Automatic Classification and Segmentation of Tunnel Cracks Based on Deep Learning and Visual Explanations

Shijin Feng; Xiaolei Zhang; Yihan Chen; Yong Feng; Yong Zhao

REVIEW 3 major objections 5 minor 24 references

Automatic Classification and Segmentation of Tunnel Cracks Based on Deep Learning and Visual Explanations

T0 review · 3 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read A two-stage deep learning pipeline, using DenseNet-169 for classification and DeepLabV3+ for segmentation, outperforms single-model alternatives on tunnel crack detection, with Score-CAM heatmaps explaining where the segmenter looks.

desk verdict Competent but very conventional two-stage crack detection paper whose 'superiority' claims rest on degenerate classification baselines, within-noise segmentation margins, and an unmeasured end-to-end benefit. read the letter →

arxiv 2507.14010 v1 pith:Q7UQXZQJ submitted 2025-07-18 cs.CV

Yong Feng , Xiaolei Zhang , Shijin Feng , Yong Zhao , Yihan Chen This is my paper

classification cs.CV

keywords TunnelcrackDeeplearningImageclassificationSemanticsegmentationVisualexplanationDenseNet-169LabV3+Score-CAM

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper sets out to make tunnel crack inspection automatic and trustworthy by splitting the problem into two steps. A DenseNet-169 classifier first separates crack images from defect-free tunnel lining images, achieving 92.23% accuracy and 39.80 frames per second on the NUAACrack-2000 dataset, ahead of six CNN and Transformer alternatives. The selected crack images then go into a DeepLabV3+ segmenter, which reaches 57.01% intersection-over-union (a pixel-overlap measure) and 67.44% F1, ahead of four segmentation baselines. The paper also uses Score-CAM heatmaps to show that the segmenter attends globally while encoding and gradually fixes on the crack itself while decoding. If these results hold, the two-stage design is a practical template for fast, explainable screening of tunnel linings.

What carries the argument

The argument runs on three named components. DenseNet-169 is a convolutional classifier whose layers receive feature maps from all preceding layers, maximizing information flow; it acts as the coarse filter that discards defect-free images. DeepLabV3+ is an encoder-decoder segmentation network built on ResNet-101, with atrous spatial pyramid pooling in the encoder using dilation rates 6, 12, and 18 to capture crack context at multiple scales, and a decoder that fuses low-level spatial features with upsampled high-level features; it is trained with dice loss to counter the imbalance between few crack pixels and many background pixels. Score-CAM is the explanation technique: it up-samples and normalizes each channel's activation map, weights the map by how much it changes the model's output relative to a baseline input, and sums the weighted maps into a heatmap. Together, the two-stage design keeps pixel-level segmentation from running on the majority of images, while Score-CAM makes the segmenter's attention inspectable.

What would settle it

A concrete check has two parts. First, rerun the same training and testing protocol on NUAACrack-2000 with multiple random seeds; if DenseNet-169 does not consistently beat the six comparison classifiers, or DeepLabV3+ does not consistently beat the four segmenters, the claimed superiority is not stable. Second, apply the trained models to tunnel lining images from a different tunnel, camera, or lighting condition than NUAACrack-2000; a large drop in accuracy or pixel-overlap would show the numbers do not transfer to practice.

Watch

Extended reading notes

Core claim

The central claim is that tunnel crack detection is best structured as a two-stage pipeline rather than as a single monolithic model. In the first stage, DenseNet-169 classifies each tunnel lining image as crack or defect-free, reaching 92.23% accuracy and 39.80 FPS; the paper reports it beats DenseNet-201, EfficientNet-B0, ResNet-50, ResNet-101, Swin Transformer, and Vision Transformer by 2.07 to 18.65 percentage points. In the second stage, DeepLabV3+ segments crack pixels in the images the classifier kept, reaching 57.01% IoU, 67.44% F1, 62.95% precision, and 82.12% recall, beating DeepLabV3, PSPNet, UNet, and UNet++. The visual-explanation experiments with Score-CAM show DeepLabV3+'s encoder starts with broad attention and its decoder narrows onto the crack, and that high-level fused features encode global context while low-level fused features encode local spatial detail.

