Cracks are the defects formed by cyclic loading, fatigue, shrinkage, creep, and so on. In addition, they represent the deterioration of the structures over some time. Therefore, it is essential to detect and classify them according to the condition grade at the early stages to prevent the collapse of structures. Deep learning-based semantic segmentation convolutional neural network (CNN) has millions of learnable parameters. However, depending on the complexity of the CNN, it takes hours to days to train the network fully. In this study, an encoder network DenseNet and modified LinkNet with five upsampling blocks were used as a decoder network. The proposed network is referred to as the “CrackDenseLinkNet” in this work. CrackDenseLinkNet has 19.15 million trainable parameters, although the input image size is 512 × 512 and has a deeper encoder. CrackDenseLinkNet and four other state-of-the-art (SOTA) methods were evaluated on three public and one private datasets. The proposed CNN, CrackDenseLinkNet, outperformed the best SOTA method, CrackSegNet, by 2.2% of F1-score on average across the four datasets. Lastly, a crack profile analysis demonstrated that the CrackDenseLinkNet has lesser variance in relative errors for the crack width, length, and area categories against the ground-truth data. The code and datasets can be downloaded at https://github.com/preethamam/CrackDenseLinkNet-DeepLearning-CrackSegmentation.
ZhuZGermanSBrilakisI.Visual retrieval of concrete crack properties for automated post-earthquake structural safety evaluation. Autom Constr2011; 20(7): 874–883.
2.
YangYSWuClHsuTT, et al. Image analysis method for crack distribution and width estimation for reinforced concrete structures. Autom Constr2018; 91: 120–132.
3.
TongXGuoJLingY, et al. A new image-based method for concrete bridge bottom crack detection. In: 2011 international conference on image analysis and signal processing, Hubei, 21–23 October 2011, IEEE, pp. 568–571.
4.
FujitaYHamamotoY.A robust method for automatically detecting cracks on noisy concrete surfaces. In: International conference on industrial, engineering and other applications of applied intelligent systems, Springer, Berlin, Heidelberg, 2009, pp. 76–85.
5.
LeeBYKimYYYiST, et al. Automated image processing technique for detecting and analysing concrete surface cracks. Struct Infrastruct Eng2013; 9(6): 567–577.
6.
JahanshahiMRMasriSFPadgettCW, et al. An innovative methodology for detection and quantification of cracks through incorporation of depth perception. Mach Vision Appl2013; 24(2): 227–241.
7.
JahanshahiMRMasriSF.Adaptive vision-based crack detection using 3D scene reconstruction for condition assessment of structures. Autom Constr2012; 22: 567–576.
8.
SchmidhuberJ.Deep learning in neural networks: an overview. arXiv Preprint arXiv:1404.7828, 2014.
ChenFCJahanshahiM.ARF-crack: rotation invariant deep fully convolutional network for pixel-level crack detection. Mach Vision Appl2020; 31: 47.
11.
ZhangLYangFZhangYD, et al. Road crack detection using deep convolutional neural network. In: 2016 IEEE international conference on image processing (ICIP), Phoenix, AZ, USA, IEEE, 2016, pp. 3708–3712.
12.
ChenFCJahanshahiMR.NB-CNN: deep learning-based crack detection using convolutional neural network and naïve bayes data fusion. IEEE Trans Ind Electron2017; 65(5): 4392–4400.
13.
ChenFCJahanshahiMR.Real-time crack detection from nuclear inspection videos using fully convolutional network and parametric data fusion. IEEE Trans Instrum Meas2019; 69(8): 5325–5334.
KrizhevskyASutskeverIHintonGE.Imagenet classification with deep convolutional neural networks. Commun ACM2017; 60(6): 84–90.
16.
DorafshanSThomasRJMaguireM.Comparison of deep convolutional neural networks and edge detectors for image-based crack detection in concrete. Constr Build Mater2018; 186: 1031–1045.
17.
SimonyanKZissermanA.Very deep convolutional networks for large-scale image recognition. arXiv Preprint arXiv:14091556, 2014.
18.
SzegedyCLiuWJiaY, et al. Going deeper with convolutions. In: Proceedings of the IEEE conference on computer vision and pattern recognition, IEEE, Boston, MA, USA, 2015, pp. 1–9.
19.
HeKZhangXRenS, et al. Deep residual learning for image recognition. In: Proceedings of the IEEE conference on computer vision and pattern recognition, Las Vegas, NV, USA, 2016, pp. 770–778.
20.
LiuYYaoJLuX, et al. Deepcrack: a deep hierarchical feature learning architecture for crack segmentation. Neurocomputing2019; 338: 139–153.
21.
YangXLiHYuY, et al. Automatic pixel-level crack detection and measurement using fully convolutional network. Comput-Aided Civ Infrastruct Eng2018; 33(12): 1090–1109.
