Abstract
Surface bughole is a major quality defect of the concrete surface, which has nonnegligible impact on the evaluation of surface quality of fair-faced concrete. However, most existing deep learning methods use image segmentation to detect bugholes or other defects on the surface of formed concrete, there is a lack of a method that can detect bughole on the concrete surface instantly during the pouring process. In addition, the inability to effectively detect small-scale bughole is also an issue that cannot be ignored. This application scenario requires a method with high detection accuracy, fast inference speed and ease of deployment. Aiming at these problems, this paper proposes an improved YOLOv5 network, we propose a detector scale (DS) that is added in the model to detect small size bughole on the fair-faced concrete surface with high accuracy. The concatenations are introduced for the feature transmission during the backbone and head part of the model. This allows for the fusing of low-level and high-level features and improves the perception of the detection model on minor flaws. We also construct a dataset of fair-faced concrete surface bugholes and compare our modified YOLOv5 with the baseline version. Our proposed model has improved mAP@0.5 and mAP@.5:.95 to 89.9% and 65.7%, its performance is superior to YOLOv5, while also retaining good inference speed with approximately 13.4 ms to detect one \(1280 \times 1024\) images on single GPU.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Samuelsson, P.: Voids in concrete surfaces. J. Proc. 67(11), 868–874 (1970)
Lemaire, G., Escadeillas, G., Ringot, E.: Evaluating concrete surfaces using an image analysis process. Constr. Build. Mater. 19(8), 604–611 (2005)
Ozkul, T., Kucuk, I.: Design and optimization of an instrument for measuring bughole rating of concrete surfaces. J. Franklin Inst. 348(7), 1377–1392 (2011)
Da Silva, W., Štemberk, P.: Expert system applied for classifying self-compacting concrete surface finish. Adv. Eng. Softw. 64, 47–61 (2013)
Peterson, K., Carlson, J., Sutter, L., et al.: Methods for threshold optimization for images collected from contrast enhanced concrete surfaces for air-void system characterization. Mater. Charact. 60(7), 710–715 (2009)
Zhu, Z., Brilakis, I.: Machine vision-based concrete surface quality assessment. J. Constr. Eng. Manag. 136(2), 210–218 (2010)
Liu, B., Yang, T.: Image analysis for detection of bugholes on concrete surface. Constr. Build. Mater. 137, 432–440 (2017)
Yoshitake, I., Maeda, T., Hieda, M.: Image analysis for the detection and quantification of concrete bugholes in a tunnel lining. Case Stud. Const. Mater. 8, 116–130 (2018)
Wei, F., Yao, G., Yang, Y., et al.: Instance-level recognition and quantification for concrete surface bughole based on deep learning. Autom. Constr. 107, 102920 (2019)
Wei, F., Shen, L., Xiang, Y., et al.: Deep learning-based automatic detection and evaluation on concrete surface bugholes. CMES-Comput. Model. Eng. Sci. 131(2), 619–637 (2022)
Yao, G., Wei, F., Yang, Y., et al.: Deep-learning-based bughole detection for concrete surface image. In: Advances in Civil Engineering 2019 (2019)
Sun, Y., Yang, Y., Yao, G., et al.: Autonomous crack and bughole detection for concrete surface image based on deep learning. IEEE Access 9, 85709–85720 (2021)
Wei, W., Ding, L., Luo, H., et al.: Automated bughole detection and quality performance assessment of concrete using image processing and deep convolutional neural networks. Constr. Build. Mater. 281, 122576 (2021)
Fan, J., Huo, T., Li, X.: A review of one-stage detection algorithms in autonomous driving. In: 2020 4th CAA International Conference on Vehicular Control and Intelligence, Hangzhou, China, pp. 210–214. IEEE (2020)
Ren, S., He, K., Girshick, R., et al.: Faster R-CNN: towards real-time object detection with region proposal networks. IEEE Trans. Pattern Anal. Mach. Intell. 39(6), 1137–1149 (2017)
Redmon, J., Farhadi, A.: Yolov3: an incremental improvement. arXiv preprint arXiv:1804.02767 (2018)
Author, F.: Yolov4: optimal speed and accuracy of object detection. arXiv preprint arXiv:2004.10934 (2020)
Liu, S., Qi, L., Qin, H., et al.: Path aggregation network for instance segmentation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA, pp. 8759–8768. IEEE (2018)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Li, B., Di, F., Li, Y., Luo, Q., Cao, J., Zhao, Q. (2023). A Bughole Detection Approach for Fair-Faced Concrete Based on Improved YOLOv5. In: Zhang, H., et al. International Conference on Neural Computing for Advanced Applications. NCAA 2023. Communications in Computer and Information Science, vol 1870. Springer, Singapore. https://doi.org/10.1007/978-981-99-5847-4_2
Download citation
DOI: https://doi.org/10.1007/978-981-99-5847-4_2
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-99-5846-7
Online ISBN: 978-981-99-5847-4
eBook Packages: Computer ScienceComputer Science (R0)