Load-bearing premise

The load-bearing premise is that NUAACrack-2000, with its 512×375 tunnel lining images and ground-truth crack labels, represents the variety of lighting, crack widths, backgrounds, and camera conditions found in real tunnel inspections; if it does not, the reported accuracy and pixel-overlap figures will not transfer to practice.

Editorial extensions

If this is right

If the two-stage pipeline is adopted, crack segmentation is applied only to images already flagged as containing cracks, so computation and inspection time are concentrated where they matter.
The reported figures support the practicality of automated screening on ordinary hardware: 39.80 FPS means the classifier can process roughly forty tunnel images per second on the test setup.
DeepLabV3+'s recall of 82.12% against precision of 62.95% matches the engineering priority of not missing cracks, at the cost of extra false crack pixels.
The Score-CAM finding that the model shifts from global to crack-focused attention during decoding offers a concrete way to verify that the segmenter is responding to crack geometry rather than to unrelated image cues.
Because the dense CNN outperformed both Transformer models on this dataset, the paper implies that for small, imbalanced tunnel-image collections, CNN-based architectures remain a strong default.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

Beyond the paper, a natural extension is to test the same two-stage pipeline on images with wet patches, shadows, or cable clutter; NUAACrack-2000 may not contain those, so transfer performance is unverified.
The Score-CAM heatmaps could be turned into a debugging tool: firms deploying crack models could inspect heatmaps on new tunnels to see whether the model is keying on crack-like stains rather than actual fractures, and retrain accordingly.
A multi-class version of the first stage could screen for leakage, spalling, and rebar exposure together, then route each flagged image to the appropriate segmenter; the paper only demonstrates binary crack screening.
The dice-loss choice suggests the authors expect crack pixels to be a small minority; a useful stress test would be measuring how both stages behave on very wide or very faint cracks that shift the pixel-class balance.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Quick take: this is a clean but entirely conventional engineering application of DenseNet-169 for classification and DeepLabV3+ for segmentation on the NUAACrack-2000 tunnel dataset. The abstract's numbers are real, but the paper's central claim—that the two-stage framework outperforms other models and improves detection—is much weaker than the evidence supports.

What's new: the specific DenseNet-169 + DeepLabV3+ + Score-CAM combination has not been reported on this dataset, and the Score-CAM heatmaps inside the DeepLabV3+ decoder are a small useful addition. The model descriptions and Score-CAM equations are correct, and the experimental setup is clearly documented.

The soft spots are exactly where the stress-test note lands. In Table 2, EfficientNet-B0, Swin Transformer, and Vision Transformer all hit 75.65% accuracy, which equals the majority-class baseline (146 of 193 test images are cracks). Those comparators collapsed to predicting the majority class, so the reported 16.58% improvements over them prove nothing about DenseNet-169's superiority. The authors even note that Swin and ViT misclassified every defect-free image, confirming these baselines were not properly trained or tuned. A fair comparison would need hyperparameter tuning or class weighting for the comparators. In Table 3, the IoU margins over UNet (0.17 pp) and UNet++ (0.59 pp) on a 146-image test set are within run-to-run noise, and the paper itself reports lower F1 than UNet and lower recall than both UNet and UNet++. No error bars, no repeated runs, no significance tests.

The bigger structural flaw is that the two-stage benefit is never measured. The classification stage misses 12 of 146 crack images (8.2% false negatives), and those images never reach segmentation. No end-to-end metric is reported. A simple product estimate puts crack-pixel recall near 75% (0.9223 × 0.8212), below the 82.12% of the segmentation model run alone. So the claimed efficiency and accuracy advantage of the two-stage design is asserted, not demonstrated.

What I do credit: the paper acknowledges its own limitations (small dataset, class imbalance, need for preprocessing and lightweight models), and the visual explanation section is honest interpretation work. It is a plausible engineering report, not a research advance.