22.
YangFZhangLYuS, et al. Feature pyramid and hierarchical boosting network for pavement crack detection. IEEE Trans Intell Transp Syst2019; 21(4): 1525–1535.
23.
GaoYMosalamKM.Deep transfer learning for image-based structural damage recognition. Comput-Aided Civ Infrastruct Eng2018; 33(9): 748–768.
24.
AlfazNHasnatAKhanAMRN, et al. Bridge crack detection using dense convolutional network (densenet). In: Proceedings of the 2nd International Conference on Computing Advancements, ICCA ’22, New York, NY, USA, 2022, pp. 509–515.
25.
QiaoWMaBLiuQ, et al. Computer vision-based bridge damage detection using deep convolutional networks with expectation maximum attention module. Sensors2021; 21(3): 824.
26.
LiHZongJNieJ, et al. Pavement crack detection algorithm based on densely connected and deeply supervised network. IEEE Access2021; 9: 11835–11842.
27.
HuangGLiuZVan Der MaatenL, et al. Densely connected convolutional networks. In: Proceedings of the IEEE conference on computer vision and pattern recognition, Honolulu, 2017, Pp. 4700–4708.
28.
LiSZhaoX.Image-based concrete crack detection using convolutional neural network and exhaustive search technique. Adv Civ Eng2019; 2019: 6520620.
29.
LeTTNguyenVHLeMV.Development of deep learning model for the recognition of cracks on concrete surfaces. Appl Comput Intell Soft Comput2021; 2021: 8858545.
30.
ZhangYYuenKV.Crack detection using fusion features-based broad learning system and image processing. Comput-Aided Civ Infrastruct Eng2021; 36: 1568–1584.
31.
RaoASNguyenTPalaniswamiM, et al. Vision-based automated crack detection using convolutional neural networks for condition assessment of infrastructure. Struct Health Monit2021; 20(4): 2124–2142.
32.
KyalCRezaMVaruB, et al. Image-based concrete crack detection using random forest and convolution neural network. In: Computational Intelligence in Pattern Recognition. Springer, Singapore, 2022, pp. 471–481.
33.
KimBChoS.Automated vision-based detection of cracks on concrete surfaces using a deep learning technique. Sensors2018; 18(10): 3452.
34.
KimHAhnEShinM, et al. Crack and noncrack classification from concrete surface images using machine learning. Struct Health Monit2019; 18(3): 725–738.
35.
ParkSBangSKimH, et al. Patch-based crack detection in black box images using convolutional neural networks. J Comput Civ Eng2019; 33(3): 04019017.
36.
ZhangXRajanDStoryB.Concrete crack detection using context-aware deep semantic segmentation network. Comput-Aided Civ Infrastruct Eng2019; 34(11): 951–971.
37.
JiangSZhangJ.Real-time crack assessment using deep neural networks with wall-climbing unmanned aerial system. Comput-Aided Civ Infrastruct Eng2020; 35(6): 549–564.
38.
DengJLuYLeeVCS. Concrete crack detection with handwriting script interferences using faster region-based convolutional neural network. Comput-Aided Civ Infrastruct Eng2020; 35(4): 373–388.
39.
ParkSEEemSHJeonH.Concrete crack detection and quantification using deep learning and structured light. Constr Build Mater2020; 252: 119096.
40.
ZhangCChangCcJamshidiM.Concrete bridge surface damage detection using a single-stage detector. Comput-Aided Civ Infrastruct Eng2020; 35(4): 389–409.
41.
LiYLiHWangH.Pixel-wise crack detection using deep local pattern predictor for robot application. Sensors2018; 18(9): 3042.
42.
NiFZhangJChenZ.Zernike-moment measurement of thin-crack width in images enabled by dual-scale deep learning. Comput-Aided Civ Infrastruct Eng2019; 34(5): 367–384.
43.
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 Monit2019; 18(3): 653–674.
44.
HoskereVNarazakiYHoangT, et al. Vision-based structural inspection using multiscale deep convolutional neural networks. arXiv Preprint arXiv:180501055, 2018.
45.
WangBLiYZhaoW, et al. Effective crack damage detection using multilayer sparse feature representation and incremental extreme learning machine. Appl Sci2019; 9(3): 614.
46.
DrouyerS.An’all terrain’crack detector obtained by deep learning on available databases. Image Process Line2020; 10: 105–123.
47.
AlipourMHarrisDK.Increasing the robustness of material-specific deep learning models for crack detection across different materials. Eng Struct2020; 206: 110157.
48.
RenYHuangJHongZ, et al. Image-based concrete crack detection in tunnels using deep fully convolutional networks. Constr Build Mater2020; 234: 117367.
49.
MeiQGülMAzimMR.Densely connected deep neural network considering connectivity of pixels for automatic crack detection. Autom Constr2020; 110: 103018.
