Detection Method for Insulator Defect Based on Improved YOLOv5s

PENG Yanfei; YUAN Xiaolong; ZHAO Tao; CHEN Yankang

doi:10.13296/j.1001-1609.hva.2026.01.017

您当前的位置：

首页 >

文章列表页 >

Detection Method for Insulator Defect Based on Improved YOLOv5s

Research & Analysis | 更新时间：2026-04-13

- Detection Method for Insulator Defect Based on Improved YOLOv5s
- High Voltage Apparatus Vol. 62, Issue 1, Pages: 134-142(2026)
- 作者机构：
  
  辽宁工程技术大学电子与信息工程学院，辽宁葫芦岛　125105
- 作者简介：
- 基金信息：
- DOI：10.13296/j.1001-1609.hva.2026.01.017
  CLC：
- Received：21 June 2025，
  
  Revised：2025-09-18，
  
  Published：16 January 2026
- 稿件说明：
移动端阅览
PENG Yanfei, YUAN Xiaolong, ZHAO Tao, et al. Detection Method for Insulator Defect Based on Improved YOLOv5s[J]. High Voltage Apparatus, 2026, 62(1): 134-142.
DOI：

PENG Yanfei, YUAN Xiaolong, ZHAO Tao, et al. Detection Method for Insulator Defect Based on Improved YOLOv5s[J]. High Voltage Apparatus, 2026, 62(1): 134-142. DOI： 10.13296/j.1001-1609.hva.2026.01.017.

摘要

基于无人机航拍的电力巡检成为目前绝缘子缺陷检测方法的主流，但当遇到图像特征不够明显或干扰特征较多等问题时，绝缘子缺陷识别困难，检测精度不高。由此，提出了一种基于改进YOLOv5s的绝缘子缺陷检测方法。首先，重新设计卷积模块，然后将CA注意力机制与其相融合，并且在主干网络加入注意力机制与颈部网络的特征图进行多尺度特征融合，抑制复杂环境下的干扰特征，专注缺陷特征提取；其次，对空间金字塔池化结构（SPPF）进行改进，扩大感受野，减少被模型过滤掉的有用信息；接着，将Transformer与C3模块中的残差结构（Bottleneck）相结合，增强模型对绝缘子缺陷特征的识别能力；最后，使用K-means算法对数据集进行聚类分析，重新计算最适合的锚框尺寸。在数据集上进行验证，改进后的方法平均精度达到97.4%，召回率达到94.8%，均值平均精度为97.6%，该方法有效提升了复杂环境下的绝缘子缺陷检测能力，进一步满足了对绝缘子缺陷检测精度的需求。

Abstract

Electric power inspection based on UAV aerial photography has become presently the mainstream method for insulator defect detection. However

in case of such issues as insufficiently distinct image features or excessive interfering features

the defect identification of the insulator is difficult and the detection accuracy is not high. Therefore

a method for insulator defect detection based on improved YOLOv5s is proposed. First

the convolutional module is redesigned

and then integrated with the CA attention mechanism. Furthermore

the attention mechanism is incorporated into the backbone network to perform multi-scale feature fusion with feature maps from the neck network. This helps to suppress interference features in complex environments and enhance focus on defect feature extraction. Then

the spatial pyramid pool structure (SPPF) is improved to enlarge the receptive field and reduce the useful information filtered out by the model. After that

a Bottleneck in the C3 module is combined with Transformer to enhance the model's ability to recognize insulator defect characteristics. Finally

the K-means algorithm is used to perform cluster analysis on the data set and recalculate the most suitable anchor frame size. Verification is performed on the data set

and the average accuracy of the improved method reaches 97.4%

the recall rate reaches 94.8%

and the average accuracy is 97.6%. This method effectively improves the ability of insulator defect detection in complex environments

and further meets the demand for defect detection accuracy of insulator .

关键词

Keywords

references

刘悦,黄新波,刘天娇.基于深度学习的架空输电线路绝缘子缺陷检测方法研究综述[J].电力电容器与无功补偿,2024,45 (3):167-173.

LIU Yue,HUANG Xinbo,LIU Tanjao.Overview on defect detection methods of insulator for overhead transmission lines based on deep learning[J]. Power Capacitor & Reactive Power Compensation,2024,45(3):167-173.

刘志颖,缪希仁,陈静,等.电力架空线路巡检可见光图像智能处理研究综述[J].电网技术,2020,44(3):1057-1069.