Bottom line: this deserves a serious referee only if the editor insists on a revised version with fixed baselines, error bars, and a direct end-to-end evaluation. As it stands, I would not cite the quantitative claims in my own work, though the paper could become a useful data point after major revision.

Referee Report

3 major / 5 minor

Summary. The paper proposes a two-stage deep-learning framework for tunnel lining crack detection: DenseNet-169 first classifies each image as crack or defect-free, and DeepLabV3+ then segments cracks in the images classified as containing cracks. Score-CAM is applied to the segmentation model to produce visual explanations. Experiments on the open NUAACrack-2000 dataset report 92.23% classification accuracy and 39.80 FPS for DenseNet-169, and 57.01% IoU / 67.44% F1 for DeepLabV3+. Comparisons are made with six classification models and four segmentation models, and the paper claims that the proposed two-stage method outperforms these alternatives.

Significance. If the claims were established, the work would be of practical interest for tunnel inspection, because it combines image-level screening with pixel-level crack segmentation and adds interpretability through Score-CAM. The manuscript is clearly organized and uses a publicly available dataset, and it compares against a reasonable set of CNN and Transformer baselines. The visual explanations in Section 4.4 provide a useful qualitative check on the segmentation model's behavior. However, as presented, the central claims are not supported by the evidence: key classification baselines collapse to the majority-class baseline, the segmentation margins over UNet and UNet++ are within run-to-run noise, and no end-to-end metric for the two-stage pipeline is reported. The paper's empirical contribution is therefore currently uncertain until these issues are addressed.

major comments (3)

[§4.2.3, Table 2] The claimed classification superiority is not established because several comparators collapse to the majority-class baseline. EfficientNet-B0, Swin Transformer, and Vision Transformer all report 75.65% accuracy, which is exactly the accuracy of always predicting the crack class on the 193-image test set (146/193). The stated improvements of 16.58% over these models therefore compare DenseNet-169 against effectively untrained or degenerate classifiers. The authors should re-tune or re-train these baselines, verify that their training protocols lead to non-degenerate predictions (e.g., report confusion matrices or a majority-class baseline), and then reassess the claimed improvement.
[§4.3.3, Table 3] The segmentation comparison does not support the claim that DeepLabV3+ outperforms other state-of-the-art models. DeepLabV3+ achieves 57.01% IoU versus 56.84% for UNet and 56.42% for UNet++, differences of 0.17 and 0.59 percentage points on a 146-image test set. The paper reports no repeated runs, confidence intervals, or statistical significance tests, and DeepLabV3+ actually has a lower F1 score (67.44 vs. 67.68) and lower recall (82.12 vs. 83.91) than UNet. These margins are within plausible run-to-run variation, so the reported numbers do not demonstrate a reliable advantage. The authors should provide uncertainty estimates, e.g., via multiple training runs with different seeds or a paired bootstrap test on the test set.
[§2.1, §4.3.2] The central two-stage benefit is asserted but never measured end-to-end. The classification stage misclassifies 12 of 146 crack images in the test set, and those images are then excluded from the segmentation evaluation; no metric reports what the full pipeline produces on the entire 193-image test set. The paper also does not compare the end-to-end speed of the two-stage pipeline against a one-stage segmentation model on all images. To support the claim that the two-stage design improves detection accuracy and efficiency, the authors should report an end-to-end measure (e.g., pipeline recall/precision over all test images, combined with a latency/throughput comparison of the whole system).

minor comments (5)