50.
QuZMeiJLiuL, et al. Crack detection of concrete pavement with cross-entropy loss function and improved VGG16 network model. IEEE Access2020; 8: 54564–54573.
51.
LiSZhaoX.Automatic crack detection and measurement of concrete structure using convolutional encoder-decoder network. IEEE Access2020; 8: 134602–134618.
52.
LiGMaBHeS, et al. Automatic tunnel crack detection based on u-net and a convolutional neural network with alternately updated clique. Sensors2020; 20(3): 717.
53.
FengCZhangHWangH, et al. Automatic pixel-level crack detection on dam surface using deep convolutional network. Sensors2020; 20(7): 2069.
54.
LiGLiuQZhaoS, et al. Automatic crack recognition for concrete bridges using a fully convolutional neural network and naive bayes data fusion based on a visual detection system. Meas Sci Technol2020; 31(7): 075403.
55.
KangDBenipalSSGopalDL, et al. Hybrid pixel-level concrete crack segmentation and quantification across complex backgrounds using deep learning. Autom Constr2020; 118: 103291.
56.
HuangYZhangHLiH, et al. Recovering compressed images for automatic crack segmentation using generative models. Mech Syst Signal Process2021; 146: 107061.
57.
YangQJiX.Automatic pixel-level crack detection for civil infrastructure using unet++ and deep transfer learning. IEEE Sens J2021; 21: 19165–19175.
58.
ChaurasiaACulurcielloE.Linknet: exploiting encoder representations for efficient semantic segmentation. In: 2017 IEEE Visual Communications and Image Processing (VCIP), IEEE, St. Petersburg, FL, USA, 2017, pp. 1–4.
59.
DengJDongWSocherR, et al. Imagenet: a large-scale hierarchical image database. In: 2009 IEEE conference on computer vision and pattern recognition, IEEE, Miami, FL, USA, 2009, pp. 248–255.
60.
WangPChungAC.Focal dice loss and image dilation for brain tumor segmentation. In: Deep learning in medical image analysis and multimodal learning for clinical decision support: 4th international workshop, DLMIA 2018, and 8th international workshop, MLCDS 2018, held in conjunction with MICCAI 2018, Granada, Spain, September 20, 2018, Proceedings 4. Springer, pp. 119–127.
61.
IantsenAVisvikisDHattM.Squeeze-and-excitation normalization for automated delineation of head and neck primary tumors in combined pet and ct images. In: Head and Neck Tumor Segmentation: First Challenge, HECKTOR 2020, Held in Conjunction with MICCAI 2020, Lima,Peru, October 4, 2020,Proceedings 1, Springer, Cham, 2020, pp. 37–43.
62.
ZhuWHuangYZengL, et al. Anatomynet: deep learning for fast and fully automated whole-volume segmentation of head and neck anatomy. Med Phys2019; 46(2): 576–589.
63.
YeungMSalaESch··onliebCB, et al. Unified focal loss: generalising dice and cross entropy-based losses to handle class imbalanced medical image segmentation. Comput Med Imaging Graphics2022; 95: 102026.
64.
ManjunathAJoisSSeelamantulaCS.Robust segmentation of optic disc and cup from fundus images using deep neural networks. arXiv Preprint arXiv: 201207128, 2020.
65.
LiuZCaoYWangY, et al. Computer vision-based concrete crack detection using u-net fully convolutional networks. Autom Constr2019; 104: 129–139.
66.
ZhangLShenJZhuB.A research on an improved unet-based concrete crack detection algorithm. Struct Health Monit2021; 20(4): 1864–1879.
67.
LinTYGoyalPGirshickR, et al. Focal loss for dense object detection. IEEE Trans Pattern Anal Mach Intell2020; 42(2): 318–327.
68.
DrozdzalMVorontsovEChartrandG, et al. The importance of skip connections in biomedical image segmentation. In: Deep learning and data labeling for medical applications, Springer, Cham, 2016, pp. 179–187.
69.
ShelhamerELongJDarrellT.Fully convolutional networks for semantic segmentation. IEEE Trans Pattern Anal Mach Intell2017; 39: 640–651.
70.
HeKZhangXRenS, et al. Delving deep into rectifiers: surpassing human-level performance on imagenet classification. In: Proceedings of the IEEE international conference on computer vision, Santiago, Chile, 2015, pp. 1026–1034.
71.
TeleaAvan WijkJJ.An augmented fast marching method for computing skeletons and centerlines. In: Proceedings of the Symposium on Data Visualisation 2002, VISSYM ’02, DEU, Eurographics Association, Goslar, Germany, 2002, pp. 251–259.
72.
QiuSWangWWangS, et al. Methodology for accurate AASHTO PP67-10–based cracking quantification using 1-mm 3D pavement images. J Comput Civ Eng2017; 31(2): 04016056.
