Performance Comparison of Multiple Convolutional Neural Networks for Concrete Defects Classification
Open accessThe DOI matches its Crossref record; 43 of the 43 references it cites were found in a registry or an index.
Abstract
Periodical vision-based inspection is a principal form of structural health monitoring (SHM) technique. Over the last decades, vision-based artificial intelligence (AI) has successfully facilitated an effortless inspection system owing to its exceptional ability of accuracy of defects’ pattern recognition. However, most deep learning (DL)-based methods detect one specific type of defect, whereas DL has a high proficiency in multiple object detection. This study developed a dataset of two types of defects, i.e., concrete crack and spalling, and applied various pre-built convolutional neural network (CNN) models, i.e., VGG-19, ResNet-50, InceptionV3, Xception, and MobileNetV2 to classify these concrete defects. The dataset developed for this study has one of the largest collections of original images of concrete crack and spalling and avoided the augmentation process to replicate a more real-world condition, which makes the dataset one of a kind. Moreover, a detailed sensitivity analysis of hyper-parameters (i.e., optimizers, learning rate) was conducted to compare the classification models’ performance and identify the optimal image classification condition for the best-performed CNN model. After analyzing all the models, InceptionV3 outperformed all the other models with an accuracy of 91%, precision of 83%, and recall of 100%. The InceptionV3 model performed best with optimizer stochastic gradient descent (SGD) and a learning rate of 0.001.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Figures











Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Arafin et al. [132] compared the performance of multiple CNN networks, namely VGG-19, ResNet-50, InceptionV3, Xception, and MobileNetV2. The accuracy, precision, and recall rates of the InceptionV3 model were higher than those of other models.”
From Deep Learning for Structural Health Monitoring: Data, Algorithms, Applications, Challenges, and Trends (2023) · cites this paper as [132]
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“DL-based SHM mainly uses structural damage images to classify or segment surface defects to achieve crack detection [130,131,132], bolt loosening detection [47], steel bar exposure detection [49,133], or vehicle recognition through vehicle images captured by road cameras [134,135].”
From Deep Learning for Structural Health Monitoring: Data, Algorithms, Applications, Challenges, and Trends (2023) · cites this paper as [132]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474 · cited 12×
“For example, while the VGG-19, followed by ResNet-50, has the lowest trainable layers, they have shown acceptable performance prospects with their unique architecture.”
BackgroundMethodInfluentialGoing deeper with convolutions
Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., & Rabinovich, A. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298594 · Cited by 48,562
Object Detection With Deep Learning: A Review
Zhao, Z.-Q., Zheng, P., Xu, S.-T., & Wu, X. (2019) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2018.2876865 · Cited by 5,545 · cited 1×
“CNNs have shown efficacy for defect detection purposes [15].”
BackgroundBoosting for transfer learning
Dai, W., Yang, Q., Xue, G.-R., & Yu, Y. (2007) · Proceedings of the 24th international conference on Machine learning · doi:10.1145/1273496.1273521 · Cited by 1,724 · cited 1×
“To consider the pre-trained models for new tasks, usually the original model should have a certain amount of better generalization adaptability to perform satisfactorily with new unseen data [36].”
BackgroundAutomatic Pixel‐Level Crack Detection and Measurement Using Fully Convolutional Network
Yang, X., Li, H., Yu, Y., Luo, X., Huang, T., & Yang, X. (2018) · Computer-Aided Civil and Infrastructure Engineering · doi:10.1111/mice.12412 · Cited by 852 · cited 1×
“For instance, most of the researchers [2,32,33] have considered an 80–20% train–test split ratio for their CNN models.”
BackgroundAzimi, M., Eslamlou, A., & Pekcan, G. (2020) · Sensors · doi:10.3390/s20102778 · Cited by 769 · cited 1×
“Moreover, DL can analyze a large amount of data simultaneously with multiple categorizations, which facilitated damage evaluation in the SHM system [14].”
BackgroundStructural Damage Detection with Automatic Feature‐Extraction through Deep Learning
Lin, Y., Nie, Z., & Ma, H. (2017) · Computer-Aided Civil and Infrastructure Engineering · doi:10.1111/mice.12313 · Cited by 634 · cited 1×
“A CNN-based DL method was developed to detect concrete cracks by [16,17] and other defects by [18].”