[§2.4, Eqs. (1)–(4)] The equations and surrounding notation are poorly typeset: e.g., "k lA", "H s Up A", "co c S re CAML", and the use of roman and italic letters is inconsistent. The authors should rewrite the Score-CAM equations cleanly, defining all symbols (including X_b, f, and the upsampling operator).
[Abstract and §4.3.3] The abstract claims that the segmentation model's F1 score "outperforms other state-of-the-art models," but Table 3 shows UNet has a higher F1 score (67.68 vs. 67.44). The abstract should be corrected to state the actual pattern of results.
[§3 and §4.3.2] Section 3 states that the original images have a pixel size of 512×375, while Section 4.3.2 states that images were resized to 512×384 before segmentation. This discrepancy, and the reason for the different aspect ratio, should be clarified.
[References] Several references are incomplete or inconsistently formatted. For instance, the Zhou et al. (2023) entry reads "Computer aided Civil Eng mice.13003" instead of a full citation, and the two Feng et al. (2023) entries are not distinguished with year suffixes.
[§4.2.3] The text states that Swin Transformer and Vision Transformer "wrongly classified all defect-free images as crack images," but no confusion matrix is shown for any model. Reporting confusion matrices for all classification models would make the comparison transparent and would clarify the majority-class collapse.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the paper reports empirical measurements from trained models and does not derive predictions from assumptions that encode the answers.

full rationale

This paper is an empirical systems paper, not a derivation chain. The classification accuracy and segmentation IoU are direct measurements on held-out test images after training standard models (DenseNet-169 and DeepLabV3+) on the NUAACrack-2000 dataset. No equation in the paper maps a fitted parameter into a claimed prediction; the reported numbers come from applying the trained models to test data. The Score-CAM visual explanations are used only for post hoc interpretation of the segmentation model and are not claimed to be evidence of correctness. The paper's comparisons with other models, even if flawed because some baselines collapsed to majority-class prediction or IoU differences are small, are empirical comparisons rather than circular reasoning. The self-citations in the introduction provide background motivation about tunnel infrastructure and prior defect-detection work, but they are not load-bearing for the paper's central accuracy claims. The weakest assumption identified by the reader, that NUAACrack-2000 is representative of real tunnel imagery, is a data-quality and generalization concern, not a circularity concern. No step in the manuscript reduces its output to its input by construction, so the appropriate circularity score is 0.

Assumptions & free parameters 6 free parameters · 4 assumptions · 0 invented entities

The paper introduces no new entities, parameters, or mathematical structures. It relies on standard deep learning training assumptions and the correctness of the public dataset. The free parameters listed are hand-chosen training hyperparameters, none of which are fitted to optimize the final reported numbers in a search-based sense, but they influence the results.

free parameters (6)

Classification learning rate schedule = 0.005 initial, decay x0.1 every 10 epochs
Hand-chosen; no sensitivity analysis.
Classification input size = 224x224
Resized from 512x375; standard choice.
Classification batch size = 4
Hand-chosen; no ablation.
Segmentation learning rate schedule = 0.001 for first 50 epochs, 0.0001 for last 50
Hand-chosen; no sensitivity analysis.
Segmentation batch size = 8
Hand-chosen.
Segmentation input size = 512x384
Hand-chosen, close to original 512x375.

assumptions (4)

domain assumption Training, validation, and test splits are independently and identically distributed and representative of the same distribution.
The paper divides the dataset with a 7:2:1 ratio without stratifying by source or imaging conditions; if images from different tunnels are not independent, the reported metrics may be optimistic.
domain assumption Ground-truth crack labels in NUAACrack-2000 are correct and consistent.
The paper relies entirely on these labels for training and evaluation; no label quality check or inter-annotator agreement is reported.
domain assumption Comparison models are trained with equivalent effort and hyperparameter tuning.
The paper reports results for several baselines but does not describe whether each baseline had its own hyperparameter search, so differences may partly reflect tuning effort.
domain assumption Score-CAM heatmaps faithfully reflect the model's decision-making.
The visual explanation is presented as evidence of model behavior without quantitative validation such as perturbation tests or comparison with ground-truth attention.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Automatic Classification and Segmentation of Tunnel Cracks Based on Deep Learning and Visual Explanations." pith.science (2026). https://pith.science/paper/Q7UQXZQJ

