Restricted accessResearch articleFirst published online 2026
An intelligent monitoring framework for crack anomaly detection and early warning in concrete bridges using GAF-based classification and GRU prediction
Crack development in concrete is a crucial indicator for the safety assessment of bridge structures, as abnormal crack behavior often reflects underlying structural deterioration. Under normal service conditions, a strong negative correlation typically exists between crack width and ambient temperature. This study presents an intelligent framework for concrete bridge crack monitoring by establishing a temperature–crack correlation model. Temperature-induced crack features are first extracted using wavelet packet decomposition and transformed into Gramian Angular Field images. These are input into a convolutional neural network for automated classification, which demonstrates high accuracy (98.67%) in identifying anomalous channels deviating from group-level behavior. This step facilitates the detection of sensor faults and localized structural anomalies, thereby ensuring reliable inputs for further modeling. Subsequently, a gated recurrent unit-based model is developed for crack width prediction, achieving R2 scores of 0.9637 and 0.9310 on the training and test sets, respectively. To enhance anomaly detection, a residual-based dynamic threshold warning method is introduced, maintaining a low outlier rate (0.56%–1.86%) on typical datasets, and robust performance (below 6%) even with significant noise. Field monitoring results confirm the robustness and practical applicability of the proposed method. Overall, this framework offers an effective approach for early warning of potential safety hazards in concrete bridges, contributing to improved operational safety and extended service life.
MaYBaoTLiY, et al. A framework for automatic Real-Time Pixel-Level segmentation of underwater dam concrete cracks utilizing the CRTransU-Net model. Adv Eng Inform2025; 66: 103415.
2.
HakimiOLiuHAbudayyehO.Deep learning-driven multi-level data fusion framework for predictive maintenance and structural health monitoring of concrete bridge decks. Automat Constr2025; 175: 106180.
3.
LiHWuXNieQ, et al. Lifetime prediction of damaged or cracked concrete structures: a review. Structures2025; 71: 108095.
4.
FanCDingYLiuX, et al. A review of crack research in concrete structures based on data-driven and intelligent algorithms. Structures2025; 75: 108800.
5.
ZhangYFengHZhaoJ, et al. Fatigue crack propagation for Ultra-High-Performance Concrete-normal strength concrete interface with restrained strain. Int J Fatigue2025; 199: 109039.
6.
LeiJXuCLüC, et al. A data-driven prediction for concrete crack propagation path based on deep learning method. Case Stud Constr Mater2024; 21: e03883.
7.
ZhangXDingYHouS, et al. Intelligent perception for multiple intrusion events of urban long-line infrastructures with distributed optical fiber sensing. Measurement2025; 248: 116921.
8.
KasaiNLeMSekinoK, et al. Crack detection using a cutting-edge flexible eddy current sensor with voltage and phase measurement techniques. Mech Syst Sig Process2024; 219: 111613.
9.
SiahkouhiMRashidiMMashiriF, et al. Application of self-sensing concrete sensors for bridge monitoring − A review of recent developments, challenges, and future prospects. Measurement2024; 245: 116543.
10.
WangPZhuJScaioniM.A rapid and precise detection and measurement method for road cracks based on video keyframes and DPPNet. Measurement2025; 254: 117493.
11.
XuGZhangYYueQ, et al. A deep learning framework for real-time multi-task recognition and measurement of concrete cracks. Adv Eng Inform2025; 65: 103127.
12.
SahRKKumarAGautamA, et al. Temperature independent FBG based displacement sensor for crack detection in civil structures. Optical Fiber Technol2022; 74: 103137.
13.
ChenGBianZJingH, et al. Crack identification of concrete structures based on high-precision multi-level deep learning model. Structures2025; 75: 108720.
14.
LiuYZhouTHongY, et al. Neighborhood shortest distance method for concrete crack width detection in images. Eng Struct2025; 326: 119519.
15.
ZhaoHWDingYLLiAQ, et al. Digital modeling approach of distributional mapping from structural temperature field to temperature-induced strain field for bridges. J Civil Struct Health Monitor2023; 13(1): 251–267.
16.
YueZDingYZhaoH, et al. Mechanics-guided optimization of an LSTM network for real-time modeling of temperature-induced deflection of a cable-stayed bridge. Eng Struct2022; 252: 113619.
17.
GongXSongXCaiC S, et al. A temperature-driven approach for quantitative assessment of strengthening effect of continuous bridges using structural health monitoring data. Struct Health Monitor2024; 23(2): 1053–1070.
18.
DanDZengGPanR, et al. Block-wise recursive sliding variational mode decomposition method and its application on online separating of bridge vehicle-induced strain monitoring signals. Mech Syst Sig Process2023; 198: 110389.
19.
LiSXinJJiangY, et al. Temperature-induced deflection separation based on bridge deflection data using the TVFEMD-PE-KLD method. J Civil Struct Health Monitor2023; 13(2): 781–797.
20.
ZhaoHDingYLiA, et al. Digital modeling on the nonlinear mapping between multi-source monitoring data of in-service bridges. Struct Contr Health Monitor2020; 27(11): e2618.
21.
YangKDingYGengF, et al. A multi-sensor mapping Bi-LSTM model of bridge monitoring data based on spatial-temporal attention mechanism. Measurement2023; 217: 113053.
22.
ZhuSYangMXiangT, et al. Advanced time-series prediction of bridge long-term deflection using the learning models. Structures2024; 67: 106967.
23.
YangNWangSFuY.Analysis of crack characteristics in Tibetan-style historic masonry buildings based on monitoring data and their correlation with temperature. China Civil Eng J. Epub ahead of print 24January2026. DOI:10.15951/j.tmgcxb.24090719.
24.
DengYJuHZhaiW, et al. Correlation model of deflection, vehicle load, and temperature for in-service bridge using deep learning and structural health monitoring. Struct Contr Health Monitor2022; 29(12): e3113.
25.
JinWChenSYunJ, et al. Time series analysis of bridge cracks under temperature effect. CE/Papers2025; 8(2): 1765–1771.
26.
LvZDingYZhangY.Study on long-term temperature variation characteristics of concrete bridge tower cracks based on deep learning. Sensors2025; 25(1): 207.
27.
ChenNLiuGMaR, et al. Wind field characterization with skip-connected variational autoencoder for data cleaning under deck disturbance effects. Eng Struct2025; 330: 119905.
28.
YiLDingYHouJ, et al. Structural health monitoring data cleaning based on Bayesian robust tensor learning. Struct Health Monitor2023; 22(4): 2169–2192.
29.
LeeTJinSSKimST, et al. Online anomaly detection for long-term structural health monitoring of caisson quay walls. Eng Struct2025; 323: 119197.
30.
YangKDingYJiangH, et al. A two-stage data cleansing method for bridge global positioning system monitoring data based on bi-direction long and short term memory anomaly identification and conditional generative adversarial networks data repair. Struct Contr Health Monitor2022; 29(9): e2993.
31.
HuangTChakrabortyPSharmaA.Deep convolutional generative adversarial networks for traffic data imputation encoding time series as images. Int J Transp Sci Technol2023; 12(1): 1–18.
32.
DongSMengYYinS, et al. Tool wear state recognition study based on an MTF and a vision transformer with a Kolmogorov-Arnold network. Mech Syst Sig Process2025; 228: 112473.
33.
MaoJWangHSpencerBFJr.Toward data anomaly detection for automated structural health monitoring: exploiting generative adversarial nets and autoencoders. Struct Health Monitor2021; 20(4): 1609–1626.
34.
GongXSongXLiG, et al. Deep learning based anomaly identification of temperature effects in bridge structural health monitoring data. Structures2024; 69: 107478.
35.
ChenLXuDYangL, et al. Classification and identification of extreme wind events by CNNs based on Shapelets and improved GASF-GADF. J Wind Eng Industr Aerodyn2024; 253: 105852.
36.
BianSWangZSongW, et al. Feature extraction and classification of time-varying power load characteristics based on PCANet and CNN+ Bi-LSTM algorithms. Electr Power Syst Res2023; 217: 109149.
37.
ZhaoYYangZHouZ, et al. Deep learning based on image analysis for refrigerant charging and leakage detection in building heat pump. Energ Build2025; 328: 115157.
38.
ZhangHMaLYuanZ, et al. Enhanced concrete crack detection and proactive safety warning based on I-ST-UNet model. Automat Constr2024; 166: 105612.
