Computer vision-based concrete crack detection using U-net fully convolutional networks

Liu, Zhengqing; Cao, Yiwen; Wang, Yi–Ze; Wang, Wei

doi:10.1016/j.autcon.2019.04.005

Cited by 565 publications

(244 citation statements)

References 22 publications

Supporting

Mentioning

239

Contrasting

Unclassified

Order By: Relevance

“…First, we selected the best learning rate through training and testing. Then we empirically demonstrate the effectiveness of the proposed U-CliqueNet on the tunnel crack datasets that we set up and compared with these most recent algorithms: FCN [23] proposed by Yang et al, U-net [25] proposed by Liu et al, Bang's SegNet [38] and the MFCD [14] proposed by Li et al In addition, skeleton extraction was carried out for the predicted binary image, next the area, length and width of the crack are calculated.…”

Section: Resultsmentioning

confidence: 99%

“…In addition to the above three evaluation indicators, there are also three general indicators in crack detection, precision, recall and F1-score [19,25,37]. The precision represents the proportion of actual crack pixels in the predicted crack pixels, and the recall represents the proportion of correctly predicted crack pixels in the real crack pixels.…”

Section: Performance Evaluation Indicatorsmentioning

confidence: 99%

See 1 more Smart Citation

Automatic Tunnel Crack Detection Based on U-Net and a Convolutional Neural Network with Alternately Updated Clique

et al. 2020

Sensors

View full text Add to dashboard Cite

Regular crack inspection of tunnels is essential to guarantee their safe operation. At present, the manual detection method is time-consuming, subjective and even dangerous, while the automatic detection method is relatively inaccurate. Detecting tunnel cracks is a challenging task since cracks are tiny, and there are many noise patterns in the tunnel images. This study proposes a deep learning algorithm based on U-Net and a convolutional neural network with alternately updated clique (CliqueNet), called U-CliqueNet, to separate cracks from background in the tunnel images. A consumer-grade DSC-WX700 camera (SONY, Wuxi, China) was used to collect 200 original images, then cracks are manually marked and divided into sub-images with a resolution of 496 × 496 pixels. A total of 60,000 sub-images were obtained in the dataset of tunnel cracks, among which 50,000 were used for training and 10,000 were used for testing. The proposed framework conducted training and testing on this dataset, the mean pixel accuracy (MPA), mean intersection over union (MIoU), precision and F1-score are 92.25%, 86.96%, 86.32% and 83.40%, respectively. We compared the U-CliqueNet with fully convolutional networks (FCN), U-net, Encoder–decoder network (SegNet) and the multi-scale fusion crack detection (MFCD) algorithm using hypothesis testing, and it’s proved that the MIoU predicted by U-CliqueNet was significantly higher than that of the other four algorithms. The area, length and mean width of cracks can be calculated, and the relative error between the detected mean crack width and the actual mean crack width ranges from −11.20% to 18.57%. The results show that this framework can be used for fast and accurate crack semantic segmentation of tunnel images.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Performance Evaluation Indicatorsmentioning

confidence: 99%

Automatic Tunnel Crack Detection Based on U-Net and a Convolutional Neural Network with Alternately Updated Clique

et al. 2020

Sensors

View full text Add to dashboard Cite

show abstract

“…To improve existing approaches, other characteristics, such as the direction or shape of the cracks, should be considered in a smarter machine learning approach. More recently Deep Learning-based algorithms using convolutional neural networks have been introduced in this domain, which demonstrated promising results [5][6][7][8]. However, the examples of demonstration and implementation of these methods regarding the entire structure in large scale (not for a local detection problem) have been limited.…”

Section: Related Workmentioning

confidence: 99%

Automated Transverse Crack Mapping System with Optical Sensors and Big Data Analytics

