Radiographic detection of welding defects based on neural ordinary differential equations

WU Wenyong; CHENG Hongyu; ZHANG Qican

doi:10.12073/j.hjxb.20241031001

TRANSACTIONS OF THE CHINA WELDING INSTITUTION > 2025 > 46(5): 72-80. > DOI: 10.12073/j.hjxb.20241031001

WU Wenyong, CHENG Hongyu, ZHANG Qican. Radiographic detection of welding defects based on neural ordinary differential equations[J]. TRANSACTIONS OF THE CHINA WELDING INSTITUTION, 2025, 46(5): 72-80. DOI: 10.12073/j.hjxb.20241031001

Citation:

PDF (13882 KB)

Radiographic detection of welding defects based on neural ordinary differential equations

1.
College of Electronics and Information Engineering, Sichuan University, Chengdu, 610065, China
2.
Chengdu Institute of Special Equipment Inspection and Testing, Chengdu, 610000, China

More Information

Received Date: October 30, 2024
Available Online: April 26, 2025

Graphical Abstract

Abstract

Abstract

Aiming at the problems of strong subjectivity, high work load, repetitive tasks, and low efficiency in defects evaluation in the process of radiographic testing (RT), a neural memory ordinary differential equations based residual network model was proposed in this paper, for the objective, accurate, and intelligent classification of weld defects. The data set of defect images containing 7 types of defects including crack, lack of fusion, incomplete penetration, concave, undercut, slag inclusion, and porosity was collected to ensure the diversity of defects, and the corresponding image preprocessing and expansion was carried out. The typical artificial neural networks, ResNet18, ResNet34, ResNet50, and ResNet101, were iteratively trained. The ResNet34 model got the highest accuracy and was selected as the backbone network to avoid under-learning or overfitting. Then, the nmODE-ResNet was built to improve the classification performance by exploring the excellent nonlinear mapping ability of the nmODE module. The experimental results show that, compared to ResNet34, nmODE-ResNet can significantly improve the accuracy of defect classification in the seven-classification task on RT images of welds’ defects by 1.56% without increasing the number of parameters, and the performance is comparable to that of qualified inspectors.
- radiographic testing,
- welding defects classification,
- residual network,
- neural memory ordinary differential equation

FullText(HTML)

References (27)

