[1]王合佳,林宁,林振超,等.改进YOLO的X射线管道焊缝检测算法[J].华侨大学学报(自然科学版),2024,45(6):766-775.[doi:10.11830/ISSN.1000-5013.202403040]
　WANG Hejia,LIN Ning,LIN Zhenchao,et al.X-Ray Pipe Weld Detection Algorithm of Improved YOLO[J].Journal of Huaqiao University(Natural Science),2024,45(6):766-775.[doi:10.11830/ISSN.1000-5013.202403040]
点击复制

改进YOLO的X射线管道焊缝检测算法()
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6期
页码:
766-775
栏目:
出版日期:
2024-11-15

文章信息/Info

Title:
X-Ray Pipe Weld Detection Algorithm of Improved YOLO
文章编号:
1000-5013(2024)06-0766-10
作者:
王合佳1;  林宁2;  林振超2;  黄凯2;  牛顿2;  郑力新1
1. 华侨大学 工学院, 福建 泉州 362021;2. 华侨大学 福建省特种设备检验研究院泉州分院, 福建 泉州 362021
Author(s):
WANG Hejia1;  LIN Ning2;  LIN Zhenchao2;  HUANG Kai2;  NIU Dun2;  ZHENG Lixin1
1. College of Engineering, Huaqiao University, Quanzhou 362021, China; 2. Quanzhou Branch of Special Equipment Inspection Research Institue, Huaqiao University, Quanzhou 362021, China
关键词:
焊接缺陷;  缺陷检测;  MLCA模块;  YOLOv8算法;  检测帧率
Keywords:
weld defect;  defect detection;  mixed local channel attention module;  YOLOv8 algorithm;  detection frame rate
分类号:
TP391.41;TU229
DOI:
10.11830/ISSN.1000-5013.202403040
文献标志码:
A
摘要:
提出一种基于YOLOv8n算法改进的YOLOv8n-MG算法,用于解决目标小、遮挡重叠、算法参数量大等问题。首先,引入GSConv和VoV-GSCSP模块降低算法复杂度,提高算法对缺陷粗糙边缘的检测能力;其次,使用轻量级的上采样算子Carafe替换原有的传统上采样,保留更多的细节特征;最后,引入混合局部通道注意力(MLCA)机制,以较低的计算成本和参数量保留更多的空间特征信息,并利用Adam优化器提高算法对复杂参数空间的学习能力。结果表明:优化后的算法与YOLOv8n算法相比,参数量减少了11.3%,检测帧率提高了7.7%,平均精度提高了2.8%。
Abstract:
An improved YOLOv8n-MG algorithm based on the YOLOv8n algorithm is proposed to solve issues such as the small targets, overlapping occlusion, and large number of algorithm parameters, etc. Firstly, the GSConv and VoV-GSCSP modules are introduced to reduce the complexity of the the algorithm and enhance its ability to detect rough edges of defects. Secondly, a lightweight up-sampling operator, Carafe, is used to replace the traditional up-sampling, preserving more detailed features. Finally, a mixed local channel attention(MLCA)mechanism is introduced to retain more spatial feature information with lower computational cost and parameters, and the Adam optimizer is used to improve the algorithm’s learning ability in complex parameter spaces. The results show that compared with the YOLOv8n algorithm, the optimized algorithm reduces the number of parameters by 11.3%, improves the detection frame rate by 7.7%, and improves the average accuracy by 2.8%.

参考文献/References:

