A surface defect detection method for steel pipe based on improved YOLO

Lili Wang; Chunhe Song; Guangxi Wan; Shijie Cui; Lili Wang; Chunhe Song; Guangxi Wan; Shijie Cui

doi:10.3934/mbe.2024134

Mathematical Biosciences and Engineering

2024, Volume 21, Issue 2: 3016-3036. doi: 10.3934/mbe.2024134

Previous Article Next Article

Research article Special Issues

A surface defect detection method for steel pipe based on improved YOLO

1.
State Key Laboratory of Robotics, Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang 110016, China
2.
Key Laboratory of Networked Control Systems, Chinese Academy of Sciences, Shenyang 110016, China
3.
Institutes for Robotics and Intelligent Manufacturing, Chinese Academy of Sciences, Shenyang 110169, China
4.
College of Information, Liaoning University, Shenyang 110036, China

Academic Editor: Shangce Gao

Received: 05 November 2023 Revised: 14 December 2023 Accepted: 26 December 2023 Published: 30 January 2024

Surface defect detection is of great significance as a tool to ensure the quality of steel pipes. The surface defects of steel pipes are charactered by insufficient texture, high similarity between different types of defects, large size differences, and high proportions of small targets, posing great challenges to defect detection algorithms. To overcome the above issues, we propose a novel steel pipe surface defect detection method based on the YOLO framework. First, for the problem of a low detection rate caused by insufficient texture and high similarity among different types of defects of steel pipes, a new backbone block is proposed. By increasing high-order spatial interaction and enhancing the capture of internal correlations of data features, different feature information for similar defects is extracted, thereby alleviating the false detection rate. Second, to enhance the detection performance for small defects, a new neck block is proposed. By fusing multiple features, the accuracy of steel pipe defect detection is improved. Third, for the problem of a low detection rate causing large size differences in steel pipe surface defects, a novel regression loss function that considers the aspect ratio and scale is proposed, and the focal loss is introduced to further solve the sample imbalance problem in steel pipe defect datasets. The experimental results show that the proposed method can effectively improve the accuracy of steel pipe surface defect detection.
- deep learning,
- defect detection,
- steel pipe,
- X-ray image,
- YOLOv5
Citation: Lili Wang, Chunhe Song, Guangxi Wan, Shijie Cui. A surface defect detection method for steel pipe based on improved YOLO[J]. Mathematical Biosciences and Engineering, 2024, 21(2): 3016-3036. doi: 10.3934/mbe.2024134

Related Papers:

Abstract

Surface defect detection is of great significance as a tool to ensure the quality of steel pipes. The surface defects of steel pipes are charactered by insufficient texture, high similarity between different types of defects, large size differences, and high proportions of small targets, posing great challenges to defect detection algorithms. To overcome the above issues, we propose a novel steel pipe surface defect detection method based on the YOLO framework. First, for the problem of a low detection rate caused by insufficient texture and high similarity among different types of defects of steel pipes, a new backbone block is proposed. By increasing high-order spatial interaction and enhancing the capture of internal correlations of data features, different feature information for similar defects is extracted, thereby alleviating the false detection rate. Second, to enhance the detection performance for small defects, a new neck block is proposed. By fusing multiple features, the accuracy of steel pipe defect detection is improved. Third, for the problem of a low detection rate causing large size differences in steel pipe surface defects, a novel regression loss function that considers the aspect ratio and scale is proposed, and the focal loss is introduced to further solve the sample imbalance problem in steel pipe defect datasets. The experimental results show that the proposed method can effectively improve the accuracy of steel pipe surface defect detection.

