基于多域多尺度深度特征自适应融合的焊缝缺陷检测研究

摘要
图/表
参考文献(22)
相关文章 (10)

全文: PDF (3619 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要针对焊缝缺陷检测信号信息丰富度低、深度网络架构人工依赖性强等问题，开展基于多域多尺度深度特征自适应融合的焊缝缺陷检测研究。首先构建时域数据集并衍生至实数域与复数域中，丰富检测信号的特征表达；其次设计多域信息融合模型，充分融合特征域信息；最后提出面向卷积神经网络多维超参数自寻优的模型优化策略，提高模型的效率和性能。试验表明，所提方法对五类焊缝缺陷识别准确率为96.54%，能够在提升识别准确率同时保持较少的参数量和计算消耗，具有较强的实用性和泛化性。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	张睿1
	2
	高美蓉1
	2
	傅留虎1
	2
	张鹏云1
	白晓露1
	赵娜1
	2

关键词 ：焊缝缺陷, 超声检测, 多域多尺度特征融合, 卷积神经网络模型优化策略, 模型自优化

Abstract：In order to solve the problems of low information richness of weld defect detection signal and strong artificial dependence on depth network architecture, the research of weld defect detection based on adaptive fusion of multi-domain and multi-scale depth features is carried out. Firstly, the time-domain data set is constructed and derived to the real domain and complex domain to enrich the feature expression of the detection signal; secondly, a multi-domain information fusion model is designed to fully fuse the feature domain information; finally, a model optimization strategy for convolution neural network multi-dimensional hyperparameter self-optimization is proposed to improve the efficiency and performance of the model. The experimental results show that the accuracy of the proposed method for five types of weld defects is 96.54%. It can improve the recognition accuracy while maintaining a small number of parameters and calculation consumption, and has strong practicability and generalization.

Key words： Weld defect ultrasonic detection multi-domain and multi-scale feature fusion convolutional neural network model optimization strategy; model self-optimization

收稿日期: 2022-05-16 出版日期: 2023-09-15

引用本文:

张睿1,2，高美蓉1,2，傅留虎1,2，张鹏云1，白晓露1，赵娜1,2. 基于多域多尺度深度特征自适应融合的焊缝缺陷检测研究[J]. 振动与冲击, 2023, 42(17): 294-305.
ZHANG Rui1,2, GAO Meirong1,2, FU Liuhu1,2, ZHANG Pengyun1, BAI Xiaolu1, ZHAO Na1,2. Weld defect detection based on adaptive fusion of multi-domain and multi-scale deep features. JOURNAL OF VIBRATION AND SHOCK, 2023, 42(17): 294-305.

链接本文:

