A Model Coupled Shuffle Attention and Deformable Convolution for Overhead Transmission Line Hardware Object Detection

XU Guowei; WU Liang; ZHAI Han; LI Xin

doi:10.13203/j.whugis20240409

您当前的位置：

首页 >

文章列表页 >

A Model Coupled Shuffle Attention and Deformable Convolution for Overhead Transmission Line Hardware Object Detection

Photogrammetry and Remote Sensing | 更新时间：2026-11-05

- A Model Coupled Shuffle Attention and Deformable Convolution for Overhead Transmission Line Hardware Object Detection
- Geomatics and Information Science of Wuhan University Vol. 51, Issue 3, Pages: 544-556(2026)
- 作者机构：
  
  1.广东电网有限责任公司汕头供电局，广东汕头，515041
  2.中国地质大学(武汉)地理与信息工程学院，湖北武汉，430074
  3.武汉大学遥感信息工程学院，湖北武汉，430079
- 作者简介：
- 基金信息：
- DOI：10.13203/j.whugis20240409
  CLC： P237
- Received：26 January 2025，
  
  Published：05 March 2026
- 稿件说明：
移动端阅览
许国伟,吴亮,翟晗，等.耦合置换注意力和可形变卷积的输电线金具目标检测[J].武汉大学学报（信息科学版）,2026,51(3):544-556.

XU Guowei,WU Liang,ZHAI Han,et al.A Model Coupled Shuffle Attention and Deformable Convolution for Overhead Transmission Line Hardware Object Detection[J].Geomatics and Information Science of Wuhan University,2026,51(3):544-556.
许国伟,吴亮,翟晗，等.耦合置换注意力和可形变卷积的输电线金具目标检测[J].武汉大学学报（信息科学版）,2026,51(3):544-556. DOI：10.13203/j.whugis20240409

XU Guowei,WU Liang,ZHAI Han,et al.A Model Coupled Shuffle Attention and Deformable Convolution for Overhead Transmission Line Hardware Object Detection[J].Geomatics and Information Science of Wuhan University,2026,51(3):544-556. DOI：10.13203/j.whugis20240409

摘要

金具部件的准确识别是架空输电线路安全巡检的一项关键性基础工作。针对金具目标纹理特征多变、形态复杂等问题，提出一种耦合置换注意力和可形变卷积的YOLOv8模型，用于架空输电线路无人机高分辨率可见光影像金具目标识别。通过置换注意力捕获通道和像素间的依赖关系，增强模型空间通道特征挖掘能力；耦合可形变卷积网络，精细挖掘各类金具目标的复杂空间形态特征，实现金具目标的高精度定位；采用基于动态非单调聚焦机制的边界框损失进行网络模型训练，提高模型的样本学习能力；采用无人机实地采集汕头架空输电线路高分辨率可见光影像作为数据源，构建金具目标数据集进行实验，并与多种先进的金具目标检测模型进行对比。实验结果表明，所提模型金具目标检测性能明显优于现有主流方法，在交并比阈值为0.5时，平均精度达到84.4%，所提模型能够快速、高精度识别各类金具目标，保障输电基础设施的安全维护。

Abstract

Objectives

The accurate identification of hardware components is a crucial foundational task for the safety inspection of overhead transmission lines. This paper proposes a novel YOLOv8 model coupling shuffle attention and deformable convolution (YOLOv8-SADC) to address the issues of variable texture features and complex shapes of hardware targets

which is used for overhead transmission line hardware object detection based on high-resolution visible images by unmanned aerial vehicles.

Methods

The shuffle attention is constructed to capture the dependencies between channels and pixels to enhance spatial channels feature mining ability of the model. The deformable convolutional network is used to finely mine the complex spatial morphological features of various hardware targets to achieve high-precision positioning of hardware targets. An intersection over union (IoU)-based loss with a dynamic non-monotonic focusing mechanism based bounding box loss (named Wies-IoU) is adopted for network training to improve sample learning ability of the model.

