towers. Therefore, the early identification of cracks is a critical procedure in promptly evaluating structural integrity. This study proposed an ad-hoc encoder–decoder network based on DeepLabv3+ with depth separable convolutions to automatically segment cracks from real-world images captured from various concrete wind towers. The combined advantages of the improved DeepLabv3+ and the lightweight MobileNet v2 are suitable as a benchmark due to their high performance and universality. Four experiments were conducted to determine the model design choice and crack feature measurement capability: (1) six parametric tests using various pre-trained base networks and algorithm optimisers, (2) the influence of complex background noise (i.e., handwriting script) on crack segmentation performance, (3) comparative studies with cutting-edge pixel-wise segmentation models and (4) crack feature measurement (i.e., length and width). The research outcome demonstrated that DeepLabv3+ with MobileNet v2 can potentially be applied for efficient and accurate crack segmentation in concrete wind towers with complex backgrounds.
AdhikariRSMoselhiOBagchiA. Image-based retrieval of concrete crack properties for bridge inspection. Autom Constr2014; 39: 180–194.
2.
KochCGeorgievaKKasireddyV, et al. A review on computer vision based defect detection and condition assessment of concrete and asphalt civil infrastructure. Adv Eng Inform2015; 29: 196–210.
3.
YangXLiHYuY, et al. Automatic pixel-level crack detection and measurement using fully convolutional network. Comput Civ Infrastruct Eng2018; 33: 1090–1109.
4.
KimHYoonJSimSH. Automated bridge component recognition from point clouds using deep learning. Struct Control Health Monit2020; 27: e2591.
5.
DengJSinghAZhouY, et al. Review on computer vision-based crack detection and quantification methodologies for civil structures. Constr Build Mater2022; 356: 129238.
6.
YuYHoshyarANSamaliB, et al. Corrosion and coating defect assessment of coal handling and preparation plants (CHPP) using an ensemble of deep convolutional neural networks and decision-level data fusion. Neural Comput Appl2023; 35: 18697–18718.
7.
YuZShenYShenC. A real-time detection approach for bridge cracks based on YOLOv4-FPM. Autom Constr2021; 122: 103514.
8.
KangDChaY. Autonomous UAVs for structural health monitoring using deep learning and an ultrasonic beacon system with geo-tagging. Comput Civ Infrastruct Eng2018; 33: 885–902.
9.
WaqasAKangDChaYJ. Deep learning-based obstacle-avoiding autonomous UAVs with fiducial marker-based localization for structural health monitoring. Struct Health Monit2024; 23: 971–990.
10.
HuangHSunYXueY, et al. Inspection equipment study for subway tunnel defects by grey-scale image processing. Adv Eng Inform2017; 32: 188–201.
11.
SunLQianZ. Multi-scale wavelet transform filtering of non-uniform pavement surface image background for automated pavement distress identification. Measurement2016; 86: 26–40.
12.
XuXZhangX. Crack detection of reinforced concrete bridge using video image. J Cent South Univ2013; 20: 2605–2613.
13.
JahanshahiMRMasriSF. A new methodology for non-contact accurate crack width measurement through photogrammetry for automated structural safety evaluation. Smart Mater Struct2013; 22: 035019.
14.
HuaLLuYDengJ, et al. 3D reconstruction of concrete defects using optical laser triangulation and modified spacetime analysis. Autom Constr2022; 142: 104469.
15.
JahanshahiMRMasriSFPadgettCW, et al. An innovative methodology for detection and quantification of cracks through incorporation of depth perception. Mach Vis Appl2013; 24: 227–241.
16.
ZhuZGermanSBrilakisI. Visual retrieval of concrete crack properties for automated post-earthquake structural safety evaluation. Autom Constr2011; 20: 874–883.
17.
ChenYLiangJGuX, et al. An improved minimal path selection approach with new strategies for pavement crack segmentation. Measurement2021; 184: 109877.
18.
YamaguchiTHashimotoS. Practical image measurement of crack width for real concrete structure. Electron Commun Japan2009; 92: 1–12.
19.
ZiouDTabboneS. Edge detection techniques-an overview. Pattern Recognit Image Anal Adv Math Theory Appl1998; 8: 537–559.
20.
ChaYJAliRLewisJ, et al. Deep learning-based structural health monitoring. Autom Constr2024; 161: 105328.
21.