References

[1]	中国机械工程学会焊接学会. 焊接手册. 第1卷, 焊接方法及设备. 第3版(修订本)[M]. 北京: 机械工业出版社, 2016. The Chinese Mechanical Engineering Society Welding Society. Welding Handbook. Volume 1, Welding method & equipment. 3rd edition (Revised edition) [M]. Beijing: China Machine Press, 2016.
[2]	SHRANK C, RADHIKA N, DEEPAK KUMAR N H, et al. A review on welding techniques: properties, characterisations and engineering applications[J]. Advances in Materials and Processing Technologies, 2023, 10(2): 1126 − 1181.
[3]	李家伟. 无损检测手册. 第2版 [M]. 北京: 机械工业出版社, 2012. LI Jiawei. Non-destructive testing manual. 2nd edition [M]. Beijing: China Machine Press, 2012.
[4]	SHARMA K. Analysis of non-destructive testing for improved inspection and maintenance strategies[J]. The e-Journal of Nondestructive Testing, 2023. https://doi.org/10.58286/28287.
[5]	DEEPAK J R, RAJA V K B, SRIKANTH D, et al. Non-destructive testing (NDT) techniques for low carbon steel welded joints: A review and experimental study[J]. Materials Today: Proceedings, 2021, 44: 3732 − 3737. doi: 10.1016/j.matpr.2020.11.578
[6]	WANG D, ZHENG Y, DAI W, et al. Deep network-assisted quality inspection of laser welding on power Battery[J]. Sensors, 2023, 23(21): 8894. doi: 10.3390/s23218894
[7]	RAMÍREZ D P, VEITÍA B D R, ARIOSA P F, et al. Pore segmentation in industrial radiographic images using adaptive thresholding and Morphological analysis[J]. Trends in Agricultural and Environmental Sciences, 2023: e230008. DOI: 10.46420/TAES.e230008.
[8]	PRUNELLA M, SCARDIGNO R M, BUONGIORNO D, et al. Deep learning for automatic vision-based recognition of industrial surface defects: a survey[J]. IEEE Access, 2023, 11: 43370 − 43423. doi: 10.1109/ACCESS.2023.3271748
[9]	王睿, 胡云雷, 刘卫朋, 等. 基于边缘AI的焊缝X射线图像缺陷检测[J]. 焊接学报, 2022, 43(1): 79 − 84. WANG Rui, HU Yunlei, LIU Weipeng, et al. Defect detection of weld X-ray images based on edge AI[J]. Transactions of the China Welding Institution, 2022, 43(1): 79 − 84.
[10]	KUMARESAN S, AULTRIN K S J, KUMAR S S, et al. Transfer learning with CNN for classification of weld defect[J]. IEEE Access, 2021, 9: 95097 − 95108. doi: 10.1109/ACCESS.2021.3093487
[11]	AMARNATH M, SUDHARSHAN N, SRINIVAS P. Automatic detection of defects in welding using deep learning-a systematic review[J]. Materials Today: Proceedings, 2023. https://doi.org/10.1016/j.matpr.2023.03.268.
[12]	PATIL R V, REDDY Y P. Multiform weld joint flaws detection and classification by sagacious artificial neural network technique[J]. The International Journal of Advanced Manufacturing Technology, 2023, 125(1): 913 − 943
[13]	侯文慧. 基于深度学习的焊缝图像缺陷识别方法研究[D]. 合肥: 中国科学技术大学, 2019. HOU Wenhui. Weld defect image recognition method based on the deep learning research [D]. Hefei: University of Science and Technology of China, 2019.
[14]	LIU T, ZHENG P, BAO J. Deep learning-based welding image recognition: A comprehensive review[J]. Journal of Manufacturing Systems, 2023, 68: 601 − 625. doi: 10.1016/j.jmsy.2023.05.026
[15]	HOU W, WEI Y, GUO J, et al. Automatic detection of welding defects using deep neural network[J]. Journal of Physics: Conference Series. IOP Publishing, 2018, 933: 012006.
[16]	樊丁, 胡桉得, 黄健康, 等. 基于改进卷积神经网络的管焊缝X射线图像缺陷识别方法[J]. 焊接学报, 2020, 41(1): 7 − 11. FAN Ding, HU Ande, HUANG Jiankang, et al. Defect recognition method of pipe weld in X-ray image based on improved convolutional neural network[J]. Transactions of the China Welding Institution, 2020, 41(1): 7 − 11.
[17]	PALMA-RAMÍREZ D, ROSS-VEITÍA B D, FONT-ARIOSA P, et al. Deep convolutional neural network for weld defect classification in radiographic images[J]. Heliyon, 2024, 10(9): e30590. doi: 10.1016/j.heliyon.2024.e30590
[18]	SAY D, ZIDI S, QAISAR S M, et al. Automated categorization of multiclass welding defects using the X-ray image augmentation and convolutional neural network[J]. Sensors, 2023, 23(14): 6422. doi: 10.3390/s23146422
[19]	HE K, ZHANG X, REN S, et al. Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016: 770 − 778.
[20]	CHEN R T Q, RUBANOVA Y, BETTENCOURT J, et al. Neural ordinary differential equations[C]//Advances in Neural Information Processing Systems, 32nd Conference on Neural Infornation Processing Systems (NeurIPS 2018), Montréal, Canada, 2018: 6572 − 6583.
[21]	YI Z. nmODE: neural memory ordinary differential equation[J]. Artificial Intelligence Review, 2023, 56(12): 14403 − 14438. doi: 10.1007/s10462-023-10496-2
[22]	NIU H, ZHOU Y, YAN X, et al. On the applications of neural ordinary differential equations in medical image analysis[J]. Artificial Intelligence Review, 2024, 57(9): 236. doi: 10.1007/s10462-024-10894-0
[23]	TAN X, NIU H, HU J, et al. Precise fluence map prediction for intensity-modulated radiation therapy by leveraging nmODE[C]//2023 International Annual Conference on Complex Systems and Intelligent Science (CSIS-IAC). IEEE, 2023: 596 − 602.
[24]	TESHIMA T, TOJO K, IKEDA M, et al. Universal approximation property of neural ordinary differential equations[J]. arXiv preprint arXiv: 2012.02414, 2020. https://doi.org/10.48550/arXiv.2012.02414.
[25]	CHEN Y, XU K, ZHOU P, et al. Improved cross entropy loss for noisy labels in vision leaf disease classification[J]. IET Image Processing, 2022, 16(6): 1511 − 1519. doi: 10.1049/ipr2.12402
[26]	HU J, YU C, YI Z, et al. Enhancing robustness of medical image segmentation model with neural memory ordinary differential equation[J]. International Journal of Neural Systems, 2023, 33(12): 2350060. doi: 10.1142/S0129065723500600
[27]	SELVARAJU R R, COGSWELL M, DAS A, et al. Grad-CAM: visual explanations from deep networks via gradient-based localization[J]. International Journal of Computer Vision, 2020, 128: 336 − 359. doi: 10.1007/s11263-019-01228-7