Results

The high-resolution visible light images of Shantou overhead transmission lines were collected by unmanned aerial vehicles (UAVs) as data sources

and a hardware target dataset was constructed for experiments

and compared with several advanced hardware target detection models. (1) Comparative analysis was carried out on CPLID

2511 and self-built datasets

and the CPLID dataset included 600 images of normal insulators taken by UAVs; the 2511 dataset uses a blue mat as the background

and shoots four types of fittings from multiple angles: insulators

bolt pins

anti-vibration hammers

and tension clamps

with a total of 2 511 images. The self-built dataset includes 6 types of fittings including connecting fittings

plate fittings

fan-shaped fittings

triangular fittings

short connecting fittings and long fittings

with a total of 516 images and 1 745 labels. Experimental results show that YOLOv8-SADC has excellent performance on CPLID and 2511 datasets

and can fully adapt to scenarios with simple backgrounds and single object categories. (2) The comparative experimental results show that the target detection performance of the proposed model is significantly better than that of the existing mainstream algorithms

and the mean average precision (mAP) reached 84.4% when the IoU threshold is 0.5

indicating that the method demonstrates good robustness to various types of fittings and can effectively detect all kinds of fittings. Among them

the mAP50 of the connecting category reached 90.52%

and the mAP50 of the long category reached 93.52%

indicating that the method has good detection ability for all kinds of fittings

especially the rectangular shape of the target fittings. For the sector fittings with small targets and a large number of occlusions

the detection accuracy is also improved. (3) There were 116 samples with blue sky background

646 samples with cloud background

495 samples with buildings as background

and 488 samples with natural environment as background. There is less interference in the blue sky background

and the target characteristics of the fittings are prominent

and it is easy to extract key information. The natural environment background is mostly trees and grasslands

and the detection accuracy decreases slightly. The cloud texture in the cloudy background may introduce some background interference

resulting in the reduced feature extraction efficiency of the model. There is a certain similarity between the structural characteristics of the building (such as straight lines

edges

etc.) and the shape of the fittings

which can easily lead to false detection. The performance improvement of YOLOv8-SADC varies in different backgrounds

with the natural environment and building background being relatively complex

and the detection accuracy is significantly higher in simple backgrounds (such as blue sky and cloudy background).

Conclusions

YOLOv8-SADC

based on high-resolution visible light images captured by drones

enhances the extraction capability of spatial channels features through coupled shuffle attention

and introduces deformable convolution to delve deeper into the complex morphological features of metal hardware. Additionally

a Wise-IoU loss function based on a dynamic non-monotonic focusing mechanism is employed to optimize localization accuracy. Experiments conducted on a drone-acquired dataset along the coastline of Shantou

Guangdong demonstrate that this method significantly improves the detection performance of various metal fastener targets in complex scenes

effectively ensuring the integrity and detail of recognition results. This represents an efficient solution for metal fastener identification.

关键词

Keywords

references

方苏，李立学，郑益慧，等 . 基于激光测距成像和图像处理的输电线路防护技术［J］. 电气自动化， 2017 ， 39 （ 3 ）： 6 - 8 .

FANG Su ， LI Lixue ， ZHENG Yihui ， et al . Protection Technology for Transmission Lines Based on Laser Range Imaging and Image Processing ［J］. Electrical Automation ， 2017 ， 39 （ 3 ）： 6 - 8 .

王炜，袁奇，顾俊杰，等 . X射线无损探伤技术在检测输电线路压接金具中的应用［J］. 上海交通大学学报， 2018 ， 52 （ 10 ）： 1189 - 1194 .

WANG Wei ， YUAN Qi ， GU Junjie ， et al . Application of X-Ray Nondestructive Testing Technology in Detecting the Pressure Welding Fittings of Transmission Lines ［J］. Journal of Shanghai Jiao Tong University ， 2018 ， 52 （ 10 ）： 1189 - 1194 .

张秋雁，杨忠，姜遇红，等 . 一种基于CNN的航拍输电线路图像分类方法［J］. 应用科技， 2019 ， 46 （ 6 ）： 41 - 45 .

ZHANG Qiuyan ， YANG Zhong ， JIANG Yuhong ， et al . CNN-Based Aerial Image Classification Method for Aerial Transmission Lines ［J］. Applied Science and Technology ， 2019 ， 46 （ 6 ）： 41 - 45 .

赵文清，程幸福，赵振兵，等 . 注意力机制和Faster RCNN相结合的绝缘子识别［J］. 智能系统学报， 2020 ， 15 （ 1 ）： 92 - 98 .

ZHAO Wenqing ， CHENG Xingfu ， ZHAO Zhenbing ， et al . Insulator Recognition Based on Attention Mechanism and Faster RCNN ［J］. CAAI Transactions on Intelligent Systems ， 2020 ， 15 （ 1 ）： 92 - 98 .

DENG C ， WANG S W ， HUANG Z ， et al . Unmanned Aerial Vehicles for Power Line Inspection： A Cooperative Way in Platforms and Communications ［J］. Journal of Communications ， 2014 ， 9 （ 9 ）： 687 - 692 .

王万国，张晶晶，韩军，等 . 基于无人机图像的输电线断股与异物缺陷检测方法［J］. 计算机应用， 2015 ， 35 （ 8 ）： 2404 - 2408 .