LIU Zhiying,MIAO Xiren,CHEN Jing,et al. Review of visible image intelligent processing for transmission line inspection[J]. Power System Technology,2020,44(3):1057-1069.

张东东,常泽中,万武艺,等.基于局部电场的零值绝缘子检测技术研究[J].电力工程技术,2024,43(4):193-201.

ZHANG Dongdong,CHANG Zezhong,WAN Wuyi,et al. Zero-value insulator detection technology based on local electric field[J]. Electric Power Engineering Technology,2024,43(4):193-201.

唐捷,黄志都,朱登杰,等.交叉跨越段输电线绝缘子发热检测方法研究[J]. 电网与清洁能源,2024,40(6):115-120.

TANG Jie,HUANG Zhidu,ZHU Dengjie,et al. A study on the heat detection method of insulators in the cross section of transmission lines[J]. Power System and Clean Energy,2024,40(6):115-120.

尹向雷,解永芳,屈少鹏,等.基于动态蛇形卷积和非跨步卷积的绝缘子缺陷检测[J]. 电力系统保护与控制,2024,52(20):177-187.

YIN Xianglei,XIE Yongfang,QU Shaopeng,et al.Insulator defect detection based on dynamic snake convolution and space-to-depth convolution[J]. Power System Protection and Control,2024,52(20):177-187.

何昱燊,张志劲,傅海涛,等.自然环境下覆冰绝缘子长串交流闪络电压及电弧发展现象[J]. 电网技术,2021,45(7):2904-2912.

HE Yushen,ZHANG Zhijin,FU Haitao,et al. AC flashover voltage and arc development of long insulators strings in natural icing environment[J]. Power System Technology,2021,45(7):2904-2912.

李敬如,李红军,马良,等.基于人工智能的配电网规划技术研究现状与展望[J].电力建设,2025,46(4):1-15.

LI Jingru,LI Hongjun,MA Liang,et al. Research status and prospects of distribution network planning technology based on artificial intelligence[J]. Electric Power Construction,2025,46(4):1-15.

王道累,张正刚,张世恒,等.基于密集连接网络的航拍绝缘子旋转目标精准定位方法[J]. 电力系统保护与控制,2024,52(1):35-43.

WANG Daolei,ZHANG Zhenggang,ZHANG Shiheng,et al. Accurate positioning method of insulator rotating target in aerial photography based on dense connection network[J]. Power System Protection and Control,2024,52(1):35-43.

李斌,曾筠婷,朱新山,等.基于多尺度上下文感知的绝缘子缺陷检测网络[J].高电压技术,2022,48(8):2905-2914.

LI Bin,ZENG Junting,ZHU Xinshan,et al. Detection network for insulator defects based on multi-scale context awareness[J]. High Voltage Engineering,2022,48(8):2905-2914.

REDMON J,DIVVALA S,GIRSHICK R,et al. You only look once:Unified,real-time object detection[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition(CVPR). USA:IEEE,2016:779-788.

REDMON J,FARHADI A. YOLO9000:Better,faster,stronger[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition(CVPR). USA:IEEE,2017:6517-6525.

LIU Wei,ANGUELOV D,ERHAN D,et al. SSD:Single shot MultiBox detector[C]//Computer Vision-ECCV 2016. Cham:Springer International Publishing,2016:21-37.

GIRSHICK R,DONAHUE J,DARRELL T,et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]//2014 IEEE Conference on Computer Vision and Pattern Recognition. USA:IEEE,2014:580-587.

REN Shaoqing,HE Kaiming,GIRSHICK R,et al. Faster R-CNN:Towards real-time object detection with region proposal networks[C]//Proceedings of the 28th International Conference on Neural Information Processing Systems-Volume 1. Cambridge,MA,USA:MIT Press,2015:91-99.

HE Kaiming,GKIOXARI G,DOLLÁR P,et al. Mask R-CNN[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020,42(2):386-397.

丘灵华,朱铮涛.基于深度学习的输电线路绝缘子缺陷检测研究[J].计算机应用研究,2020,37(s1):358-360.

QIU Linghua,ZHU Zhengtao. Research on insulator defect detection of transmission line based on deep learning[J]. Application Research of Computers,2020,37(s1):358-360.