References

[1]	M. Wang, C. P. J Cheng, A uniffed convolutional neural network integrated with conditional random ffeld for pipe defect segmentation, Comput. Aided Civ. Inf., 35 (2020), 162–177. https://doi.org/10.1111/mice.12481 doi: 10.1111/mice.12481
[2]	B. Jesica, K. Bartosz, M. Igor, Defects and incompatibilities of pipes manufactured by pilgrim method, New Trends Prod. Eng., 2 (2019), 24–35. https://doi.org/10.2478/ntpe-2019-0069 doi: 10.2478/ntpe-2019-0069
[3]	C. Song, S. Liu, G. Han, P. Zeng, H. Yu, Q. Zheng, Edge-intelligence-based condition monitoring of beam pumping units under heavy noise in industrial internet of things for industry 4.0, IEEE IoT J., 10 (2023), 3037–3046. https://doi.org/10.1109/JIOT.2022.3141382 doi: 10.1109/JIOT.2022.3141382
[4]	S. Liu, C. Song, T. Wu, P. Zeng, A lightweight fault diagnosis method of beam pumping units based on dynamic warping matching and parallel deep network, IEEE Trans. Syst. Man Cybern.: Syst., 2023 (2023), 1–11. https://doi.org/10.1109/TSMC.2023.3328731 doi: 10.1109/TSMC.2023.3328731
[5]	B. Chen, Defects classiffcation of steel tube based on spectrogram and CNN using magnetic flux leakage signals, in 2023 IEEE 3rd International Conference on Information Technology, Big Data and Artificial Intelligence (ICIBA), 3 (2023), 1137–1140. https://doi.org/10.1109/ICIBA56860.2023.10165124
[6]	X. Liu, F. Xue, L. Teng, Surface defect detection based on gradient LBP, in 2018 IEEE 3rd International Conference on Image, Vision and Computing (ICIVC), (2018), 133–137. https://doi.org/10.1109/ICIVC.2018.8492798
[7]	H. Wang, J. Zhang, Y. Tian, H. Chen, H. Sun, K. Liu, A simple guidance template-based defect detection method for strip steel surfaces, IEEE Trans. Ind. Inf., 15 (2019), 2798–2809. https://doi.org/10.1109/TII.2018.2887145 doi: 10.1109/TII.2018.2887145
[8]	S. Ren, K. He, R. Girshick, J. Sun, Faster R-CNN: Towards real-time object detection with region proposal networks, IEEE Trans. Pattern Anal. Mach. Intell., 39 (2017), 1137–1149. https://doi.org/10.1109/TPAMI.2016.2577031 doi: 10.1109/TPAMI.2016.2577031
[9]	J. Redmon, S. Divvala, R. Girshick, A. Farhadi, You only look once: Uniffed, real-time object detection, in 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), (2016), 779–788. https://doi.org/10.1109/CVPR.2016.91
[10]	R. Liu, C. Ren, M. Fu, Z. Chu, J. Guo, Platelet detection based on improved YOLOv3, Cyborg Bionic Syst., 2022 (2022), 1–9. https://doi.org/10.34133/2022/9780569 doi: 10.34133/2022/9780569
[11]	W. Liu, D. Anguelov, D. Erhan, C. Szegedy, S. Reed, C. Fu, SSD: Single shot multibox detector, in Computer Vision–ECCV 2016: 14th European Conference, Amsterdam, The Netherlands, October 11–14, 2016, Proceedings, Part I 14, Springer International Publishing, (2016), 21–37. https://doi.org/10.48550/arXiv.1512.02325
[12]	S. Zhang, L. Wen, X. Bian, Z. Lei, S. Z. Li, Single-shot reffnement neural network for object detection, in 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, (2018), 4203–4212. https://doi.org/10.1109/CVPR.2018.00442