WANG Wanguo ， ZHANG Jingjing ， HAN Jun ， et al . Broken Strand and Foreign Body Fault Detection Method for Power Transmission Line Based on Unmanned Aerial Vehicle Image ［J］. Journal of Computer Applications ， 2015 ， 35 （ 8 ）： 2404 - 2408 .

赵建坤，王璋奇，刘世钊 . 基于灰度共生矩阵纹理特征的输电导线识别［J］. 云南电力技术， 2015 ， 43 （ 2 ）： 126 - 129 .

ZHAO Jiankun ， WANG Zhangqi ， LIU Shizhao . Research on Transmission Line Recognition Based on GLCM Texture Feature ［J］. Yunnan Electric Power ， 2015 ， 43 （ 2 ）： 126 - 129 .

王伟，刘国海 . 绝缘子图像的边缘检测［J］. 微计算机信息， 2008 ， 24 （ 27 ）： 308 - 309 .

WANG Wei ， LIU Guohai . Image Edge Detection of the Insulator ［J］. Microcomputer Information ， 2008 ， 24 （ 27 ）： 308 - 309 .

陈晓娟，吴英石，赵亮 . 基于随机Hough变换的OPGW防震锤识别［J］. 黑龙江电力， 2010 ， 32 （ 1 ）： 1 - 2 .

CHEN Xiaojuan ， WU Yingshi ， ZHAO Liang . Identification of OPGW Vibration Damper Based on Random Hough Transformation ［J］. Heilongjiang Electric Power ， 2010 ， 32 （ 1 ）： 1 - 2 .

宋伟，左丹，邓邦飞，等 . 高压输电线防震锤锈蚀缺陷检测［J］. 仪器仪表学报， 2016 ， 37 （ S1 ）： 113 - 117 .

SONG Wei ， ZUO Dan ， DENG Bangfei ， et al . Corrosion Defect Detection of Earthquake Hammer for High Voltage Transmission Line ［J］. Chinese Journal of Scientific Instrument ， 2016 ， 37 （ S1 ）： 113 - 117 .

黄乐，吴功平，叶旭辉 . 输电线巡检机器人弱光条件下的障碍物识别研究［J］. 光学学报， 2018 ， 38 （ 9 ）： 0915006 .

HUANG Le ， WU Gongping ， YE Xuhui . Research on Obstacle Recognition for Transmission Line Inspection Robots Under Low-Light Conditions ［J］. Acta Optica Sinica ， 2018 ， 38 （ 9 ）： 0915006 .

刘进，杨洁，庞瑞帆，等 . 一种基于多尺度特征簇的舰船目标快速定位与识别方法［J］. 武汉大学学报（信息科学版）， 2016 ， 41 （ 1 ）： 111 - 116 .

LIU Jin ， YANG Jie ， PANG Ruifan ， et al . A Method for Fast Ship Detection and Recognition in Sea-Sky Background Based on Multi-scale Feature Cluster ［J］. Geomatics and Information Science of Wuhan University ， 2016 ， 41 （ 1 ）： 111 - 116 .

杨洪朝，席燕辉，项胜，等 . 基于深度学习的输电线路巡检无人机图像分类方法［J］. 湖南电力， 2024 ， 44 （ 1 ）： 53 - 58 .

YANG Hongzhao ， XI Yanhui ， XIANG Sheng ， et al . Image Classification Method Based on Deep Learning for Inspection Unmanned Aerial Vehiclesin Transmission Line ［J］. Hunan Electric Power ， 2024 ， 44 （ 1 ）： 53 - 58 .

谢小瑜，周俊煌，张勇军 . 深度学习在泛在电力物联网中的应用与挑战［J］. 电力自动化设备， 2020 ， 40 （ 4 ）： 77 - 87 .

XIE Xiaoyu ， ZHOU Junhuang ， ZHANG Yongjun . Application and Challenge of Deep Learning in Ubiquitous Power Internet of Things ［J］. Electric Power Automation Equipment ， 2020 ， 40 （ 4 ）： 77 - 87 .

杨罡，孙昌雯，张娜，等 . 基于多尺度特征融合的输电线路关键部件检测［J］. 电测与仪表， 2020 ， 57 （ 3 ）： 54 - 59 .

YANG Gang ， SUN Changwen ， ZHANG Na ， et al . Detection of Key Components of Transmission Lines Based on Multi-scale Feature Fusion ［J］. Electrical Measurement & Instrumentation ， 2020 ， 57 （ 3 ）： 54 - 59 .