Won

Sim

2020

Sensors

View full text Add to dashboard Cite

Transverse cracks on bridge decks provide the path for chloride penetration and are the major reason for deck deterioration. For such reasons, collecting information related to the crack widths and spacing of transverse cracks are important. In this study, we focused on developing a data pipeline for automated crack detection using non-contact optical sensors. We developed a data acquisition system that is able to acquire data in a fast and simple way without obstructing traffic. Understanding that GPS is not always available and odometer sensor data can only provide relative positions along the direction of traffic, we focused on providing an alternative localization strategy only using optical sensors. In addition, to improve existing crack detection methods which mostly rely on the low-intensity and localized line-segment characteristics of cracks, we considered the direction and shape of the cracks to make our machine learning approach smarter. The proposed system may serve as a useful inspection tool for big data analytics because the system is easy to deploy and provides multiple properties of cracks. Progression of crack deterioration, if any, both in spatial and temporal scale, can be checked and compared if the system is deployed multiple times.

show abstract

“…Among the developed methods for crack detection, edge-detection algorithms (Abdel-Qader et al, 2003), mathematical morphology (Sinha, Fieguth, 2006), high-speed percolation (Yamaguchi, Hashimoto, 2010), Principal Component Analysis (PCA) (Abdel-Qader et al, 2006), Extreme Learning Machine (Zhang et al, 2014) and Support Vector Machine (SVM) (Nashat et al, 2014) have been adopted. Furthermore, to improve the performances of image-based techniques and develop a method able to cover unexpected real-world situations, deep learning approaches (CNN -Convolutional Neural Network) have been also recently investigated (Li et al, 2018, Cha et al, 2017, Gopalakrishnan et al, 2017, Zou et al, 2018, Liu et al, 2019, Ren et al, 2020. Indeed, CNNs represent very powerful techniques for automatic feature extraction and classification problem and they have received considerable attention in the field of infrastructure monitoring thanks also to the spread of drones and other mobile mapping systems which can acquire a large amount of data.…”

Section: Introductionmentioning

confidence: 99%

“…The last approach provides pixel-level classification by detecting all the pixels which belong to the crack. It represents, therefore, the most accurate method for this specific task (Liu et al, 2019).…”

Section: Introductionmentioning

confidence: 99%

Tack Project: Tunnel and Bridge Automatic Crack Monitoring Using Deep Learning and Photogrammetry

Belloni

Sjölander²,

Ravanelli

et al. 2020

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

Abstract. Civil infrastructures, such as tunnels and bridges, are directly related to the overall economic and demographic growth of countries. The aging of these infrastructures increases the probability of catastrophic failures that results in loss of lives and high repair costs; all over the world, these factors drive the need for advanced infrastructure monitoring systems. For these reasons, in the last years, different types of devices and innovative infrastructure monitoring techniques have been investigated to automate the process and overcome the main limitation of standard visual inspections that are used nowadays. This paper presents some preliminary findings of an ongoing research project, named TACK, that combines advanced deep learning techniques and innovative photogrammetric algorithms to develop a monitoring system. Specifically, the project focuses on the development of an automatic procedure for crack detection and measurement using images of tunnels and bridges acquired with a mobile mapping system. In this paper, some preliminary results are shown to investigate the potential of a deep learning algorithm in detecting cracks occurred in concrete material. The model is a CNN (Convolutional Neural Network) based on the U-Net architecture; in this study, we tested the transferability of the model that has been trained on a small available labeled dataset and tested on a large set of images acquired using a customized mobile mapping system. The results have shown that it is possible to effectively detect cracks in unseen imagery and that the primary source of errors is the false positive detection of crack-like objects (i.e., contact wires, cables and tile borders).

show abstract

Computer vision-based concrete crack detection using U-net fully convolutional networks

Cited by 565 publications

References 22 publications

Automatic Tunnel Crack Detection Based on U-Net and a Convolutional Neural Network with Alternately Updated Clique

Automatic Tunnel Crack Detection Based on U-Net and a Convolutional Neural Network with Alternately Updated Clique

Automated Transverse Crack Mapping System with Optical Sensors and Big Data Analytics

Tack Project: Tunnel and Bridge Automatic Crack Monitoring Using Deep Learning and Photogrammetry

Contact Info

Product

Resources

About