计算机科学

计算机科学 ›› 2025, Vol. 52 ›› Issue (6): 286-296.doi: 10.11896/jsjkx.240300146

• 计算机图形学&多媒体 • 上一篇    下一篇

基于局部和全局特征表示的小样本绝缘子缺陷检测

崔克彬1,2, 胡真真1   

  1. 1 华北电力大学计算机系 河北 保定 071003
    2 复杂能源系统与智能计算教育部工程研究中心 河北 保定 071003
  • 收稿日期:2024-03-20 修回日期:2024-07-05 出版日期:2025-06-15 发布日期:2025-06-11
  • 通讯作者: 胡真真(zzjtcblm@163.com)
  • 作者简介:(ncepuckb@163.com)
  • 基金资助:
    国家电网有限公司总部管理科技项目(5700-202340289A-1-1-ZN)

Few-shot Insulator Defect Detection Based on Local and Global Feature Representation

CUI Kebin1,2, HU Zhenzhen1   

  1. 1 Department of Computer,North China Electric Power University,Baoding,Hebei 071003,China
    2 Engineering Research Center of Intelligent Computing for Complex Energy Systems,Ministry of Education,Baoding,Hebei 071003,China
  • Received:2024-03-20 Revised:2024-07-05 Online:2025-06-15 Published:2025-06-11
  • About author:CUI Kebin,born in 1979,Ph.D,professor.His main research interests include image processing,pattern recognition and machine learning.
    HU Zhenzhen,born in 2000,postgra-duate.Her main research interests include object detection and few-shot learning.
  • Supported by:
    State Grid Corporation of China Headquarters Management Science and Technology Project(5700-202340289A-1-1-ZN).

PDF (PC)

58

摘要： 为解决绝缘子缺陷样本数量少且缺陷目标小导致目前绝缘子缺陷检测精度偏低这一问题,提出一种结合CNN与Transformer的小样本目标检测模型(C-TFSIDD),通过融合图像局部和全局特征来更有效地实现绝缘子缺陷检测。首先,采用融合CNN局部细节捕捉能力与Transformer全局信息整合能力的Next-ViT作为特征提取模块,精准捕获绝缘子图像局部和全局特征信息;其次,采用改进路径聚合特征金字塔网络(Path Aggregation Feature Pyramid Network,PAFPN)进行双向多尺度特征融合,增强底层特征表示,以改善小目标的检测效果;最后,提出一个基于度量的判别性损失函数,在微调阶段优化分类器学习更具判别性的特征表示,以增加类别之间的可分性,减少类内变化的影响。在两个公开的绝缘子缺陷数据集上进行训练和评估,实验结果表明,与基线模型TFA相比,C-TFSIDD在样本为5shot,10shot,20shot的检测结果分别提升28.7%,35.5%,47.7%;与小样本目标检测模型FSCE相比,C-TFSIDD分别提升21.8%,26.7%,21.1%。结果表明,C-TFSIDD能有效提升小样本条件下的绝缘子缺陷检测精度。

关键词: 缺陷检测, 绝缘子, 小样本, CNN-Transformer, 度量学习

Abstract: In order to solve the problem that the small number of insulator defect samples and small defect targets lead to the current low accuracy of insulator defect detection,this paper proposes a few-shot object detection model(C-TFSIDD) combining CNN and Transformer,which fuses local and global features of images to realize insulator defect detection more effectively.Firstly,Next-ViT,which integratesthe local detail capture capability of CNN and the global information integration capability of Transformer,is used as the feature extraction module to accurately capture local and global feature information of insulator images.Secondly,the improved path aggregation feature pyramid network(PAFPN) is used for bidirectional multi-scale feature fusion to enhance the underlying feature representation and improve the detection effect of small targets.Finally,a metric-based discriminative loss is proposed to optimize the classifier in the fine-tuning stage to learn more discriminative feature representations to increase the separability between classes and reduce the effect of intra-class variations.Trained and evaluated on two public insulator defect datasets,the experimental results show that C-TFSIDD improves the detection results with samples of 5shot,10shot,and 20shot by 28.7%,35.5%,and 47.7%,respectively,compared to the baseline model TFA,and compared with the few-shot object detection model FSCE,the proposed method improved by 21.8%,26.7%,and 21.1%,respectively.The results show that C-TFSIDD can effectively improve the defect detection accuracy of few-shot insulator samples.

Key words: Defect detection, Insulator, Few-shot, CNN-Transformer, Metric learning

中图分类号: 