[1]	HONG Yuxiang, YING Qiluo, LIN Kai, WANG Kaiming, WANG Yaoqi. Arc welding molten pool image recognition based on attention mechanism and transfer learning[J]. TRANSACTIONS OF THE CHINA WELDING INSTITUTION, 2025, 46(4): 94-102. DOI: 10.12073/j.hjxb.20240112003
[2]	LI Chengwen, JI Haibiao, YAN Zhaohui, LIU Zhihong, MA Jianguo, WANG Rui, WU Jiefeng. Prediction of residual stress and deformation of 316L multi-layer multi-pass welding based on GA-BP neural network[J]. TRANSACTIONS OF THE CHINA WELDING INSTITUTION, 2024, 45(5): 20-28. DOI: 10.12073/j.hjxb.20230520002
[3]	CHEN Chen, ZHOU Fangzheng, LI Chenglong, LIU Xinfeng, JIA Chuanbao, XU Yao. Prediction method of plasma arc welding molten pool melting state based on spatial and channel characteristics[J]. TRANSACTIONS OF THE CHINA WELDING INSTITUTION, 2023, 44(4): 30-38. DOI: 10.12073/j.hjxb.20220516001
[4]	GAO Changlin, SONG Yanli, ZUO Hongzhou, ZHANG Cheng. Cause diagnosis of welding defects based on adaptive PSO-BP neural network with dynamic weighting[J]. TRANSACTIONS OF THE CHINA WELDING INSTITUTION, 2022, 43(1): 98-106. DOI: 10.12073/j.hjxb.20210515001
[5]	FAN Ding, HU Ande, HUANG Jiankang, XU Zhenya, XU Xu. X-ray image defect recognition method for pipe weld based on improved convolutional neural network[J]. TRANSACTIONS OF THE CHINA WELDING INSTITUTION, 2020, 41(1): 7-11. DOI: 10.12073/j.hjxb.20190703002
[6]	CHEN Yuquan, GAO Xiangdong. Neural network compensation for micro-gap weld detection by magneto-optical imaging[J]. TRANSACTIONS OF THE CHINA WELDING INSTITUTION, 2016, 37(10): 33-36.
[7]	CHEN Zhenhua, SHI Yaowu, ZHAO Haiyan. Ultrasonic testing of spot weld based on spectrum analysis and artificial neural network[J]. TRANSACTIONS OF THE CHINA WELDING INSTITUTION, 2009, (10): 76-80.
[8]	DI Xinjie, LI Wushen, BAI Shiwu, LIU Fangming. Metal magnetic memory signal recognition by neural network for welding crack[J]. TRANSACTIONS OF THE CHINA WELDING INSTITUTION, 2008, (3): 13-16.
[9]	SHI Yu, FAN Ding, CHEN Jian-hong. Predication of properties of welded joints based on neural network[J]. TRANSACTIONS OF THE CHINA WELDING INSTITUTION, 2004, (2): 73-76.
[10]	PENG Pai, WU Lin, TIAN JIN-Song, WANG Xue-feng, FENG Ying-jun. Application of Neural Networks in Welding Parameters's Planning of Robots[J]. TRANSACTIONS OF THE CHINA WELDING INSTITUTION, 2001, (4): 39-42.

Cited By

Cited by

Periodical cited type(7)

1.	高东，李永利，邓颖，周好斌. 旁路耦合电弧TIG焊原理及工艺研究. 热加工工艺. 2025(01): 65-69 .
2.	孟美情，韩俭，朱瀚钊，梁哲滔，蔡养川，张欣，田银宝. 基于多丝电弧增材制造研究现状. 材料工程. 2025(05): 46-62 .
3.	王梦真，万占东，林健. 电弧增材制造工艺及数值仿真研究进展. 大型铸锻件. 2024(01): 7-12 .
4.	李博洋，巴现礼，陈帅帅，徐国敏，刘黎明. 不同路径下的低碳钢三丝间接电弧增材制造组织与性能. 焊接技术. 2024(10): 1-6+145 .
5.	张加恒，黄祎，郭顺，杨东青，闫德俊，李东，王克鸿. 超音频MIG辅助三丝电弧增材制造工艺研究. 电焊机. 2023(02): 104-110 .
6.	吴涛，谭振，王立伟，梁志敏，汪殿龙. 异质双丝间接电弧增材制造Al-Mg-Cu合金组织与力学性能. 焊接学报. 2023(10): 64-70+136 . 本站查看
7.	朱强，姚屏，许斯帆，许可昱. 316L不锈钢电弧增材制造工艺研究. 精密成形工程. 2023(11): 164-170 .

Other cited types(1)

Get Citation

PDF

XML

Article views (34) PDF downloads (7) Cited by(8)

Turn off MathJax

Article Contents

Abstract

References

Radiographic detection of welding defects based on neural ordinary differential equations