Computer vision framework for crack detection of civil infrastructure - A review

Automated bridge surface crack detection and segmentation using computer vision-based deep learning model

Rubber gaskets are commonly adopted as the waterproof component in shield tunnels for their outstanding sealing performance. The contact pressure between surfaces generated by the assembly stress ensures that the gaskets resist certain water pressure without leaking. However, with the continuous occurrence of leakage accidents, attention has been drawn to the topic of the waterproof performance of gasketed joint shield tunnels. In this article, prominent contributions to the waterproof performance of sealing gasket in shield tunnels are listed and sorted into four sections: (1) structural behavior of lining and joint; (2) material constitutive model and durability; (3) numerical simulation methods; (4) thermal-mechanical coupling analysis. First, examples of leakage are discussed and tests on gaskets are elucidated, which is followed by a summary of the progress on material mechanical properties and durability. Then, the development of the simulation methods is presented. Finally, the existing research on the thermal-mechanical coupling analysis is summarized. It is found that the contributions to gaskets’ waterproof performance are fruitful, however, with stringent construction conditions, such as the material constitutive model and aging mechanism under special conditions, such as high temperature, numerical simulation, and laboratory test methods, which need to be further explored.

/pdf/waterproof-performance-of-sealing-gasket-in-shield-tunnel-a-2vtpvsth.pdf

Waterproof Performance of Sealing Gasket in Shield Tunnel: A Review

Automatic tunnel lining crack detection via deep learning with generative adversarial network-based data augmentation

Automatic defogging, deblurring, and real-time segmentation system for sewer pipeline defects

Aiming to solve the challenges of low detection accuracy, poor anti‐interference ability, and slow detection speed in the traditional tunnel lining defect detection methods, a novel deep learning‐based model, named You Only Look Once network v4 enhanced by EfficientNet and depthwise separable convolution (DSC; YOLOv4‐ED), is proposed. In the YOLOv4‐ED, EfficientNet is used as the backbone to improve the identification accuracy of indistinguishable defect targets in complex tunnel background and light conditions. Furthermore, DSC block is introduced to reduce the storage space of the model and thereby enhance the detection efficiency. The experimental results indicate that the mean average precision, F1 score, Model Size, and FPS of YOLOv4‐ED are 81.84%, 81.99%, 49.3 MB, and 43.5 f/s, respectively, which is superior to the comparison models in both detection accuracy and efficiency. Based on robust and cost‐effective YOLOv4‐ED, a tunnel lining defect detection platform (TLDDP) with the capacity of automated inspection of various lining defects (i.e., water leakage, crack, rebar‐exposed) is built. The established TLDDP can realize the high‐precision and automatic detection of multiple tunnel lining defects under different lighting and complex background conditions of the practical in‐service tunnel.

Automatic detection method of tunnel lining multi‐defects via an enhanced You Only Look Once network

In the field of tunnel lining crack identification, the semantic segmentation algorithms based on convolution neural network (CNN) are extensively used. Owing to the inherent locality of CNN, these algorithms cannot make full use of context semantic information, resulting in difficulty in capturing the global features of crack. Transformer-based networks can capture global semantic information, but this method also has the deficiencies of strong data dependence and easy loss of local features. In this paper, a hybrid semantic segmentation algorithm for tunnel lining crack, named SCDeepLab, is proposed by fusing Swin Transformer and CNN in the encoding and decoding framework of DeepLabv3+ to address the above issues. In SCDeepLab, a joint backbone network is introduced with CNN-based Inverse Residual Block and Swin Transformer Block. The former is used to extract the local detailed information of the crack to generate the shallow feature layer, whereas the latter is used to extract the global semantic information to obtain the deep feature layer. In addition, Efficient Channel Attention enhanced Feature Fusion Module is proposed to fuse the shallow and deep features to combine the advantages of the two types of features. Furthermore, the strategy of transfer learning is adopted to solve the data dependency of Swin Transformer. The results show that the mean intersection over union (mIoU) and mean pixel accuracy (mPA) of SCDeepLab on the data sets constructed in this paper are 77.41% and 84.42%, respectively, which have higher segmentation accuracy than previous CNN-based and transformer-based semantic segmentation algorithms. 

Hybrid semantic segmentation for tunnel lining cracks based on Swin Transformer and convolutional neural network

During the daily operation of high-speed railways, unexpected events such as geological disasters or equipment failure may lead to railway segment blockages. A long-duration segment blockage usually leads to large-scale train delays, which cause significant inconvenience to passengers.Adjusting train schedules in real time during segment blockages is an important issue for the dispatch and operation of high-speed railways.For a segment blockage of a known time interval, a mixed integer nonlinear programming (MINLP) model for train speed adjustment and train diagram adjustment is constructed.In this model, in addition to the common scheduling strategies involving train cancellation and departure delay, the deceleration strategy is considered. Under the constraints of traffic safety and station capacity, the objective is to minimize the weighted delays of the three strategies.Additionally, the delay time owing to train speed reduction is calculated based on the train kinematic model, and the safe interval of train operation is controlled using the moving block principle.The model established in this paper can be directly simulated using commercial optimization software CPLEX, and the scheduling scheme is guaranteed to be in real time. This paper also discusses the influence of the three adjustment strategies, i.e., train cancellation, departure delay, and deceleration, on the total delay time. Finally, the model is verified using an example of the Beijing-Shanghai high-speed railway. Results show that the model is reasonable and effective. By implementing the model, a reasonable scheduling scheme can be achieved to address practical problems, and the punctuality rate of trains can be improved.

/pdf/moving-block-principle-based-multi-strategy-optimal-14nkn8qco7.pdf

Moving block principle-based multi-strategy optimal scheduling method for trains in case of segment blockages

ABSTRACT To solve slow speed and low accuracy of traditional detection methods of tunnel lining cracks, especially under the complicated situation of tunnel in operation, this work proposed an improved You Only Look Once version X (YOLOX) tunnel lining crack image detection algorithm. First, Mobilenetv3 was used to replace YOLOX’s CSPDarknet network. The Efficient Channel Attention (ECA) module was then added to the enhanced feature extraction network, and the IOU loss function was replaced by the generalised IOU (GIOU) loss function. A tunnel crack image data set was constructed and used to compare the performance of the improved YOLOX algorithm with that of five other algorithms. The improved YOLOX algorithm solves the shortcomings of the other five algorithms. The results showed that the improved YOLOX algorithm had 82.48% F1 score and 87.28% AP value, which is higher than that of the other five algorithms at varying degrees. In addition, the data size of the improved YOLOX model was 51.2 M, which is 75.27% compressed compared to the YOLOX model. The time was 16.52 ms, and the FPS was 60.52 frames/s. Therefore, the proposed improved YOLOX algorithm can realise the high-speed, high-precision, real-time dynamic detection of tunnel lining cracks in complicated environments.

Junjie Zhang

Papers

Automatic detection method of tunnel lining multi‐defects via an enhanced You Only Look Once network

Hybrid semantic segmentation for tunnel lining cracks based on Swin Transformer and convolutional neural network

Moving block principle-based multi-strategy optimal scheduling method for trains in case of segment blockages

Real-time tunnel lining crack detection based on an improved You Only Look Once version X algorithm

Fast detection algorithm for cracks on tunnel linings based on deep semantic segmentation