HüthwohlPBrilakisI. Detecting healthy concrete surfaces. Adv Eng Informatics2018; 37: 150–162.
22.
SpencerBFHoskereVNarazakiY. Advances in computer vision-based civil infrastructure inspection and monitoring. Engineering2019; 5: 199–222.
23.
WangYLiuMZhengP, et al.. A smart surface inspection system using faster R-CNN in cloud-edge computing environment. Adv Eng Informs2020; 43: 101037.
24.
HuangH-wLiQ-tZhangD-m. Deep learning based image recognition for crack and leakage defects of metro shield tunnel. Tunn Undergr Space Technol2018; 77: 166–176.
25.
HoskereVNarazakiYHoangTA, et al. Vision-based structural inspection using multiscale deep convolutional neural networks, arXiv Prepr, arXiv1805.01055, 2018. https://doi.org/10.48550/arXiv.1805.01055.
26.
LiSZhaoXZhouG. Automatic pixel-level multiple damage detection of concrete structure using fully convolutional network. Comput Civ Infrastruct Eng2019; 34: 616–634.
27.
ChowJKSuZWuJ, et al. Anomaly detection of defects on concrete structures with the convolutional autoencoder. Adv Eng Inform2020; 45: 101105.
28.
XuYBaoYChenJ, et al. Surface fatigue crack identification in steel box girder of bridges by a deep fusion convolutional neural network based on consumer-grade camera images. Struct Health Monit2018; 18(12): 147592171876487.
29.
TongZGaoJZhangH. Recognition, location, measurement. and 3D reconstruction of concealed cracks using convolutional neural networks. Constr Build Mater2017; 146: 775–787.
30.
ChaYJChoiWBüyüköztürkO. Deep learning-based crack damage detection using convolutional neural networks. Comput Civ Infrastruct Eng2017; 32: 361–378.
31.
YuYRashidiMSamaliB, et al. Crack detection of concrete structures using deep convolutional neural networks optimized by enhanced chicken swarm algorithm. Struct Health Monitoring2022; 21: 2244–2263.
32.
ChaYJChoiWSuhG, et al. Autonomous structural visual inspection using region-based deep learning for detecting multiple damage types. Comput Civ Infrastruct Eng2018; 33: 731–747.
33.
LiDXieQGongX, et al. Automatic defect detection of metro tunnel surfaces using a vision-based inspection system. Adv Eng Inform2021; 47: 101206.
34.
DengJLuYLeeVCS. Concrete crack detection with handwriting script interferences using faster region-based convolutional neural network. Comput Civ Infrastruct Eng2020; 35: 373–388.
35.
KangDBenipalSSGopalDL, et al. Hybrid pixel-level concrete crack segmentation and quantification across complex backgrounds using deep learning. Autom Constr2020; 118: 103291.
36.
AliRKangDSuhG, et al. Real-time multiple damage mapping using autonomous UAV and deep faster region-based neural networks for GPS-denied structures. Autom Constr2021; 130: 103831.
37.
DengJLuYLeeVCS. Imaging-based crack detection on concrete surfaces using You Only Look Once network. Struct Health Monit2020; 20(2): 147592172093848.
38.
LiuJYangXLauS, et al. Automated pavement crack detection and segmentation based on two-step convolutional neural network. Comput Civ Infrastruct Eng2020; 35: 1291–1305.
39.
AlipourMHarrisDKMillerGR. Robust pixel-level crack detection using deep fully convolutional neural networks. J Comput Civ Eng2019; 33: 04019040.
40.
LeeJSHwangSHChoiIY, et al. Estimation of crack width based on shape-sensitive kernels and semantic segmentation. Struct Control Health Monit2020; 27: e2504.
41.
ZhangAWangKCPLiB, et al. Automated pixel-level pavement crack detection on 3D asphalt surfaces using a deep-learning network. Comput Civ Infrastruct Eng2017; 32: 805–819.
42.
RenYHuangJHongZ, et al. Image-based concrete crack detection in tunnels using deep fully convolutional networks. Constr Build Mater2020; 234: 117367.
43.
DengJLuYLeeVCS. A hybrid lightweight encoder-decoder network for automatic bridge crack assessment with real-world interference. Meas J Int Meas Confed2023; 216: 112892.
44.
RaoASNguyenTLeST, et al. Attention recurrent residual U-Net for predicting pixel-level crack widths in concrete surfaces, Struct Health Monit2022: 21: 147592172110688.
45.