ZHAO Z B ， XU G Z ， QI Y C ， et al . Multi-Patch Deep Features for Power Line Insulator Status Classification from Aerial Images ［C］// International Joint Conference on Neural Networks ， Vancouver， BC， Canada， 2016 .

赵振兵，江爱雪，戚银城，等 . 嵌入遮挡关系模块的SSD模型的输电线路图像金具检测［J］. 智能系统学报， 2020 ， 15 （ 4 ）： 656 - 662 .

ZHAO Zhenbing ， JIANG Aixue ， QI Yincheng ， et al . Fittings Detection in Transmission Line Images with SSD Model Embedded Occlusion Relation Module ［J］. CAAI Transactions on Intelligent Systems ， 2020 ， 15 （ 4 ）： 656 - 662 .

赵振兵，马迪雅，丁洁涛，等 . 基于SAW-PCL的输电线路缺销螺栓弱监督检测方法［J］. 北京航空航天大学学报， 2024 ， 50 （ 11 ）： 3319 - 3326 .

ZHAO Zhenbing ， MA Diya ， DING Jietao ， et al . Weak Supervision Detection Method for Pin-Missing Bolts of Transmission Lines Based on SAW-PCL ［J］. Journal of Beijing University of Aeronautics and Astronautics ， 2024 ， 50 （ 11 ）： 3319 - 3326 .

ZHANG K ， ZHAO K ， GUO X W ， et al . HRM-CenterNet： A High-Resolution Real-Time Fittings Detection Method ［C］// IEEE International Conference on Systems， Man， and Cybernetics ， Melbourne， Australia ， 2022 .

WANG N ， ZHANG K ， ZHU J W ， et al . Fittings Detection Method Based on Multi-Scale Geometric Transformation and Attention-Masking Mechanism ［J］. Sensors ， 2023 ， 23 （ 10 ）： 4923 .

ZHANG K ， LOU W S ， WANG J C ， et al . PA-DETR： End-to-End Visually Indistinguishable Bolt Defects Detection Method Based on Transmission Line Knowledge Reasoning ［J］. IEEE Transactions on Instrumentation and Measurement ， 2023 ， 72 ： 5016914 .

CHEN G Q ， WANG S K ， LIU Y K ， et al . Research on Transmission Line Hardware Identification Based on Improved YOLOv5 and DeblurGANv2 ［J］. IEEE Access ， 2023 ， 11 ： 133351 - 133362 .

WANG S S ， TAN W W ， YANG T F ， et al . High-Voltage Transmission Line Foreign Object and Power Component Defect Detection Based on Improved YOLOv5 ［J］. Journal of Electrical Engineering & Technology ， 2024 ， 19 （ 1 ）： 851 - 866 .

汤寓麟，边少锋，翟国君，等 . 侧扫声呐检测沉船目标的改进YOLOv5法［J］. 武汉大学学报（信息科学版）， 2024 ， 49 （ 6 ）： 977 - 985 .

TANG Yulin ， BIAN Shaofeng ， ZHAI Guojun ， et al . An Improved YOLOv5 Method for Shipwreck Target Detection by Side-Scan Sonar Images ［J］. Geomatics and Information Science of Wuhan University ， 2024 ， 49 （ 6 ）： 977 - 985 .

ZHAO Z B ， XIONG J ， HAN Y ， et al . Transmission Line Bolts and Their Defects Detection Method Based on Position Relationship ［J］. Frontiers in Energy Research ， 2023 ， 11 ： 1269087 .

张瑞倩，邵振峰， Aleksei Portnov ，等 . 多尺度空洞卷积的无人机影像目标检测方法［J］. 武汉大学学报（信息科学版）， 2020 ， 45 （ 6 ）： 895 - 903 .

ZHANG Ruiqian ， SHAO Zhenfeng ， PORTNOV A ， et al . Multi-scale Dilated Convolutional Neural Network for Object Detection in UAV Images ［J］. Geomatics and Information Science of Wuhan University ， 2020 ， 45 （ 6 ）： 895 - 903 .

张宇，何楚，石博，等 . 逐层特征选择的多层部件模型用于遥感图像飞机目标检测［J］. 武汉大学学报（信息科学版）， 2014 ， 39 （ 12 ）： 1406 - 1411 .

ZHANG Yu ， HE Chu ， SHI Bo ， et al . Multi-layer Feature Selection Based Hierarchal Component Model for Aero-Plane Detection on Remote Sensing Image ［J］. Geomatics and Information Science of Wuhan University ， 2014 ， 39 （ 12 ）： 1406 - 1411 .

