计算机科学

计算机科学 ›› 2025, Vol. 52 ›› Issue (6A): 240800044-8.doi: 10.11896/jsjkx.240800044

• 图像处理&多媒体技术 • 上一篇    下一篇

基于迁移学习与改进YOLOv8s的输电线路故障识别方法

黄柏澄1, 王晓龙2, 安国成2, 张涛1   

  1. 1 上海交通大学电子信息与电气工程学院北斗导航与位置服务上海市重点实验室 上海 200240
    2 上海华讯网络系统有限公司行业数智事业部 成都 610074
  • 出版日期:2025-06-16 发布日期:2025-06-12
  • 通讯作者: 王晓龙(wxlong@eccom.com.cn)
  • 作者简介:(122035910078@sjtu.edu.cn)
  • 基金资助:
    国家重点研发计划(2023YFC3006700)

Transmission Line Fault Identification Method Based on Transfer Learning and Improved YOLOv8s

HUANG Bocheng1, WANG Xiaolong2, AN Guocheng2, ZHANG Tao1   

  1. 1 Shanghai Key Laboratory of Beidou Navigation and Location Services,College of Electronic Information and Electrical Engineering,Shanghai Jiao Tong University,Shanghai 200240,China
    2 Artificial Intelligence Research Institute of Shanghai Huaxun Network System Co.,Ltd.,Chengdu 610074,China
  • Online:2025-06-16 Published:2025-06-12
  • About author:HUANG Bocheng,born in 2000,postgraduate. His main research interests include target detection and deep lear-ning.
    WANG Xiaolong,born in 1980,master,senior engineer. His main research interests include intelligent information processing and data analysis.
  • Supported by:
    National Key R&D Program of China(2023YFC3006700).

PDF (PC)

39

摘要： 目前,输电线路部分故障类别识别存在样本严重不足、无人机拍摄远距离小目标定位困难等问题,导致输电线路故障识别精度较低。为此,提出一种基于迁移学习与改进YOLOv8s的输电线路故障识别方法。首先,为改善小样本情况下的故障识别效果,该算法以YOLOv8s作为基线模型,使用迁移学习方法对模型进行预训练,并提出一种基于双向相关性的迁移学习样本选择模块,筛选出与目标域具有强相关性的样本类别,避免使用迁移学习时可能产生的负迁移问题,更好地辅助故障识别任务。其次,针对小目标定位困难问题,通过设计小目标注意检测层,将80*80输出特征图与浅层特征图进行特征融合后,引入EMA多尺度注意力机制,增强小目标特征信息;在预测框回归损失中使用NWD损失替换CIoU损失,采取Wasserstein距离度量小目标预测框与真值框的相似性,解决了IoU对小目标位置偏差敏感的问题,有效提升了小目标检测精度。实验结果表明:在小样本与小目标情况下,所提方法在输电线路故障数据集中mAP为51.1%,相较于YOLOv8s基线模型提升了8.2%,有效提升了故障识别精度,为小样本与小目标输电线路故障识别提供了新的解决思路与办法。

关键词: 故障识别, YOLOv8s, 迁移学习, 样本选择, 注意力机制, 损失函数

Abstract: At present,there are serious problems when identifying some fault categories in transmission lines,such as insufficient samples and difficulty in locating small targets at long distances captured by drones,resulting in low accuracy in fault identification of transmission lines.To address the above issues,a novel transmission line fault identification method based on transfer learning and improved YOLOv8s is proposed.Firstly,taking YOLOv8s as the baseline,the transfer learning method is used to improve the recognition performance in few-shot,and a bidirectional correlation sample selection module is proposed to obtain sample categories with strong correlation with the target domain,avoiding the problem of negative transfer that may occur when using transfer learning,effectively improving the fault recognition performance of the model.Secondly,aiming at the difficult problem of small target localization,after fusing the 80*80 output feature map and the shallow feature map,EMA is introduced into multi-scale attention to enhance the feature information of the small target for designing small target attention detection layer.To improve the loss function,the CIoU loss is replaced by NWD loss in the prediction frame regression loss,which solves the problem that IoU is sensitive to small target position deviation.Specifically,the Wasserstein distance is used to measure the similarity between prediction box and the truth box.Experimental results show that in the case of few shot and small targets,the proposed method has a mAP of 51.1% in the transmission line fault dataset,which is 8.2% higher than that of the YOLOv8s baseline model,effectively improving the accuracy of fault recognition and providing new solution and method for few-shot and small target transmission line fault identification.