KangDHChaY.-J. Efficient attention-based deep encoder and decoder for automatic crack segmentation. Struct Health Monit2022; 21: 2190–2205.
46.
KimBChoS. Image-based concrete crack assessment using mask and region-based convolutional neural network. Struct Control Health Monit2019; 26: e2381.
47.
XuYLiDXieQ, et al. Automatic defect detection and segmentation of tunnel surface using modified Mask R-CNN. Measurement2021; 178: 109316.
48.
HouSDongBWangH, et al. Inspection of surface defects on stay cables using a robot and transfer learning. Autom Constr2020; 119: 103382.
49.
ChoiWChaYJ. SDDNet: real-time crack segmentation. IEEE Trans Ind Electron2020; 67: 8016–8025.
50.
LiGLiXZhouJ, et al. Pixel-level bridge crack detection using a deep fusion about recurrent residual convolution and context encoder network. Measurement2021; 176: 109171.
51.
PanYZhangGZhangL. A spatial-channel hierarchical deep learning network for pixel-level automated crack detection. Autom Constr2020; 119: 103357.
52.
LiGLiuQRenW, et al. Automatic recognition and analysis system of asphalt pavement cracks using interleaved low-rank group convolution hybrid deep network and SegNet fusing dense condition random field. Measurement2021; 170: 108693.
53.
ZhengYGaoYLuS, et al. Multistage semisupervised active learning framework for crack identification, segmentation, and measurement of bridges. Comput Civ Infrastruct Eng2022; 37: 1089–1108.
54.
ChenLCPapandreouGKokkinosI, et al. DeepLab: semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected CRFs. In: IEEE Transactions on Pattern Analysis and Machine, 2018, pp. 834–848.
55.
ChenL.-CZhuYPapandreouG, et al. Encoder-decoder with atrous separable convolution for semantic image segmentation. In: The European conference on computer vision (ECCV), Munich, Germany, Springer International Publishing, 2018, pp. 801–818.
56.
HowardAGZhuMChenB, et al. MobileNets: efficient convolutional neural networks for mobile vision applications. ArXiv Prepr., ArXiv, 2017. http://arxiv.org/abs/1704.04861.
57.
SandlerMHowardAZhuM, et al. MobileNetV2: inverted residuals and linear bottlenecks. In: Proceedings of IEEE computer society conference on computer vision and pattern recognition, Salt Lake City, Utah, 2018, pp. 4510–4520.
58.
CholletF. Xception: deep learning with depthwise separable convolutions. In: Proceedings of the IEEE conference on computer vision and pattern recognition (CVPR), Honolulu, Hawaii, 2017, pp. 1251–1258.
59.
ChenL.-CPapandreouGSchroffF, et al. Rethinking atrous convolution for semantic image segmentation, ArXiv Prepr, arXiv:1706.05587, 2017. http://arxiv.org/abs/1706.05587.
60.
ZhangWLiXJiaXD, et al. Machinery fault diagnosis with imbalanced data using deep generative adversarial networks. Measurement2020; 152: 107377.
61.
BrostowGJFauqueurJCipollaR. Semantic object classes in video: a high-definition ground truth database. Pattern Recognit Lett2009; 30: 88–97.
62.
GoodfellowIBengioYCourvilleA. Deep learning. Cambridge, MA: MIT press, 2017.
63.
HeKZhangXRenS, et al. Deep residual learning for image recognition. In: Proceeding IEEE computer society conference on computer vision and pattern recognition, Las Vegas, NV, USA, December 2016, pp. 770–778. https://doi.org/10.1109/CVPR.2016.90.
64.
KingmaDPBaJ. Adam: a method for stochastic optimization. arXiv prepr, arXiv1412.6980, 2014. http://arxiv.org/abs/1412.6980.
65.
PascanuRMikolovTBengioY. On the difficulty of training recurrent neural networks. In: International conference on machine learning, Atlanta, GA, 2013, pp. 1310–1318.
66.
RobertC. Machine learning, a probabilistic perspective. Chance2014; 27: 62–63.
67.
ZhouZRahman SiddiqueeMMTajbakhshN, et al. Unet++: a nested U-net architecture for medical image segmentation. In: Deep learning in medical image analysis and multimodal learning for clinical decision support: 4th international workshop, DLMIA, Granada, Spain, 2018, pp. 3–11. https://doi.org/10.48550/arXiv.1807.10165.