MethodTrain faster, generalize better: stability of stochastic gradient descent
Hardt, M., Recht, B., & Singer, Y. (2016) · International Conference on Machine Learning · Cited by 541
The Marginal Value of Adaptive Gradient Methods in Machine Learning
Wilson, A., Roelofs, R., Stern, M., Srebro, N., & Recht, B. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1705.08292 · Cited by 522 · cited 1×
“The authors of [45] studied experimental and empirical analysis to prove that for classification tasks, SGD converged better than other adaptive methods.”
BackgroundSDDNet: Real-Time Crack Segmentation
Choi, W., & Cha, Y.-J. (2020) · IEEE Transactions on Industrial Electronics · doi:10.1109/tie.2019.2945265 · Cited by 462 · cited 2×
“[16] tested a CNN model with a dataset of targeted and noise-free images.”
Method
Research works printed but not linked: 2
- Simonyan, K., and Zisserman, A. (2014). Very deep convolutional networks for large-scale image recognition. arXiv.
- Goodfellow, I., Bengio, Y., and Courville, A. (2016). Deep Learning, MIT Press.
Cited by
Jia, J., & Li, Y. (2023) · Sensors · doi:10.3390/s23218824 · Cited by 182
Matarneh, S. T., Elghaish, F., Rahimian, F. P., Abdellatef, E., & Abrishami, S. (2024) · Automation in Construction · doi:10.1016/j.autcon.2024.105297 · Cited by 91
Pitakaso, R., Srichok, T., Khonjun, S., Golińska-Dawson, P., Gonwirat, S., Nanthasamroeng, N., Boonmee, C., Jirasirilerd, G., & Luesak, P. (2024) · Waste Management · doi:10.1016/j.wasman.2024.05.002 · Cited by 33
Pan, X., Yang, T. Y., Li, J., Ventura, C. E., Málaga‐Chuquitaype, C., Li, C., Su, R. K. L., & Brzev, S. (2025) · Archives of Computational Methods in Engineering · doi:10.1007/s11831-025-10279-8 · Cited by 52
A Review and Benchmark on State-of-the-Art Steel Defects Detection
Chazhoor, A. A. P., Ho, E. S. L., Gao, B., & Woo, W. L. (2023) · SN Computer Science · doi:10.1007/s42979-023-02436-2 · Cited by 41
“Lastly, our choice of ResNet-50 aligns with established practices in the field, facilitating comparisons with other studies and promoting the reproducibility of our work [48].”
Method
Metadata
- Authors
- Arafin, P., Issa, A., & Billah, A. H. M. M.
- Affiliations
- Lakehead University (CA)
- United Arab Emirates University (AE)
- University of Calgary (CA)
- Corresponding author
- Issa, Anas
- Billah, A. H. M. Muntasir
- Venue
- Sensors
- Publisher
- MDPI AG
- Published
- 2022-11-11
- Volume
- 22
- Issue
- 22
- Pages
- 8714
- ISSN
- 1424-8220 (ISSN-L)
- PMC
- PMC9695848
- OpenAlex
- W4309012905
- Cited by
- 34
- References
- 43
- Citations by year
- 2026: 9 · 2025: 10 · 2024: 8 · 2023: 6 · 2022: 1
- FWCI
- 2.81
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Sensors (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Funders
- NSERC (National Sciences and Engineering Research Council of Canada) (ALLRP 555258-20)
- Topics
- Physical Sciences › Engineering › Civil and Structural Engineering › Infrastructure Maintenance and Monitoring
- Physical Sciences › Engineering › Industrial and Manufacturing Engineering › Industrial Vision Systems and Defect Detection
- Physical Sciences › Engineering › Civil and Structural Engineering › Structural Health Monitoring Techniques
- Keywords
- concrete cracking, concrete spalling, structural health monitoring, image-based inspection, convolutional neural networks, InceptionV3, VGG19, ResNet50, MobileNetV2, Xception, hyperparameter sensitivity analysis, learning rate, optimizer comparison
- MeSH terms
- Artificial Intelligence, Neural Networks, Computer
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Performance Comparison of Multiple Convolutional Neural Networks for Concrete Defects Classification. https://app.cochat.ai/papers/10.3390/s22228714
- The paper
- Arafin, P., Issa, A., & Billah, A. H. M. M. (2022). Performance Comparison of Multiple Convolutional Neural Networks for Concrete Defects Classification. Sensors. https://doi.org/10.3390/s22228714
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.