[1]牛顿,林宁,林振超,等.多特征融合的焊缝图像多标签分类算法[J].华侨大学学报(自然科学版),2024,45(4):514-523.[doi:10.11830/ISSN.1000-5013.202403033]
 NIU Dun,LIN Ning,LIN Zhenchao,et al.Weld Image Multi-Label Classification Algorithm Based on Multi-Feature Fusion[J].Journal of Huaqiao University(Natural Science),2024,45(4):514-523.[doi:10.11830/ISSN.1000-5013.202403033]
点击复制

多特征融合的焊缝图像多标签分类算法()
分享到:

《华侨大学学报(自然科学版)》[ISSN:1000-5013/CN:35-1079/N]

卷:
第45卷
期数:
2024年第4期
页码:
514-523
栏目:
出版日期:
2024-07-20

文章信息/Info

Title:
Weld Image Multi-Label Classification Algorithm Based on Multi-Feature Fusion
文章编号:
1000-5013(2024)04-0514-10
作者:
牛顿1 林宁2 林振超2 黄凯2 王合佳1 郑力新1
1. 华侨大学 工学院, 福建 泉州 362021;2. 福建省特种设备检验研究院, 福建 泉州 362021
Author(s):
NIU Dun1 LIN Ning2 LIN Zhenchao2 HUANG Kai2 WANG Hejia1 ZHENG Lixin1
1. College of Engineering, Huaqiao University, Quanzhou 362021, China; 2. Fujian Special Equipment Inspection and Research Institute, Quanzhou 362021, China
关键词:
多标签分类 全局相关性 图像特征 图结构特征 特征融合
Keywords:
multi-label classification global relevance image feature graph structure feature feature fusion
分类号:
TP391.41;TU229
DOI:
10.11830/ISSN.1000-5013.202403033
文献标志码:
A
摘要:
为了实现焊缝缺陷的准确分类,提出一种多特征融合的焊缝图像多标签分类算法。首先,通过残差神经网络(ResNet-50)提取图像的特征信息,根据得到的特征图构建图结构,提出关联度引导邻域传播(RDGNP)算法优化图结构;然后,使用图卷积神经网络(GCN)提取图结构的特征信息,并设计特征融合模块实现图像特征和图结构特征的结合;最后,得到多标签分类结果。实验结果表明:文中算法能够有效地实现焊缝缺陷的多标签分类,在X射线焊缝缺陷数据集上的性能有明显提升。
Abstract:
In order to achieve accurate classification of welding defects, a weld image multi-label classification algorithm based on multi-feature fusion is proposed. Firstly, feature information of images is extracted by a residual neural network(ResNet-50), and the graph structure is constructed based on the obtained feature maps. An algorithm named relation degree guided neighborhood propagation(RDGNP)is proposed to optimize the graph structure. Then, the feature information of the graph structure is extracted using graph convolutional neural network(GCN), and a feature fusion module is designed to achieve the combination of image features and graph structure features. Finally, multi-label classification results are obtained. Experimental results show that the proposed method can effectively realize the multi-label classification of welding defects, and its performance on the X-ray welding defects dataset is significantly improved.

参考文献/References:

[1] DING Kai,NIU Zhangqi,HUI Jizhuang,et al.A weld surface defect recognition method based on improved MobileNetV2 algorithm[J].Mathematics,2022,10(19):3678.DOI:10.3390/math10193678.
[2] XU Hao,YAN Zhihong,JI Bowen,et al.Defect detection in welding radiographic images based on semantic segmentation methods[J].Measurement,2022,188:110569.DOI:10.1016/j.measurement.2021.110569.
[3] 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.
[4] KUMARESAN S,AULTRIN K S J,KUMAR S S,et al.Deep learning-based weld defect classification using VGG16 transfer learning adaptive fine-tuning[J].International Journal on Interactive Design and Manufacturing,2023,17(6):2999-3010.DOI:10.1007/s12008-023-01327-3.
[5] TOTINO B,SPAGNOLO F,PERRI S.RIAWELC: A novel dataset of radiographic images for automatic weld defects classification[J].International Journal of Electrical and Computer Engineering Research,2023,3(1):13-17.DOI:10.53375/ijecer.2023.320.
[6] 张智慧,林耀进,张小清,等.基于类别一致性的层次特征选择算法[J].闽南师范大学学报(自然科学版),2022,35(4):41-49.DOI:10.16007/j.cnki.issn2095-7122.2022.04.007.
[7] BRUNA J,ZAREMBA W,SZLAM A,et al.Spectral networks and locally connected networks on graphs[C]//International Conference on Learning Representations.Banff:[s.n.],2014:1-14.DOI:10.48550/arXiv.1312.6203.
[8] KIPF T N,WELLING M.Semi-supervised classification with graph convolutional networks[C]//International Conference on Learning Representations.Toulon:[s.n.],2017:1-14.DOI:10.48550/arXiv.1609.02907.
[9] WANG Yucheng,GAO Liang,GAO Yiping,et al.A graph guided convolutional neural network for surface defect recognition[J].IEEE Transactions on Automation Science and Engineering,2022,19(3):1392-1404.DOI:10.1109/tase.2022.3140784.
[10] BALCIOGLU Y S,SEZEN B,?ERASI C C,et al.Machine design automation model for metal production defect recognition with deep graph convolutional neural network[J].Electronics,2023,12(4):825.DOI:10.3390/electronics12040825.
[11] 周忠眉,孟威.多角度标签结构和特征融合的多标签特征选择[J].闽南师范大学学报(自然科学版),2021,34(1):64-71.DOI:10.16007/j.cnki.issn2095-7122.2021.01.011.
[12] HE Kaiming,ZHANG Xiangyu,REN Shaoqing,et al.Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Las Vegas:IEEE Press,2016:770-778.DOI:10.1109/cvpr.2016.90.
[13] GAO Shanghua,CHENG Mingming,ZHAO Kai,et al.Res2Net: A new multi-scale backbone architecture[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2021,43(2):652-662.DOI:10.1109/TPAMI.2019.2938758.
[14] HAJEBI K,ABBASI-YADKORI Y,SHAHBAZI H,et al.Fast approximate nearest-neighbor search with k-nearest neighbor graph[C]//Proceedings of the Twenty-Second International Joint Conference on Artificial Intelligence.Barcelona:AAAI Press,2011:1312-1317.DOI:10.5591/978-1-57735-516-8/IJCAI11-222.
[15] CHEEMA M A,LIN Xuemin,ZHANG Wenjie,et al.Influence zone: Efficiently processing reverse k nearest neighbors queries[C]//IEEE 27th International Conference on Data Engineering.Washington D C:IEEE Press,2011:577-588.DOI:10.1109/ICDE.2011.5767904.
[16] LIU Yongli,ZHAO Congcong,CHAO Hao.Density peak clustering based on relative density under progressive allocation strategy[J].Mathematical and Computational Applications,2022,27(5):84.DOI:10.3390/mca27050084.
[17] MERY D,RIFFO V,ZSCHERPEL U,et al.GDXray: The database of X-ray images for nondestructive testing[J].Journal of Nondestructive Evaluation,2015,34(4):42.DOI:10.1007/s10921-015-0315-7.
[18] 全国锅炉压力容器标准化技术委员会.承压设备无损检测: NB/T 47013.1-2015[S].北京:新华出版社,2015.
[19] IANDOLA F N,HAN Song,MOSKEWICZ M W,et al.SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5 MB model size[C]//International Conference on Learning Representations.Toulon:[s.n.],2017:1-13.DOI:10.48550/arXiv.1602.07360.
[20] SIMONYAN K,ZISSERMAN A.Very deep convolutional networks for large-scale image recognition[EB/OL].(2015-04-10)[2024-02-10] .https://doi.org/10.48550/arXiv.1409.1556.
[21] LI Jiafeng,WEN Ying,HE Lianghua.SCConv: Spatial and channel reconstruction convolution for feature redundancy[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.Vancouver:IEEE Press,2023:6153-6162.DOI:10.1109/cvpr52729.2023.00596.
[22] CAI Zhicheng,DING Xiaohan,SHEN Qiu,et al.Refconv: Re-parameterized refocusing convolution for powerful convnets[C]//International Conference on Learning Representations.Vienna:[s.n.],2024:1-17.DOI:10.48550/arXiv.2310.10563.
[23] HUANG Xun,BELONGIE S.Arbitrary style transfer in real-time with adaptive instance normalization[C]//Proceedings of the IEEE International Conference on Computer Vision.Venice:IEEE Press,2017:1501-1510.DOI:10.1109/iccv.2017.167.
[24] CHEN Zhaoming,WEI Xiushen,WANG Peng,et al.Multi-label image recognition with graph convolutional networks[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.Long Beach:IEEE Press,2019:5177-5186.DOI:10.1109/cvpr.2019.00532.
[25] WANG Yangtao,XIE Yanzhao,LIU Yu,et al.Fast graph convolution network based multi-label image recognition via cross-modal fusion[C]//Proceedings of the 29th ACM International Conference on Information & Knowledge Management.New York:ACM,2020:1575-1584.DOI:10.1145/3340531.3411880.
[26] LI Yaning,YANG Liu.More correlations better performance: Fully associative networks for multi-label image classification[C]//25th International Conference on Pattern Recognition.Milan:IEEE Press,2021:9437-9444.DOI:10.1109/icpr48806.2021.9412004.
[27] WANG Yangtao,XIE Yanzhao,ZENG Jiangfeng,et al.Cross-modal fusion for multi-label image classification with attention mechanism[J].Computers and Electrical Engineering,2022,101:108002.DOI:10.1016/j.compeleceng.2022.108002.
[28] PANG Wenkai,TAN Zhi.A steel surface defect detection model based on graph neural networks[J].Measurement Science and Technology,2024,35(4):046201.DOI:10.1088/1361-6501/ad1c4b.

备注/Memo

备注/Memo:
收稿日期: 2024-03-20
通信作者: 郑力新(1967-),男,教授,博士,主要从事图像分析、机器视觉、深度学习方法的研究。E-mail:zlx@hqu.edu.cn。
基金项目: 福建省科技计划项目(2020Y0039); 福建省泉州市科技计划项目(2020C042R)https://hdxb.hqu.edu.cn/
更新日期/Last Update: 2024-07-20