张烨,高玉菡,黄新波,等.基于Fine Mask RCNN的110~220 kV输电铁塔涉鸟故障识别与评估[J].电网技术,2022,46(6):2132-2140.

ZHANG Ye,GAO Yuhan,HUANG Xinbo,et al. Fault identification and evaluation of tower wading in 110~220 kV transmission line with Fine Mask RCNN[J]. Power System Technology,2022,46(6):2132-2140.

张青华,李彩林,孙延坤,等.基于深度学习的绝缘子精确识别[J].电瓷避雷器,2022(1):143-150.

ZHANG Qinghua,LI Cailin,SUN Yankun,et al. Accurate identificationof insulator based on deep learning[J]. Insulators and Surge Arresters,2022(1):143-150.

郝帅,马瑞泽,赵新生,等.基于卷积块注意模型的YOLOv3输电线路故障检测方法[J].电网技术,2021,45(8):2979-2987.

HAO Shuai,MA Ruize,ZHAO Xinsheng,et al. Fault detection of YOLOv3 transmission line based on convolutional block attention model[J]. Power System Technology,2021,45(8):2979-2987.

李利荣,张云良,陈鹏,等.基于轻量化YOLOv4的复杂场景绝缘子缺陷检测算法[J].光电子·激光,2022,33(6):598-606.

LI Lirong,ZHANG Yunliang,CHEN Peng,et al. Insulator defect detection algorithm for complex scenes based on lightweight YO-LOv4[J]. Journal of Optoelectronics Laser,2022,33(6):598-606.

丁丽丹,胡君红,胡聪.基于绝缘子缺陷检测的模糊图像增强算法研究[J].激光杂志,2022,43(10):56-61.

DING Lidan,HU Junhong,HU Cong. Research on fuzzy image enhancement algorithm based on insulator defect detection[J]. Laser Journal,2022,43(10):56-61.

HOU Qibin,ZHOU Daquan,FENG Jiashi. Coordinate attention for efficient mobile network design[C]//2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR). USA:IEEE, 2021:13708-13717.

HU L L,GENG X,LI Z,et al. Improving YOLOv5 with attention mechanism for detecting boulders from planetary images[J]. Remote Sensing,2021,13(18):3776-3776.

LIU Bin,ZHANG Xiaoyun,GAO Zhiyong,et al. Weld defect images classification with VGG16-based neural network[C]//Digital TV and Wireless Multimedia Communication. Singapore:Springer Singapore,2018:215-223.

LIU Ze,LIN Yutong,CAO Yue,et al. Swin transformer:Hierarchical vision transformer using shifted windows[C]//2021 IEEE/CVF International Conference on Computer Vision(ICCV). Canada:IEEE,2021:9992-10002.

JIANG Shan,FERREIRA J,GONZALEZ M C. Activity-based human mobility patterns inferred from mobile phone data:A case study of Singapore[J]. IEEE Transactions on Big Data,2017,3(2):208-219.

LIN T Y,DOLLÁR P,GIRSHICK R,et al. Feature pyramid networks for object detection[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition(CVPR). USA:IEEE,2017:936-944.

LIU Shu,QI Lu,QIN Haifang,et al. Path aggregation network for instance segmentation[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. USA:IEEE,2018:8759-8768.

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Research on Multi-type Defect Detection of Insulator for Transmission Line Based on GE-YOLO Algorithm

Research on Multi Type Defect Detection of Transmission Line Insulators Based on GE-YOLO Algorithm

Research on Optimal Setting of Excitation Points for Defect Detection Method of Insulating Spacer Based on Ultrasonic Guided Wave

Research on Application of Improved YOLOv11n in Detection of Main Defects of Insulators on Transmission Lines

Insulator Defect Detection Algorithm for Multi-scale Feature Extraction Based on Improved YOLOv7

Related Author

WANG Qiang

FENG Yuntao

ZHOU Chao

WANG Qiang

FENG Yuntao

ZHOU Chao

LI Yang

TANG Yan

Related Institution

Hubei Provincial Engineering Research Center of Intelligent Energy Technology

College of Electrical and New Energy, China Three Gorges University

School of Electrical and Electronic Engineering, North China Electric Power University

Lijiang Power Supply Bureau of Yunnan Power Grid Co., Ltd.

国家电网葫芦岛供电公司，辽宁葫芦岛 125000State Grid Huludao Power Supply Company, Liaoning Huludao 125000, China

⁰