复杂背景下输电线路缺陷检测算法研究

doi:10.11896/jsjkx.230500178

摘要/Abstract

摘要： 输电线路定期巡检对保障电力系统安全稳定运行具有重要的意义。针对输电线路航拍图像背景复杂、目标尺度变化大、小目标多等问题,提出了基于YOLOv5s的输电线路目标检测算法。该算法采用特征细化模块优化微小目标特征,并在网络中嵌入SimAM注意力模块,通过能量函数统一权值的方式优化模型的特征提取,最后引入NWD损失函数削弱模型对小目标位置偏差的敏感性,提升模型对小目标的识别检测能力。实验结果表明,该模型对输电线路目标的平均检测精度高达98.8%,相较于基准模型,提高了1.2%。

关键词: 输电线路缺陷检测, 注意力机制, NWD损失, 特征细化

Abstract: Regular inspection of transmission lines is of great significance to ensure the safe and stable operation of power systems.For the problems of complex background of transmission line aerial images,large changes in target scale and many small targets,a transmission line target detection algorithm based on YOLOv5s is proposed.The algorithm adopts feature refinement module to optimize tiny target features,and embeds SimAM attention module in the network to optimize the feature extraction of the model by means of unified weights of energy functions,and finally introduces NWD loss function to weaken the sensitivity of the model to small target position deviation and improve the recognition and detection ability of the model for small targets.Experimental results show that the average detection accuracy of the model for transmission line targets is as high as 98.8%,which is 1.2% higher compared with the benchmark model.

Key words: Transmission line defect detection, Attention mechanism, NWD loss, Feature refinement

中图分类号:

TP391

邬春明, 王调君. 复杂背景下输电线路缺陷检测算法研究[J]. 计算机科学, 2024, 51(6A): 230500178-6. https://doi.org/10.11896/jsjkx.230500178

WU Chunming, WANG Tiaojun. Study on Defect Detection Algorithm of Transmission Line in Complex Background[J]. Computer Science, 2024, 51(6A): 230500178-6. https://doi.org/10.11896/jsjkx.230500178

参考文献

[1]National Energy Administration.2016-2021 National Power Reliability Annual Report[R].http://prpq.nea.gov.cn/ndbg/8540.htm.
[2]CHEN K,LIU X,JIA L J,et al.Insulator defect detection based on lightweight network and enhanced multi-scale feature fusion[J].HighVoltage Technology,2013:1-14.
[3]ZHOU M,WANG J,LI B.ARG-Mask RCNN:An Infrared Insulator Fault-Detection Network Based on Improved Mask RCNN[J].Sensors,2022,22(13):4720.
[4]ZHAO W Q,XU M F,CHENG X F,et al.An insulator in transmission lines recognition and fault detection model based on improved Faster RCNN[J].IEEE Transactions on Instrumentation and Measurement,2021,70:5016408.
[5]CHEN Y Q,WANG H X,SHEN J,et al.Application of Data-Driven iterative learning algorithm in transmission line defect detection[J].Scientific Programming,2021,2120(Pt3):9976209.1-9976209.9.
[6]LI H,DONG Y,LIU X,et al.Abnormal target detection method for transmission lines based on two-stage deep network[J].Control and Decision,2022,37(7):1873-1882.
[7]GAO Z,YANG G,LI E,et al.Novel feature fusion module-based detector for small insulator defect detection[J].IEEE Sensors Journal,2021,21(15):16807-16814.
[8]XIAO Y,LI Z,ZHANG D,et al.Detection of pin defects in ae-rial images based on cascaded convolutional neural network[J].IEEE Access,2021,9:73071-73082.
[9]ZHAO Z B,QI H Y,QI Y C,et al.Detection method based on automatic visual shape clustering for pin-missing defect in transmission lines[J].IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT,2020,69(9):6080-6091.
[10]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 Cybernetics-Systems,2020,50(4):1486-1498.
[11]LI F,XIN,B,CHEN T,et al.An Automatic Detection Method of Bird’s Nest on Transmission Line Tower Based on Faster RCNN[J].IEEE Access,2020,8:164214-164221.
[12]WEN Q D,LUO Z Q,CHEN R T,et al.Deep learning approaches on defect detection in high resolution aerial images of insulators[J].Sensors,2021,21(4):1033.
[13]HE H,LI Y C,YANG J,et al.Pin-missing defect recognitionbased on feature fusion and spatial attention mechanism[J].Energy Reports,2022,8:656-663.
[14]XU S Y,DENG J C,HUANG Y R,et al.Research on insulator defect detection based on an improved mobilenetV1-YOLOv4[J].Entropy,2022,24(11):1-18.
[15]ZHENG H B,SUN Y H,LIU X H,et al.Infrared image detection of substation insulators using an improved fusion single shot multibox detector[J].IEEE Transactions on Power Deli-very,2021,36(6):3351-3359.
[16]LIU X Y,MIAO X R,JIANG H,et al.Box-Point detector:A diagnosis method for insulator faults in power lines using aerial images and convolutional neural networks[J].IEEE TRANSACTIONS ON POWER DELIVERY,2021,36(6):3765-3773.
[17]LIU C Y,WU Y Q,LIU J J,et al.Improved YOLOv3 network for insulator detection in aerial images with diverse background interference[J].Electronics,2021,10(7):771-791.
[18]LIU C Y,WU Y Q,LIU J J,et al.MTI-YOLO:A light-weight and real-time deep neural network for insulator detection in complex aerial images[J].Energies,2021,14(5):1426-1446.
[19]SHI C X,HUANG Y P.Cap-Count guided weakly supervised insulator cap missing detection in aerial images[J].IEEE Sensors Journal,2021,21(1) 685-691.
[20]ZHANG J,WANG J Y,ZHANGS H.An ultra-lightweight and ultra-fast abnormal target identification network for transmission line[J].IEEE Sensors Journal,2021,21(20):23325-23334.
[21]LIU J,JIA R,LI W,et al.High precision detection algorithmbased on improved RetinaNet for defect recognition of transmission lines[J].Energy Reports,2020,6:2430-2440.
[22]WANG C Y,YUAN H,WU H Y,et al.CSP-Net:A new backbone that can enhance learning capability of cnn[C]//Procee-dings of the IEEE Conference on Computer Vision and Patten.IEEE,2020.
[23]LIN T Y,DOLLAR P,GIRSHICK R,et al.Feature pyramidnetworks for object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition(CVPR).Norway,2017:2117-2125.
[24]LIU S,QI L,QIN H F,et al.Path aggregation network for instance segmentation[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition(CVPR).Norway,2018:8759-8768.
[25]YANG L,ZHANG R Y,LI L,et al.SimAM:A Simple,Parameter-Free Attention Module for Convolutional Neural Networks[C]//International Conference on Machine Learning.PMLR,2021.
[26]WANG J W,XU C,YANG W,et al.A Normalized GaussianWasserstein Distance for Tiny Object Detection[C]//Computer Vision and Pattern Recognition(CVPR).2022.
[27]LIU W,ANGUELOV D,ERHAN D,et al.SSD:Single ShotMultiBox Detector[C]//European Conference on Computer Vision.Cham:Springer,2016.
[28]NI H,WANG M,ZHAO L.An improved Faster R-CNN for defect recognition of key components of transmission line[J].Mathematical Biosciences and Engineering:MBE,2021,18(4):4679-4695.
[29]XIA H Y,YANG B H,LI Y L,et al.An improved CenterNet model for insulator detection using aerial imagery[J].Sensors,2022,22(8):1-17.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed