Concrete crack detection and geometric features recognition based on deep learning

Yuanyuan DENG , Hui LIU , Ming LIU , Shuwei JU , Linlin WANG

ENG. Struct. Civ. Eng ››

PDF (8188KB)
ENG. Struct. Civ. Eng ›› DOI: 10.1007/s11709-026-1339-z
RESEARCH ARTICLE
Concrete crack detection and geometric features recognition based on deep learning
Author information +
History +
PDF (8188KB)

Abstract

Traditional precise detection and identification of geometric characteristics of concrete cracks are crucial for ensuring the safety of structures, yet lacking efficiency and accuracy. In this paper, the improved DeepLabv3+ model could significantly enhance the multi-scale feature extraction capability and edge detail segmentation accuracy of concrete cracks by replacing the backbone network with a Visual Attention Network network. The refined model not only optimized the Atrous Spatial Pyramid Pooling (ASPP) module into a multi-dilation rate DenseASPP structure, but also adopted strip pooling instead of global average pooling, which introduced the Polarized Self-Attention mechanism into the decoder. The proposed model is validated by integrating multiple crack image libraries including DeepCrack, Concrete Crack Segmentation and real dataset. The results indicated that the mean Intersection over Union value could reach 93.92%, which is 1.26%, 1%, 8.81%, 1.28%, 0.31% higher than DeepLabv3, DeepLabv3+, Fully Convolutional Network (FCN), Pyramid Scene Parsing Network, SegFormer, respectively. As a result, the image corrosion thinning algorithm could be used to characterize the geometric edge features. The geometric quantification error reduced significantly when compared with actual measurement values: the coefficient of determination (R2) for the maximum crack width increased from 0.954 to 0.966, with the mean absolute error decreased by 18.5%; the mean squared error for crack area measurement decreased by 37.3%.

Graphical abstract

Keywords

crack detection / deep learning / semantic segmentation / DeepLabv3+ algorithm / crack quantification

Cite this article

Download citation ▾
Yuanyuan DENG, Hui LIU, Ming LIU, Shuwei JU, Linlin WANG. Concrete crack detection and geometric features recognition based on deep learning. ENG. Struct. Civ. Eng DOI:10.1007/s11709-026-1339-z

登录浏览全文

4963

注册一个新账户 忘记密码

References

[1]

Alexander M , Beushausen H . Durability, service life prediction, and modelling for reinforced concrete structures—Review and critique. Cement and Concrete Research, 2019, 122: 17–29

[2]

JGJ 125-2016. Standard for Dangerous Building Appraisal. Beijing: Ministry of Housing and Urban Rural Development of the People’s Republic of China, 2016 (in Chinese)

[3]

Williams M S , Sexsmith R G . Seismic damage indices for concrete structures: A state-of-the-art review. Earthquake Spectra, 1995, 11(2): 319–349

[4]

Bracci J M , Kunnath S K , Reinhorn A M . Seismic performance and retrofit evaluation of reinforced concrete structures. Journal of Structural Engineering, 1997, 123(1): 3–10

[5]

Komura I , Hirasawa T , Nagai S , Takabayashi J I , Naruse K . Crack detection and sizing technique by ultrasonic and electromagnetic methods. Nuclear Engineering and Design, 2001, 206(2–3): 351–362

[6]

Laefer D F , Truong-Hong L , Carr H , Singh M . Crack detection limits in unit based masonry with terrestrial laser scanning. NDT & E International, 2014, 62: 66–76

[7]

Liu F Y , Liu J , Wang L B . Asphalt pavement crack detection based on convolutional neural network and infrared thermography. IEEE Transactions on Intelligent Transportation Systems, 2022, 23(11): 22145–22155

[8]

Pollock. Nondestructive evaluation of materials. ASM International, 2018, 360–379

[9]

Paul K , Balu A S , BabuNarayan K S . Fracture mechanics-based meshless method for crack propagation in concrete structures. Structures, 2025, 74: 108422

[10]

Munawar H S , Hammad A W A , Haddad A , Soares C A P , Waller S T . Image-based crack detection methods: A review. Infrastructures, 2021, 6(8): 115

[11]

Azouz Z , Honarvar Shakibaei Asli B , Khan M . Evolution of crack analysis in structures using image processing technique: A review. Electronics, 2023, 12(18): 3862

[12]

Koch C , Georgieva K , Kasireddy V , Akinci B , Fieguth P . A review on computer vision based defect detection and condition assessment of concrete and asphalt civil infrastructure. Advanced Engineering Informatics, 2015, 29(2): 196–210

[13]

Dinh T H, Ha Q P, La H M. Computer vision-based method for concrete crack detection. In: 2016 14th International Conference on Control, Automation, Robotics and Vision (ICARCV). Phuket: IEEE, 2016, 1–6

[14]

Garita-Durán H , Stöcker J P , Kaliske M . Deep learning-based system for automated damage detection and quantification in concrete pavement. Results in Engineering, 2025, 25: 104546

[15]

Li F F , Mou Y L , Zhang Z Y , Liu Q , Jeschke S . A novel model for the pavement distress segmentation based on multi-level attention DeepLabV3+. Engineering Applications of Artificial Intelligence, 2024, 137: 109175

[16]

Lu X C , Li Q Q , Li J Y , Zhang L . Deep learning-based method for detection and feature quantification of microscopic cracks on the surface of concrete dams. Measurement, 2025, 240: 115587

[17]

Li M C , Yuan J Y , Ren Q B , Luo Q L , Fu J N , Li Z T . CNN-Transformer hybrid network for concrete dam crack patrol inspection. Automation in Construction, 2024, 163: 105440

[18]

Wu Z H , Tang Y C , Hong B , Liang B Q , Liu Y P . Enhanced precision in dam crack width measurement: Leveraging advanced lightweight network identification for pixel-level accuracy. International Journal of Intelligent Systems, 2023, 2023: 9940881

[19]

El-Hawwat S , Shah J K , Wang H , Venkiteela G . Detection of internal cracks in polyethylene pipes using ultrasonic imaging and deep learning. Measurement, 2025, 253: 117491

[20]

Du Q Y , Jiang Q . Improvement of crack detectivity for concrete surface of subway tunnels with anti-corrosion coatings using deep learning and image processing. Case Studies in Construction Materials, 2025, 22: e04131

[21]

Zhang J , Lu Y M , Yang Z , Zhu X , Zheng T , Liu X , Tian Y G , Li W G . Recognition of void defects in airport runways using ground-penetrating radar and shallow CNN. Automation in Construction, 2022, 138: 104260

[22]

Guo W , Liang G Y , Ren S X , Zeng C . Automatic crack detection method for high-speed railway box girder based on deep learning techniques and inspection robot. Structures, 2024, 68: 107116

[23]

Pan Z T , Zhang X P , Jiang Y J , Li B , Golsanami N , Su H , Cai Y . High-precision segmentation and quantification of tunnel lining crack using an improved DeepLabV3+. Underground Space, 2025, 22: 96–109

[24]

Hu W B , Wang W D , Ai C B , Wang J , Wang W J , Meng X F , Liu J , Tao H W , Qiu S . Machine vision-based surface crack analysis for transportation infrastructure. Automation in Construction, 2021, 132: 103973

[25]

Cha Y J , Choi W , Suh G , Mahmoudkhani S , Büyüköztürk O . Autonomous structural visual inspection using region-based deep learning for detecting multiple damage types. Computer-Aided Civil and Infrastructure Engineering, 2018, 33(9): 731–747

[26]

Dung C V , Anh L D . Autonomous concrete crack detection using deep fully convolutional neural network. Automation in Construction, 2019, 99: 52–58

[27]

Long J, Shelhamer E, Darrell T. Fully convolutional networks for semantic segmentation. In: 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Boston: IEEE, 2015, 3431–3440

[28]

Chen L C, Zhu Y K, Papandreou G, Schroff F, Adam H. Encoder-decoder with atrous separable convolution for semantic image segmentation. In: 15th European Conference on Computer Vision (ECCV). Munich: Springer, 2018, 833–851

[29]

Ronneberger O, Fischer P, Brox T. U-Net: Convolutional networks for biomedical image segmentation. In: 18th International Conference on Medical Image Computing and Computer-Assisted Intervention. Munich: Springer, 2015, 234–241

[30]

Wang S W , Xu J B , Wu X , Zhang J J , Zhang Z X , Chen X Y . Concrete crack recognition and geometric parameter evaluation based on deep learning. Advances in Engineering Software, 2025, 199: 103800

[31]

Zhou S X , Pan Y , Guan J Y , Wang Q . Multi-band feature images concrete crack segmentation framework using deep learning. KSCE Journal of Civil Engineering, 2024, 28(9): 3902–3912

[32]

Hu K W , Chen Z , Kang H W , Tang Y C . 3D vision technologies for a self-developed structural external crack damage recognition robot. Automation in Construction, 2024, 159: 105262

[33]

Chaiyasarn K , Buatik A , Mohamad H , Zhou M L , Kongsilp S , Poovarodom N . Integrated pixel-level CNN-FCN crack detection via photogrammetric 3D texture mapping of concrete structures. Automation in Construction, 2022, 140: 104388

[34]

Zhu Y T , Tang H W . Automatic damage detection and diagnosis for hydraulic structures using drones and artificial intelligence techniques. Remote Sensing, 2023, 15(3): 615

[35]

Bhardwaj M , Khan N U , Baghel V . Fuzzy C-Means clustering based selective edge enhancement scheme for improved road crack detection. Engineering Applications of Artificial Intelligence, 2024, 136: 108955

[36]

Tang Y C , Huang Z F , Chen Z , Chen M Y , Zhou H , Zhang H X , Sun J B . Novel visual crack width measurement based on backbone double-scale features for improved detection automation. Engineering Structures, 2023, 274: 115158

[37]

Guo M H , Lu C Z , Liu Z N , Cheng M M , Hu S M . Visual attention network. Computational Visual Media, 2023, 9(4): 733–752

[38]

Yuan J Y , Ren Q B , Jia C , Zhang J T , Fu J K , Li M C . Automated pixel-level crack detection and quantification using deep convolutional neural networks for structural condition assessment. Structures, 2024, 59: 105780

[39]

Su M , Wan J K , Zhou Q L , Wang R , Xie Y X , Peng H . Utilizing pretrained convolutional neural networks for crack detection and geometric feature recognition in concrete surface images. Journal of Building Engineering, 2024, 98: 111386

Rights & permissions

Higher Education Press

PDF (8188KB)

0

Accesses

0

Citation

Detail

Sections
Recommended

/

〈 〉