[13]	C. Chen, H. Wang, Y. Chen, Z. Yin, X. Yang, H. Ning, et al., Understanding the brain with attention: A survey of transformers in brain sciences, Brain-X, 1 (2023), e29. https://doi.org/10.1002/brx2.29 doi: 10.1002/brx2.29
[14]	B. Hu, J. Wang, Detection of PCB surface defects with improved Faster-RCNN and feature pyramid network, IEEE Access, 8 (2020), 108335–108345. https://doi.org/10.1109/ACCESS.2020.3001349 doi: 10.1109/ACCESS.2020.3001349
[15]	C. Song, J. Chen, Z. Lu, F. Li, Y. Liu, Steel surface defect detection via deformable convolution and background suppression, IEEE Trans. Instrum. Meas., 72 (2023), 1–9. https://doi.org/10.1109/TIM.2023.3277989 doi: 10.1109/TIM.2023.3277989
[16]	M. Zhang, L. Yin, Solar cell surface defect detection based on improved YOLOv5, IEEE Access, 10 (2022), 80804–80815. https://doi.org/10.1109/ACCESS.2022.3195901 doi: 10.1109/ACCESS.2022.3195901
[17]	J. Hang, H. Sun, X. Yu, A. J. J. Rodríguez-Andina, X. Yang, Surface defect detection in sanitary ceramics based on lightweight object detection network, IEEE Open J. Ind. Electron. Soc., 3 (2022), 473–483. https://doi.org/10.1109/OJIES.2022.3193572 doi: 10.1109/OJIES.2022.3193572
[18]	Y. Tu, Z. Ling, S. Guo, H. Wen, An accurate and real-time surface defects detection method for sawn lumber, IEEE Trans. Instrum. Meas., 70 (2021), 1–11. https://doi.org/10.1109/TIM.2020.3024431 doi: 10.1109/TIM.2020.3024431
[19]	C. Chen, K. Zhou, T. Lu, H. Ning, R. Xiao, Integration-and separation-aware adversarial model for cerebrovascular segmentation from TOF-MRA, Comput. Methods Programs Biomed., 233 (2023), 107475. https://doi.org/10.1016/j.cmpb.2023.107475 doi: 10.1016/j.cmpb.2023.107475
[20]	C. Song, W. Xu, G. Han, P. Zeng, Z. Wang, S. Yu, A cloud edge collaborative intelligence method of insulator string defect detection for power IIoT, IEEE IoT J., 8 (2021), 7510–7520. https://doi.org/10.1109/JIOT.2020.3039226 doi: 10.1109/JIOT.2020.3039226
[21]	G. Wang, C. Zhang, M. Chen, Y. Lin, X. Tan, P. Liang, et al., YOLO-MSAPF: Multiscale alignment fusion with parallel feature filtering model for high accuracy weld defect detection, IEEE Trans. Instrum. Meas., 72 (2023), 1–14. https://doi.org/10.1109/TIM.2023.3302372 doi: 10.1109/TIM.2023.3302372
[22]	Y. Liu, D. Jiang, C. Xu, Y. Sun, G. Jiang, B. Tao, et al., Deep learning based 3D target detection for indoor scenes, Appl. Intell., 53 (2023), 10218–10231. https://doi.org/10.1007/s10489-022-03888-4 doi: 10.1007/s10489-022-03888-4
[23]	D. Hong, L. Gao, Y. Naoto, J. Yao, C. Jocelyn, Q. Du, et al., More diverse means better: multimodal deep learning meets remote-sensing imagery classiffcation. IEEE Trans. Geosci. Remote Sens., 59 (2021), 4340–4354. https://doi.org/10.1109/TGRS.2020.3016820
[24]	D. Hong, B. Zhang, H. Li, Y. Li, J. Yao, C. Li, et al., Cross-city matters: A multimodal remote sensing benchmark dataset for cross-city semantic segmentation using high-resolution domain adaptation networks, Remote Sens. Environ., 299 (2023), 113856. https://doi.org/10.1016/j.rse.2023.113856 doi: 10.1016/j.rse.2023.113856
[25]	C. Li, B. Zhang, D. Hong, J. Yao, C. Jocelyn, LRR-Net: An interpretable deep unfolding network for hyperspectral anomaly detection. IEEE Trans. Geosci. Remote Sens., 61 (2023), 1–12. https://doi.org/10.1109/TGRS.2023.3279834