@misc{pith2026250714010,
  author       = {Pith},
  title        = {Pith review of: Automatic Classification and Segmentation of Tunnel Cracks Based on Deep Learning and Visual Explanations},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/Q7UQXZQJ}},
  note         = {Machine review of arXiv:2507.14010}
}

read the original abstract

Tunnel lining crack is a crucial indicator of tunnels' safety status. Aiming to classify and segment tunnel cracks with enhanced accuracy and efficiency, this study proposes a two-step deep learning-based method. An automatic tunnel image classification model is developed using the DenseNet-169 in the first step. The proposed crack segmentation model in the second step is based on the DeepLabV3+, whose internal logic is evaluated via a score-weighted visual explanation technique. Proposed method combines tunnel image classification and segmentation together, so that the selected images containing cracks from the first step are segmented in the second step to improve the detection accuracy and efficiency. The superior performances of the two-step method are validated by experiments. The results show that the accuracy and frames per second (FPS) of the tunnel crack classification model are 92.23% and 39.80, respectively, which are higher than other convolutional neural networks (CNN) based and Transformer based models. Also, the intersection over union (IoU) and F1 score of the tunnel crack segmentation model are 57.01% and 67.44%, respectively, outperforming other state-of-the-art models. Moreover, the provided visual explanations in this study are conducive to understanding the "black box" of deep learning-based models. The developed two-stage deep learning-based method integrating visual explanations provides a basis for fast and accurate quantitative assessment of tunnel health status.

Figures

Figures reproduced from arXiv: 2507.14010 by the authors.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

24 extracted references · 15 canonical work pages

[4]

Journal of Intelligent Construction 1, 9180004

Improved SOLOv2 detection method for shield tunnel lining water leakages. Journal of Intelligent Construction 1, 9180004. https://doi.org/10.26599/JIC.2023.9180004 Gao, X., Jian, M., Hu, M., Tanniru, M., Li, S.,

arXiv 2023
[8]

http://arxiv.org/abs/1608.06993 Huang, H., Li, Q., Zhang, D.,

Densely Connected Convolutional Networks. http://arxiv.org/abs/1608.06993 Huang, H., Li, Q., Zhang, D.,

arXiv
[9]

Tunnelling and Underground Space Technology 77, 166–176

Deep learning based image recognition for crack and leakage defects of metro shield tunne l. Tunnelling and Underground Space Technology 77, 166–176. https://doi.org/10.1016/j.tust.2018.04.002 Huang, H., Zhao, S., Zhang, D., Chen, J.,

work page doi:10.1016/j.tust.2018.04.002 2018
[10]

Structure and Infrastructure Engineering 18, 183–196

Deep learning -based instance segmentation of cracks from shield tunnel lining images. Structure and Infrastructure Engineering 18, 183–196. https://doi.org/10.1080/15732479.2020.1838559 Jiang, Y ., Wang, L., Zhang, B., Dai, X., Ye, J., Sun, B., Liu, N., Wang, Z., Zhao, Y .,

arXiv 2020
[11]

Automation in Construction 152, 104881

Tunnel lining detection and retr ofitting. Automation in Construction 152, 104881. https://doi.org/10.1016/j.autcon.2023.104881 Liu, F., Wang, L.,

work page arXiv 2023
[12]

Construction and Building Materials 322, 126265

UNet-based model for crack detection integrating visual explanations. Construction and Building Materials 322, 126265. https://doi.org/10.1016/j.conbuildmat.2021.126265 Liu, J., Zhao, Z., Lv, C., Ding, Y ., Chang, H., Xie, Q.,

work page arXiv 2021
[13]

Construction and Building Materials 348, 128583

An image enhancement algorithm to improve road tunnel crack tr ansfer detection. Construction and Building Materials 348, 128583. https://doi.org/10.1016/j.conbuildmat.2022.128583 Miao, X., Wang, J., Wang, Z., Sui, Q., Gao, Y ., Jiang, P.,

work page arXiv 2022
[14]

IEEE Sensors J

Automatic Recognition of Highway Tunnel Defects Based on an Improved U-Net Model. IEEE Sensors J. 19, 11413–11423. https://doi.org/10.1109/JSEN.2019.2934897 Qiu, J., Yan, X., Wang, J., Guo, Y ., Wei, M.,

arXiv 2019

Show all 24 references

[15]