http://jvs.sjtu.edu.cn/CN/ 或 http://jvs.sjtu.edu.cn/CN/Y2023/V42/I17/294

[1] 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(01):79-84+118.
WANG Rui, HU Yunlei, LIU Weipeng， et al. Weld Defect Detection based on Edge AI in X-ray Image[J]. Transactions of the China Welding Institution, 2022, 43(01):79-84+118.
[2] Ajmi C, Zapata J, Martinez-Alvarez J J, et al. Using deep learning for defect classification on a small weld X-ray image dataset[J]. Journal of Nondestructive Evaluation, 2020, 39(3):1–13.
[3] 黄焕东,胡利晨,李斌彬,等. 基于区域的快速卷积神经网络的焊缝TOFD检测缺陷识别[J]. 无损检测, 2019, 41(07):12-18.
HUANG Huandong, HU Lichen, LI Binbin, et al.Recognition of Defect in TOFD Image Based on Faster Region Convolutional Neural Networks[J]. Nondestructive Testing, 2019, 41(07):12-18.
[4] 胡宏伟,张婕,彭刚,等.基于LBP-KPCA特征提取的焊缝超声检测缺陷分类方法[J]. 焊接学报, 2019, 40(06):34-39+162.
HU Hongwei, ZHANG Jie, PENG Gang, et al. Defect classification method for ultrasonic inspection of welding seams based on LBP-KPCA feature extraction[J]. Transactions of the China Welding Institution, 2019, 40(06):34-39+162.
[5] Wang X, Guan S, Hua L, et al. Classification of spot-welded joint strength using ultrasonic signal time-frequency features and PSO-SVM method[J]. Ultrasonics, 2019, 91:161-169.
[6] Murta R, Vieira F, Santos V O, et al. Welding Defect Classification from Simulated Ultrasonic Signals[J]. Journal of Nondestructive Evaluation, 2018, 37(3):40.
[7] Silva L C, Filho E, Albuquerque M, et al. Segmented Analysis of Time-of-Flight Diffraction Ultrasound for Flaw Detection in Welded Steel Plates using Extreme Learning Machines[J]. Ultrasonics, 2019, 102(11):106057.
[8] 张训杰,张敏,李贤均.基于二维图像和CNN-BiGRU网络的滚动轴承故障模式识别[J].振动与冲击,2021, 40(23):194-201+207.
ZHANG Xunjie, ZHANG Min, Li Xianjun. Rolling bearing fault mode recognition based on 2D image and CNN-BiGRU[J]. Journal of Vibration and Shock, 2021, 40(23):194-201+207.
[9] Lu L, Wang Z. Encoding Temporal Markov Dynamics in Graph for Time Series Visualization[J]. arXiv preprint, 2016, arXiv:1610.07273.
[10] Meng X J, Qiu S, Wan S, et al. A motor imagery EEG signal classification algorithm based on recurrence plot convolution neural network[J]. Pattern Recognition Letters, 2021, 146(4):134-141.
[11] 夏平,施宇,雷帮军,等. 复小波域混合概率图模型的超声医学图像分割[J]. 自动化学报, 2021, 47(1):185-196.
Xia Ping, Shi Yu, Lei Bangjun, et al. Ultrasound medical image segmentation based on hybrid probabilistic graphical model in complex-wavelet domain[J]. Acta Automatica Sinica, 2021, 47(1):185-196.
[12] 袁彩艳,孙洁娣,温江涛,等. 多域信息融合结合改进残差密集网络的轴承故障诊断[J]. 振动与冲击, 2022, 41(4):200-208.
YUAN Caiyan, SUN Jiedi, WEN Jiangtao, et al. Bearing fault diagnosis based on information fusion and improved residual dense networks[J]. Journal of Vibration and Shock, 2022, 41(4):200-208.
[13] 王子愉,袁春,黎健成. 利用可分离卷积和多级特征的实例分割[J].软件学报, 2019, 30(04):954-961.
WANG Ziyu, YUAN Chun, LI Jiancheng. Instance Segmentation with Separable Convolutions and Multi-levelFeatures[J]. Journal of Software, 2019, 30(04):954-961.
[14] 王亚珅,黄河燕,冯冲,等. 基于注意力机制的概念化句嵌入研究[J]. 自动化学报, 2020, 46(7):1390-1400.
WANG Yashen, HUANG Heyan, FENG chong, et al. Conceptual Sentence Embeddings Based on Attention Mechanism[J]. Acta Automatica Sinica, 2020, 46(7):1390-1400.
[15] Faramarzi A, Heidarinejad M, Stephens B, et al. Equilibrium optimizer: A novel optimization algorithm[J]. Knowledge-Based Systems, 2020, 191:105190.
[16] 李守玉, 何庆, 陈俊. 改进平衡优化器算法在约束优化问题中的应用[J]. 计算机科学与探索, 2021,1-14.
LI Shouyu, HE Qing, CHEN Jun. Application of improved equilibrium optimizer algorithm to constrained optimization problems[J]. Journal of Frontiers of Computer Science & Technology, 2021,1-14.
[17] Peng G A, Fh A, Wkl B, et al. United equilibrium optimizer for solving multimodal image registration-ScienceDirect[J]. Knowledge-Based Systems, 2021, 233(5):107552.
[18] 邢致恺,贾鹤鸣,宋文龙. 基于莱维飞行樽海鞘群优化算法的多阈值图像分割[J].自动化学报, 2021, 47(2):363-377.
XING Zhikai, JIA Heming, SONG Wenlong. Levy Flight Trajectory-based Salp Swarm Algorithm for Multilevel Thresholding Image Segmentation[J]. Acta Automatica Sinica, 2021, 47(2):363-377.
[19] 周鹏,董朝轶,陈晓艳,等. 基于阶梯式Tent混沌和模拟退火的樽海鞘群算法[J]. 电子学报, 2021, 49(9):1724-1735.
ZHOU Peng, DONG Chaoyi, CHEN Xiaoyan, et al. A Salp Swarm Algorithm Based on Stepped Tent Chaos and Simulated Annealing[J]. Acta Electronica Sinica, 2021, 49(9):1724-1735.
[20] Zhu H, Ge W, Liu Z. Deep Learning-Based Classification of Weld Surface Defects[J]. Applied Sciences, 2019, 9(16):3312.
[21] Song K, Yan Y. A noise robust method based on completed local binary patterns for hot-rolled steel strip surface defects[J]. Applied Surface Science, 2013, 285(21):858-864.
[22] Vinyals O, Blundell C, Lillicrap T, et al. Matching Networks for One Shot Learning. arXiv preprint, 2017, arxiv:1606.0404080v2.