[26]	X. Wu, D. Hong, C. Jocelyn, Convolutional neural networks for multimodal remote sensing data classiffcation, IEEE Trans. Geosci. Remote Sens., 60 (2022), 1–10. https://doi.org/10.1109/TGRS.2021.3124913 doi: 10.1109/TGRS.2021.3124913
[27]	G. Yang, C. Song, Z. Yang, S. Cui, Bubble detection in photoresist with small samples based on GAN augmentations and modiffed YOLO, Eng. Appl. Artif. Intell., 123 (2023), 106224. https://doi.org/10.1016/j.engappai.2023.106224 doi: 10.1016/j.engappai.2023.106224
[28]	X. Zhu, S. Lyu, X. Wang, Q. Zhao, TPH-YOLOv5: Improved YOLOv5 based on transformer prediction head for object detection on drone-captured scenarios, in Proceedings of the IEEE/CVF International Conference on Computer Vision, (2021), 2778–2788. https://doi.org/10.1109/iccvw54120.2021.00312
[29]	J. Yang, C. Li, P. Zhang, X. Dai, B. Xiao, et al., Focal self-attention for local-global interactions in vision transformers, preprint, arXiv: 2107.00641. https://doi.org/10.48550/arXiv.2107.00641
[30]	M. Liu, Y. Chen, L. He, Y. Zhang, J. Xie, LF-YOLO: A lighter and faster YOLO for weld defect detection of X-ray image, IEEE Sens. J., 23 (2023), 7430–7439. https://doi.org/10.1109/jsen.2023.3247006 doi: 10.1109/jsen.2023.3247006
[31]	S. Liu, Y. Wang, Q. Yu, H. Liu, Z. Peng, CfEAM-YOLOv7: Improved YOLOv7 based on channel expansion and attention mechanism for driver distraction behavior detection, IEEE Access, 10 (2022), 129116–129124. https://doi.org/10.1109/access.2022.3228331 doi: 10.1109/access.2022.3228331
[32]	Z. Ye, Q. Guo, J. Wei, J. Zhang, H. Zhang, L. Bian, et al., Recognition of terminal buds of densely-planted Chinese FFR seedlings using improved YOLOv5 by integrating attention mechanism, Front. Plant Sci., 13 (2022), 991929. https://doi.org/10.3389/fpls.2022.991929 doi: 10.3389/fpls.2022.991929
[33]	W. Liu, G. Ren, R. Yu, S. Guo, J. Zhu, L. Zhang, Image-Adaptive YOLO for object detection in adverse weather conditions, in Proceedings of the AAAI Conference on Artiffcial Intelligence, 36 (2022), 1792–1800. https://doi.org/10.1609/aaai.v36i2.20072
[34]	S. Cheng, Y. Zhu, S. Wu, Deep learning based efffcient ship detection from drone-captured images for maritime surveillance, Ocean Eng., 285 (2023), 115440. https://doi.org/10.2139/ssrn.4386215 doi: 10.2139/ssrn.4386215
[35]	J. Li, J. Gu, Z. Huang, J. Wen, Application research of improved YOLO V3 algorithm in PCB electronic component detection, Appl. Sci., 9 (2019), 3750. https://doi.org/10.3390/app9183750 doi: 10.3390/app9183750
[36]	D. Yang, Y. Cui, Z. Yu, H. Yuan, Deep learning based steel pipe weld defect detection, Appl. Artif. Intell., 35 (2021), 1237–1249. https://doi.org/10.1080/08839514.2021.1975391 doi: 10.1080/08839514.2021.1975391
[37]	D. Hong, Y. Naoto, C. Jocelyn, X. Zhu, An augmented linear mixing model to address spectral variability for hyperspectral unmixing, IEEE Trans. Image Process., 28 (2019), 1923–1938. https://doi.org/10.1109/TIP.2018.2878958 doi: 10.1109/TIP.2018.2878958

Reader Comments

Your name:*

Email:*
© 2024 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)