Computer Engineering and Science 44, 845–854

Crack extraction from single tunnel image based on fully convolutional ne ural network. Computer Engineering and Science 44, 845–854. https://doi.org/10.3969/j.issn.1007-130X.2022.05.010 Ren, Y ., Huang, J., Hong, Z., Lu, W., Yin, J., Zou, L., Shen, X.,

2022 doi
[17]

http://arxiv.org/abs/1910.01279 Xu, X., Yang, H.,

Score -CAM: Score -Weighted Visual Explanations for Convolutional Neural Networks. http://arxiv.org/abs/1910.01279 Xu, X., Yang, H.,

1910 arXiv
[18]

Advances in Mechanical Engineering 11, 168781401987265

Intelligent crack extraction and analysis for tunnel structures with terrestrial laser scanning measu rement. Advances in Mechanical Engineering 11, 168781401987265. https://doi.org/10.1177/1687814019872650 Yang, X., Li, H., Y u, Y ., Luo, X., Huang, T., Yang, Xu,

doi
[19]

Computer -Aided Civil and Infrastructure Engineering 33, 1090–1109

Automatic Pixel -Level Crack Detection an d Measurement Using Fully Convolutional Network: Pixel -level crack detection and measurement using FCN. Computer -Aided Civil and Infrastructure Engineering 33, 1090–1109. https://doi.org/10.1111/mice.12412 Zhang, J., Chu, W., Tu, W.,...

doi
[20]

Computer Vision -based Monitoring Method for Differential Settlement of Shield Tunnels. J. Phys.: Conf. Ser. 2519, 012057. https://doi.org/10.1088/1742-6596/2519/1/012057 Zhang, Y ., Li, X., Qiu, J., Zhai, X., Wei, M.,

doi
[21]

(Eds.), Neural Computing for Advanced Applications, Communications in Computer and Information Science

GFU -Net: A Deep Learning Approach for Automatic Metal Crack Detection, in: Zhang, H., Yang, Z., Zhang, Z., Wu, Z., Hao, T. (Eds.), Neural Computing for Advanced Applications, Communications in Computer and Information Science. Springer Singapore, Singapore, pp. 375–388. https...

doi
[22]

Automation in Construction 132, 103934

A deep learning -based approach for refined crack evaluation from shield tunnel lining images. Automation in Construction 132, 103934. https://doi.org/10.1016/j.autcon.2021.103934 Zhou, Z., Zhang, J., Gong, C.,

2021
[23]

Computer aided Civil Eng mice.13003

Hybrid semantic segmentation for tunnel lining cracks based on Swin Transformer and convolutional neural network. Computer aided Civil Eng mice.13003. https://doi.org/10.1111/mice.13003 Zhou, Z., Zhang, J., Gong, C.,

doi
[24]

Computer aided Civil Eng 37, 762–780

Automatic detection method of tunnel lining multi ‐defects via an enhanced You Only Look Once network. Computer aided Civil Eng 37, 762–780. https://doi.org/10.1111/mice.12836

doi
[2016]

Presented at the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, Las Vegas, NV , USA, pp

Deep Residual Learning for Image Recognition, in: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Presented at the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, Las Vegas, NV , USA, pp. 770–778. https://doi.org/10.1109/CV...

2016 doi
[2018]

http://arxiv.org/abs/1802.02611 Dang, L.M., Wang, H., Li, Y ., Park, Y ., Oh, C., Nguyen, T.N., Moon, H.,

Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation. http://arxiv.org/abs/1802.02611 Dang, L.M., Wang, H., Li, Y ., Park, Y ., Oh, C., Nguyen, T.N., Moon, H.,

arXiv
[2019]

Advances in Structural Engineering 22, 2907–2921

Faster multi-defect detection system in shield tunnel using combination of FCN and faster RCNN. Advances in Structural Engineering 22, 2907–2921. https://doi.org/10.1177/1369433219849829 He, K., Zhang, X., Ren, S., Sun, J.,

doi
[2020]