LI Y ， FAN Q ， HUANG H ， et al . A Modified YOLOv8 Detection Network for UAV Aerial Image Recognition ［J］. Drones ， 2023 ， 7 （ 5 ）： 304 .

ZHANG Q L ， YANG Y B . SA-Net： Shuffle Attention for Deep Convolutional Neural Networks ［C］// IEEE International Conference on Acoustics， Speech and Signal Processing ， Toronto， ON， Canada， 2021 .

ZHU X Z ， HU H ， LIN S ， et al . Deformable ConvNets V2： More Deformable， Better Results ［C］// IEEE/CVF Conference on Computer Vision and Pattern Recognition （CVPR）， Long Beach， CA， USA， 2020 .

TONG Z J ， CHEN Y H ， XU Z W ， et al . Wise-IoU： Bounding Box Regression Loss with Dynamic Focusing Mechanism ［EB/OL］. 2023 ： arXiv ： 2301. 10051 ， https：//arxiv.org/abs/2301.10051 https://arxiv.org/abs/2301.10051 .

REN S Q ， HE K M ， GIRSHICK R ， et al . Faster R-CNN： Towards Real-Time Object Detection with Region Proposal Networks ［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence ， 2017 ， 39 （ 6 ）： 1137 - 1149 .

LIU W ， ANGUELOV D ， ERHAN D ， et al . SSD： Single Shot Multibox Detector ［C］// The 14th European Conference on Computer Vision ， Amsterdam， The Netherlands ， 2016 .

WANG C Y ， YEH I H ， LIAO H P . You Only Learn One Representation： Unified Network for Multiple Tasks ［J］. Journal of Information Science and Engineering ， 2021 ， 39 ： 691 - 709 .

WANG A ， CHEN H ， LIU L H ， et al . YOLOv10： Real-Time End-to-End Object Detection ［C］// The 38th Conference on Neural Information Processing Systems ， Vancouver， BC， Canada， 2024 .

LI D H ， LU Y ， GAO Q ， et al . LiteYOLO-ID： A Lightweight Object Detection Network for Insulator Defect Detection ［J］. IEEE Transactions on Instrumentation and Measurement ， 2024 ， 73 ： 5023812 .

LU Y ， LI D H ， LI D ， et al . A Lightweight Insulator Defect Detection Model Based on Drone Images ［J］. Drones ， 2024 ， 8 （ 9 ）： 431 .

HU J ， SHEN L ， SUN G . Squeeze-and-Excitation Networks ［C］// IEEE/CVF Conference on Computer Vision and Pattern Recognition ， Salt Lake City，UT，USA，2018 .

PADILLA R ， PASSOS W L ， DIAS T L B ， et al . A Comparative Analysis of Object Detection Metrics with a Companion Open-Source Toolkit ［J］. Electronics ， 2021 ， 10 （ 3 ）： 279 .

REZATOFIGHI H ， TSOI N ， GWAK J ， et al . Generalized Intersection over Union： A Metric and a Loss for Bounding Box Regression ［C］// IEEE/CVF Conference on Computer Vision and Pattern Recognition （CVPR）， Long Beach， CA， USA， 2020 .

RAIMUNDO A . Insulator Data Set — Chinese Power Line Insulator Dataset （CPLID）［DS/OL］. ［ 2024-10-12 ］ https：//ieee-dataport.org/open0access/insulator-data-set-chinese-power-line-insulator-dataset-cplid https://ieee-dataport.org/open0access/insulator-data-set-chinese-power-line-insulator-dataset-cplid .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

An Information Extraction Method for Statistical Thematic Maps Combining YOLOv8 and Syntactic Model

Autonomous Robotic Mapping for Complex Spaces： Progress， Challenges and Future Perspectives

High-Resolution L-Band InSAR Study of Intense Near-Field Deformation Associated with Shallow Moderate Earthquakes

Research Progress in Eco-Geological Environmental Monitoring and Construction of an Intelligent Monitoring System in the Qinling Mountains

Research Progress in Spatiotemporal Modeling， Analysis， and Prediction of Urban Activity Flows in Built Environment

Related Author

WANG Zhanchu

ZHAO Fei

ZHANG Yaxin

DONG Jiahao

LUAN Guize

WANG Xinrui

DU Qingyun

Related Institution

Institute of International Rivers and Eco⁃Security, Yunnan University

School of Earth Sciences, Yunnan University

Yunnan International Joint Laboratory of China⁃Laos⁃Bangladesh⁃Myanmar Natural Resources Remote Sensing Monitoring

School of Ecology and Environmental Science, Yunnan University

School of Resources and Environmental Sciences, Wuhan University

AI问答

⁰