  • TP391
[1]LIU C Y,WU Y Q.Research Progress of Vision DetectionMethods Based on Deep Learning for Transmission Lines[J].Proceedings of the CSEE,2023,43(19):7423-7446.
[2]ZHAO Z B,JIANG Z G,LI Y X,et al.Overview of visual defect detection of transmission line components[J].Journal of Image and Graphics,2021,26(11):2545-2560.
[3]LIU Y,HUANG X.Efficient Cross-Modality Insulator Aug-mentation for Multi-Domain Insulator Defect Detection in UAV Images[J].Sensors,2024,24(2):428.
[4]MA B,FU Y K,WANG C P,et al.High Performance lnsulators Location Scheme Based on YOLOv4 with GDloU Loss Function[J].Computer Science,2022,49(S1):412-417.
[5]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.2016:779-788.
[6]ZHAO W Q,CHENG X F,ZHAO Z B,et al.Insulator recognition based on attention mechanism and Faster RCNN[J].CAAI Transactions on Intelligent Systems,2020,15(1):92-98.
[7]REN S,HE K,GIRSHICK R,et al.Faster R-CNN:Towards real-time object detection with region proposal networks[C]//Proceedings of the 29th International Conference on Neural Information Processing Systems.Cambridge.MA:MIT,2015:91-99.
[8]LIU J H,ZHAO Z,FU J R,et al.Active small sample learning based the pipe weld defect detection method[J].Chinese Journal of Scientific Instrument,2022,43(11):252-261.
[9]ZHAO Z F,HUANG J H,LUO H J,et al.Simulation Research on Ultrasonic Total Focus Method Detection of Internal Defects of Composite Insulators[J].Piezoelectrics and Acoustooptics,2024,46(1):136-142.
[10]WANG X,HUANG T E,DARRELL T,et al.Frustratingly simplefew-shot object detection[J].arXiv:2003.06957,2020.
[11]LI J,XIA X,LI W,et al.Next-vit:Next generation vision transformer for efficient deployment in realistic industrial scenarios[J].arXiv:2207.05501,2022.
[12]LIU S,QI L,QIN H,et al.Path aggregation network for instance segmentation[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2018:8759-8768.
[13]ZHANG T,ZHANG X,ZHU P,et al.Generalized few-shot object detection in remote sensing images[J].ISPRS Journal of Photogrammetry and Remote Sensing,2023,195:353-364.
[14]KÖHLER M,EISENBACH M,GROSS H M.Few-shot object detection:A comprehensive survey[J].arXiv:2112.11699,2021.
[15]SHI Y D,WANG H M,JING C,et al.A Few-Shot Defect Detection Method for Transmission Lines Based on Meta-Attention and Feature Reconstruction[J].Applied Sciences,2023,13(10):5896.
[16]ZHAI Y J,YANG K,WANG Q M,et al.Disc Insulator Defect Detection Based on Mixed Sample Transfer Learning[J].Proceedings of the CSEE,2023,43(7):2867-2877.
[17]CUI K B,PAN F.A CycleGAN small sample library amplification method for faulty insulator detection[J].Computer Engineering & Science,2022,44(3):509-515.
[18]KARLINSKY L,SHTOK J,HARARY S,et al.Repmet:Representative-based metric learning for classification and few-shot object detection[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2019:5197-5206.
[19]EVERINGHAM M,VAN GOOL L,WILLIAMS C K I,et al.The pascal visual object classes(VOC) challenge[J].International Journal of Computer Vision,2010,88:303-338.
[20]LIU K P,LI B Q,QIN L,et al.Review of Application Research of Deep Learning Object Detection Algorithms in Insulator Defect Detection of Overhead Transmission Lines[J].High Vol-tage Engineering,2023,49(9):3584-3595.
[21]ZHOU L J,MAO J N.Vision Transformer-based recognitiontasks:a critical review[J].Journal of Image and Graphics,2023,28(10):2969-3003.
[22]VASWANI A,SHAZEER N,PARMAR N,et al.Attention isall you need[C]//Proceedings of the 31st International Confe-rence on Neural Information Processing Systems.Red Hook,NY:Curran Associates Inc.,2017:6000-6010.
[23]XI Y,ZHOU K,MENG L W,et al.Transmission Line Insulator Defect Detection Based on Swin Transformer and Context[J].Machine Intelligence Research,2023,20:729-740.
[24]GUO J,LI T,DU B.Segmentation Head Networks with Har-nessing Self-Attention and Transformer for Insulator Surface Defect Detection[J].Applied Sciences,2023,13(16):9109.
[25]DU Z W,ZHOU H,LI C Y,et al.Small Object Detection Based on Deep Convolutional Neural Networks:A Review[J].Computer Science,2022,49(12):205-218.
[26]ZHAO Y,YANG L.Distance metric learning based on the class center and nearest neighbor relationship[J].Neural Networks,2023,164:631-644.
[27]LIN T Y,GOYAL P,GIRSHICK R,et al.Focal loss for dense object detection[C]//Proceedings of the IEEE International Conference on Computer Vision.2017:2980-2988.
[28]RAIMUNDO A.Insulator data set-Chinese power line insulator dataset(CPLID)[DB/OL].https://github.com/InsulatorData/InsulatorDataSet.
[29]DENG J H,GUO W Q,CHEN H J,et al.Few-shot diatom detection combining multi-scale multi-head self-attention and online hard example mining[J].Journal of Computer Applications,2022,42(8):2593-2600.
[30]SUN B,LI B,CAI S,et al.FSCE:Few-shot object detection via contrastive proposal encoding[C]//Proceedings of the 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Piscataway:IEEE,2021:7352-7362.
[31]XIAO Y,LEPETIT V,MARLET R.Few-shot object detection and viewpoint estimation for objects in the wild[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2020,45(3):3090-3106.
[32]YAN X,CHEN Z,XU A,et al.Meta R-CNN:Towards general solver for instance-level low-shot learning[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision.2019:9577-9586.
[33]WU J,LIU S,HUANG D,et al.Multi-scale positive sample refinement for few-shot object detection[M]//Lecture Notes in Computer Science.Cham:Springer,2020:456-472.
[34]PANG J,CHEN K,SHI J,et al.Libra R-CNN:Towards ba-lanced learning for object detection[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2019:821-830.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发