Construction and Building Materials 234, 117367

Image-based concrete crack detection in tunnels using deep fully convolutional networks. Construction and Building Materials 234, 117367. https://doi.org/10.1016/j.conbuildmat.2019.117367 Wang, H., Wang, Z., Du, M., Yang, F., Zhang, Z., Ding, S., Mardziel, P., Hu, X.,

2019
[2021]

IOP Conf

Image -based crack recognition of tunnel lining using residual U -Net convolutional neural network. IOP Conf. Ser.: Earth Environ. Sci. 861, 072001. https://doi.org/10.1088/1755-1315/861/7/072001 Huang, G., Liu, Z., van der Maaten, L., Weinberger, K.Q.,

doi
[2022]

Tunnelling and Underground Space Technology 124, 104472

Automatic tunnel lining crack evaluation and measurement using deep learning. Tunnelling and Underground Space Technology 124, 104472. https://doi.org/10.1016/j.tust.2022.104472 Feng, S., Feng, Y ., Zhang, X., Chen, Y .,

2022
[2023]

Tunnelling and Underground Space Technology 136, 105107

Deep learning with visual explanations for leakage defect segmentation of metr o shield tunnel. Tunnelling and Underground Space Technology 136, 105107. https://doi.org/10.1016/j.tust.2023.105107 Feng, Y ., Zhang, X., Feng, S., Chen, H., Zhao, Y ., Chen, Y .,

2023

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.

[1] [4]

Journal of Intelligent Construction 1, 9180004

Improved SOLOv2 detection method for shield tunnel lining water leakages. Journal of Intelligent Construction 1, 9180004. https://doi.org/10.26599/JIC.2023.9180004 Gao, X., Jian, M., Hu, M., Tanniru, M., Li, S.,

arXiv 2023

[2] [8]

http://arxiv.org/abs/1608.06993 Huang, H., Li, Q., Zhang, D.,

Densely Connected Convolutional Networks. http://arxiv.org/abs/1608.06993 Huang, H., Li, Q., Zhang, D.,

arXiv

[3] [9]

Tunnelling and Underground Space Technology 77, 166–176

Deep learning based image recognition for crack and leakage defects of metro shield tunne l. Tunnelling and Underground Space Technology 77, 166–176. https://doi.org/10.1016/j.tust.2018.04.002 Huang, H., Zhao, S., Zhang, D., Chen, J.,

work page doi:10.1016/j.tust.2018.04.002 2018

[4] [10]

Structure and Infrastructure Engineering 18, 183–196

Deep learning -based instance segmentation of cracks from shield tunnel lining images. Structure and Infrastructure Engineering 18, 183–196. https://doi.org/10.1080/15732479.2020.1838559 Jiang, Y ., Wang, L., Zhang, B., Dai, X., Ye, J., Sun, B., Liu, N., Wang, Z., Zhao, Y .,

arXiv 2020

[5] [11]

Automation in Construction 152, 104881

Tunnel lining detection and retr ofitting. Automation in Construction 152, 104881. https://doi.org/10.1016/j.autcon.2023.104881 Liu, F., Wang, L.,

work page arXiv 2023

[6] [12]

Construction and Building Materials 322, 126265

UNet-based model for crack detection integrating visual explanations. Construction and Building Materials 322, 126265. https://doi.org/10.1016/j.conbuildmat.2021.126265 Liu, J., Zhao, Z., Lv, C., Ding, Y ., Chang, H., Xie, Q.,

work page arXiv 2021

[7] [13]

Construction and Building Materials 348, 128583

An image enhancement algorithm to improve road tunnel crack tr ansfer detection. Construction and Building Materials 348, 128583. https://doi.org/10.1016/j.conbuildmat.2022.128583 Miao, X., Wang, J., Wang, Z., Sui, Q., Gao, Y ., Jiang, P.,

work page arXiv 2022

[8] [14]

IEEE Sensors J

Automatic Recognition of Highway Tunnel Defects Based on an Improved U-Net Model. IEEE Sensors J. 19, 11413–11423. https://doi.org/10.1109/JSEN.2019.2934897 Qiu, J., Yan, X., Wang, J., Guo, Y ., Wei, M.,

arXiv 2019