Key words: Fault identification, YOLOv8s, Transfer learning, Sample selection, Attention mechanism, Loss function

中图分类号: 

  • TP391
[1]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,2023,13:5896.
[2]WANG Z Y,GAO Q,LI D,et al.Insulator Anomaly DetectionMethod Based on Few-Shot Learning[J].IEEE Access,2021,9:94970-94980.
[3]LIU C Y,WU Y Q.Research Progress of Vision DetectionMethods Based on Deep Learning for Transmission Lines[C]//Proceedings of the CSEE.2023:7423-7445.
[4]TAO X,ZHANG D P,WANG Z H,et al.Detection of Power Line Insulator Defects Using Aerial Images Analyzed With Convolutional Neural Networks[J].IEEE Transactions on Systems,Man,and Cybernetics:Systems,2020,50(4):1486-1498.
[5]ZHOU M,LI B,WANG J,et al.Fault Detection Method ofGlass Insulator Aerial Image Based on the Improved YOLOv5[J].IEEE Transactions on Instrumentation and Measurement,2023,72:1-10.
[6]HAO S,AN B Y,MA X,et al.PKAMNet:A Transmission Line Insulator Parallel- Gap Fault Detection Network Based on Prior Knowledge Transfer and Attention Mechanism[J].IEEE Transactions on Power Delivery,2023,38(5):3387-3397.
[7]ZHAI Y J,YANG K,et al.Disc Insulator Defect DetectionBased on Mixed Sample Transfer Learning[C]//Proceedings of the CSEE.2023:2867-2876.
[8]LIU C Y,WU Y Q,LIU J J,et al.Insulator Faults Detection in Aerial Images from High-Voltage Transmission Lines Based on Deep Learning Model[J].Applied Sciences,2021,11(10):4647.
[9]YI W,MA S,LI R.Insulator and Defect Detection Model Based on Improved Yolo-S[J].IEEE Access,2023,11:93215-93226.
[10]WANG T,ZHAI Y D,LI Y H,et al.Insulator Defect Detection Based on ML-YOLOv5 Algorithm[J].Sensors,2024,24(1):204.
[11]陈华杰,吕丹妮.一种基于逆迁移学习的小样本目标识别辅助数据挑选方法[P].CN112541539A,2021-03-23.
[12]SAMET N,HICSONMEZ S,AKBAS E.HoughNet:Integrating Near and Long-Range Evidence for Visual Detection[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2023,45(4):4667-4681.
[13]JOCHER G.Ultralytics YOLOv8[EB/OL].[2023-07-05].https://github.com/ultralytics/ultralytics.
[14]WANG C Y,BOCHKOVSKIY A,LIAO H M.YOLOv7:Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[J].arXiv:2207.02696,2023.
[15]LIU S,QI L,QIN H F,et al.Path aggregation network for instance segmentation[C]//2018 IEEE Conference on Computer Vision and Pattern Recognition.2018:8759-8768.
[16]OUYANG D L,HE S,ZHANG G Z,et al.Efficient Multi-Scale Attention Module with Cross-Spatial Learning[J].arXiv:2305.13563,2023.
[17]HOU Q B,ZHOU D Q,FENG J S.Coordinate attention for efficient mobile network design[C]//IEEE Conference on Computer Vision and Pattern Recognition.2021:13708-13717.
[18]ZHENG Z H,WANG P,LIU W,et al.Distance-IoU Loss:Faster and Better Learning for Bounding Box Regression[J].AAAI Conference on Artificial Intelligence,2020,34:12993-13000.
[19]WANG J W,XU C,YANG W,et al.A Normalized GaussianWasserstein Distance for Tiny Object Detection[J].arXiv:2110.13389,2021.
[20]CHENG G,YUAN X,YAO X W,et al.Towards Large-Scale Small Object Detection:Survey and Benchmarks[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2023,45(11):13467-13488.
[21]ZHANG Y F,REN W Q,ZHANG Z,et al.Focal and efficient IoU loss for accurate bounding box regression[J].Neurocomputing,2022,506:146-157.
[22]GEVORGYAN Z.SIoU Loss:More Powerful Learning forBounding Box Regression[J].arXiv:2205.12740,2022.
[23]YANG L X,ZHANG R Y,XIE X H.SimAM:A Simple,Parameter-Free Attention Module for Convolutional Neural Networks[C]//International Conference on Machine Learning.2021.
[24]MISRA D,NALAMADA T,ARASANIPALAI A U,et al.Rotate to Attend:Convolutional Triplet Attention Module[C]//2021 IEEE Winter Conference on Applications of Computer Vision.2021:3138-3147.
[25]REN S Q,HE K M,et al.Faster R-CNN:Towards real-time object detection with region proposal networks.IEEE Transactions on Pattern Analysis and Machine Intelligence,2017,39(6):1137-1149.
[26]TERVEN J AND CORDOVA-ESPARZA D.A comprehensive review of YOLO:From YOLOv1 to YOLOv8 and beyond[J].arXiv:2304.00501,2023.
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