39.
XieAYLiBQ.Transfer learning framework for multi-scale crack type classification with sparse microseismic networks. Int J Mining Sci Technol2024; 34(2): 167–178.
40.
GiglioniVPooleJMillsR, et al. Transfer learning in bridge monitoring: laboratory study on domain adaptation for population-based SHM of multispan continuous girder bridges. Mech Syst Sig Process2025; 224: 112151.
41.
AhmadzadehMZahraiSMBitarafM.An integrated deep neural network model combining 1D CNN and LSTM for structural health monitoring utilizing multisensor time-series data. Struct Health Monitor2025; 24(1): 447–465.
42.
ChenWWangXTongD, et al. Dynamic early-warning model of dam deformation based on deep learning and fusion of spatiotemporal features. Knowl Based Syst2021; 233: 107537.
43.
ZhuangHChengYZhouM, et al. Deep learning for surface crack detection in civil engineering: a comprehensive review. Measurement2025; 2025: 116908.
44.
LiXZhangFLXiangW, et al. Structural health monitoring system based on digital twins and real-time data-driven methods. Structures, 2024, 70: 107739.
45.
HuYLuoPLiuH, et al. Open-circuit fault diagnosis method of inverter in wind turbine yaw system based on GADF image coding. Comput Electric Eng2024; 117: 109252.
46.
TaoYXuZ DWeiY, et al. A wavelet packet deep learning model for Energy-Based structural collapse assessment under Earthquake-Fire Scenarios: framework and hybrid simulation. Mech Syst Sig Process2025; 222: 111784.
47.
HouBWangDXiaT, et al. Difference mode decomposition for adaptive signal decomposition. Mech Syst Sig Process2023; 191: 110203.
48.
LiGZhouMFuY, et al. An interpretable CNN-based model for mass classification in mammography. Knowl Based Syst2025; 316: 113372.
49.
MatarnehSElghaishFRahimianFP, et al. Evaluation and optimisation of pre-trained CNN models for asphalt pavement crack detection and classification. Automat Constr2024; 160: 105297.
50.
WangXXieGLiuW, et al. A long-term vertical displacement prediction method of concrete bridges based on meteorological shared data and optimized GRU model. Measurement2025: 117811.
51.
IslamSMaswoodMMS. Designing novel deep learning based frameworks for performance improvement of automatic modulation classification. Eng Appl Artif Intell2025; 152: 110796.
52.
FanCDingYZhengY.Bond strength and failure mode prediction model for recycled aggregate concrete based on intelligent algorithm optimized support vector machine. Structures2025; 71: 107999.
53.
TengSLiuASituZ, et al. Plug-and-play method for segmenting concrete bridge cracks using the segment anything model with a fractal dimension matrix prompt. Automat Constr2025; 170: 105906.
54.
FanCZhengYWenY, et al. Classification and prediction of deformed steel and concrete bond-slip failure modes based on SSA-ELM model. Structures2023; 57: 105131.
55.
SongZXiaoFChenZ, et al. Probabilistic ultra-short-term solar photovoltaic power forecasting using natural gradient boosting with attention-enhanced neural networks. Energy AI2025; 20: 100496.
56.
DaiJYanCTangY.A hybrid ultra-short-term wind speed forecasting model for wind farms based on temporal residual probability. Power Syst Technol2023; 47(2): 688–699.
57.
CorballyRMalekjafarianA.Bridge damage detection using operating deflection shape ratios obtained from a passing vehicle. J Sound Vib2022; 537: 117225.
58.
LiuZSongXZhangM.A packet-dropping fusion Kalman filter algorithm based on non-Gaussian noise estimation. Mech Syst Sig Process2025; 228: 112457.
59.
FanCZhengYWangB, et al. Damage identification method for tied arch bridge suspender based on quasi-static displacement influence line. Mech Syst Sig Process2023; 200: 110518.
60.
SenSAswalNZhangQ, et al. Structural health monitoring with non-linear sensor measurements robust to unknown non-stationary input forcing. Mech Syst Sig Process2021; 152: 107472.
61.
ZhouYDiSXiangC, et al. Damage detection for simply supported bridge with bending fuzzy stiffness consideration. J Shanghai Jiaotong Univ2018; 23: 308–319.