[1] 高杨.长输管道自动焊接设备及技术发展探究[J].石化技术,2022,29(12):219-221.
[2] BARSOUM Z,JONSSON B.Influence of weld quality on the fatigue strength in seam welds[J].Engineering Failure Analysis,2011,18(3):971-979.DOI:10.1016/j.engfailanal.2010.12.001.
[3] GIRSHICK R,DONAHUE J,DARRELL T,et al.Region-based convolutional networks for accurate object detection and segmentation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2015,38(1):142-158.DOI:10.1109/TPAMI.2015.2437384.
[4] GAVRILESCU R,ZET C,FOAL U C,et al.Faster R-CNN: An approach to real-time object detection[C]//2018 International Conference and Exposition on Electrical and Power Engineering.Lasi:IEEE Press,2018:0165-0168.DOI:10.1109/ICEPE.2018.8559776.
[5] KONG Tao,YAO Anbang,CHEN Yurong,et al.Hypernet: Towards accurate region proposal generation and joint object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Las Vegas:IEEE Press,2016:845-853.DOI:10.1109/CVPR.2016.98.
[6] REDMON J,DIVVALA S,GIRSHICK R,et al.You only look once: Unified, real-time object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Las Vegas:IEEE Press,2016:779-788.DOI:10.1109/CVPR.2016.91.
[7] REDMON J,FARHADI A.YOLO9000: Better, faster, stronger[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Honolulu:IEEE Press,2017:7263-7271.DOI:10.1109/CVPR.2017.690.
[8] REDMON J,FARHADI A.YOLOV3: An incremental improvement[EB/OL].(2018-04-08)[2024-01-03] .https://arxiv.org/abs/1804.02767.
[9] LIU Wei,ANGUELOV D,ERHAN D,et al.SSD: Single shot multibox detector[C]//Computer Vision-ECCV 2016: 14th European Conference.Cham:Springer International Publishing,2016:21-37.DOI:10.1007/978-3-319-46448-0_2.
[10] LIU Weipeng,SHAN Shengqi,CHEN Haiyong,et al.X-ray weld defect detection based on AF-RCNN[J].Welding in the World,2022,66(6):1165-1177.DOI:10.1007/s40194-022-01281-w.
[11] CHEN Yongbin,WANG Jingran,WANG Guitang.Intelligent welding defect detection model on improved R-CNN[J].IETE Journal of Research,2023,69(12):9235-9244.DOI:10.1080/03772063.2022.2040387.
[12] LIU Moyun,CHEN Youping,XIE Jingming,et al.LF-YOLO: A lighter and faster YOLO for weld defect detection of X-ray image[J].IEEE Sensors Journal,2023,23(7):7430-7439.DOI:10.1109/JSEN.2023.3247006.
[13] ZHANG Yi,NI Qingjian.A novel weld-seam defect detection algorithm based on the S-YOLO model[J].Axioms,2023,12(7):697.DOI:10.3390/axioms12070697.
[14] 程松,杨洪刚,徐学谦,等.基于YOLOv5的改进轻量型X射线铝合金焊缝缺陷检测算法[J].中国激光,2022,49(21):2104005.DOI:10.3788/CJL202249.2104005.
[15] YANG Jun,FU Bo,ZENG Jinquan,et al.YOLO-Xweld: Efficiently detecting pipeline welding defects in X-ray images for constrained environments[C]//International Joint Conference on Neural Networks.Padua:IEEE Press,2022:1-7.DOI:10.1109/IJCNN55064.2022.9892765.
[16] HU Jie,WANG Zhangbin,CHANG Minjie,et al.Psg-YOLOV5: A paradigm for traffic sign detection and recognition algorithm based on deep learning[J].Symmetry,2022,14(11):2262.DOI:10.3390/sym14112262.
[17] LI Hulin,LI Jun,WEI Hanbing,et al.Slim-neck by GSConv: A lightweight-design for real-time detector architectures [J].Journal of Real-Time Image Processing,2024,21(3):62.DOI:10.1007/s11554-024-01436-6.
[18] WANG Jiaqi,CHEN Kai,XU Rui,et al.Carafe: Content-aware reassembly of features[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision.Seoul:IEEE Press,2019:3007-3016.DOI:10.1109/ICCV.2019.00310.
[19] WAN Dahang,LU Rongsheng,SHEN Siyuan,et al.Mixed local channel attention for object detection[J].Engineering Applications of Artificial Intelligence,2023,123:106442.DOI:10.1016/j.engappai.2023.106442.
[20] 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.
[21] HUANG Hejun,CHEN Zuguo,ZOU Ying,et al.Channel prior convolutional attention for medical image segmentation[EB/OL].(2023-06-08)[2024-01-03] .https://arxiv.org/abs/2306.05196.
[22] LAU K W,PO L M,REHMAN Y A U.Large separable kernel attention: Rethinking the large kernel attention design in cnn[J].Expert Systems with Applications,2024,236:121352.DOI:10.1016/j.eswa.2023.121352.
[23] WOO S H,PARK J,LEE J Y,et al.Cbam: Convolutional block attention module[C]//Proceedings of the European Conference on Computer Vision.Cham:Springer International Publishing,2018:3-19.DOI:10.1007/978-3-030-01234-2_1.
[24] GUO Menghao,LU Chengze,HOU Qibin,et al.Segnext: Rethinking convolutional attention design for semantic segmentation[J].Advances in Neural Information Processing Systems,2022,35:1140-1156.DOI:10.48550/arXiv.2209.08575.

相似文献/References:

[1]郑晓玲,刘斌.采用机器视觉的铝压铸件表面缺陷检测[J].华侨大学学报(自然科学版),2016,37(2):139.[doi:10.11830/ISSN.1000-5013.2016.02.0139]
　ZHENG Xiaoling,LIU Bin.Surface Defect Detection of Aluminum Die Casting Using Machine Vision[J].Journal of Huaqiao University(Natural Science),2016,37(6):139.[doi:10.11830/ISSN.1000-5013.2016.02.0139]

备注/Memo

备注/Memo:
收稿日期: 2024-03-04
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)
更新日期/Last Update: 2024-11-20