广东工业大学学报 ›› 2023, Vol. 40 ›› Issue (02): 15-21.doi: 10.12052/gdutxb.220079

• 综合研究 • 上一篇    下一篇

基于YOLOv4-MP的绝缘子爆裂缺陷检测方法

谢国波1, 林立1, 林志毅1, 贺笛轩1, 文刚2   

  1. 1. 广东工业大学 计算机学院，广东 广州 510006;
    2. 云南电网有限责任公司 电力科学研究院，云南 昆明 650011
  • 收稿日期:2022-04-29 出版日期:2023-03-25 发布日期:2023-04-07
  • 通信作者: 林立(1998-),男,硕士研究生,主要研究方向为深度学习、目标检测,E-mail:2112005022@mail2.gdut.edu.cn
  • 作者简介:谢国波(1977-),男,教授,博士,主要研究方向为遥感大数据、深度学习
  • 基金资助:
    国家自然科学基金资助项目(61802072)；广州市科技计划项目(201902020012)

An Insulator Burst Defect Detection Method Based on YOLOv4-MP

Xie Guo-bo1, Lin Li1, Lin Zhi-yi1, He Di-xuan1, Wen Gang2   

  1. 1. School of Computer Science and Technology, Guangdong University of Technology, Guangzhou 510006, China;
    2. Yunnan Electric Power Research Institute, Yunnan Power Grid Co., Ltd, Kunming 650011, China
  • Received:2022-04-29 Online:2023-03-25 Published:2023-04-07

HTML

PDF (PC)

1148

摘要： 针对输电线路绝缘子爆裂缺陷检测中缺陷目标小、背景复杂多样导致检测精度低的问题，提出了一种基于YOLOv4改进的检测算法YOLOv4-MP。首先，为减少复杂背景的干扰，在特征提取网络中嵌入Shuffle Attention注意力模块，使模型能够提取到更加有效的特征信息。其次，为增强特征融合的效果，在空间金字塔池化中引入带空洞的池化层，能够有效增大感受野。最后，为减少低层信息的丢失，采用Mish函数作为路径增强网络的激活函数。实验结果表明，YOLOv4-MP的平均精度均值(Mean Average Precision, mAP) 达到了93.60%，比YOLOv4算法提升了6.37%。与常用的检测算法相比，YOLOv4-MP具有更好的检测性能，对于绝缘子爆裂缺陷检测具有较大应用价值。

关键词: 目标检测, 绝缘子检测, 卷积神经网络, 特征融合, 输电线路巡检

Abstract: Aiming at the problems of small defect targets and complex backgrounds, which leads to low accuracy in insulator burst defect detection, an improved detection algorithm YOLOv4-MP based on YOLOv4 is proposed. First, the Shuffle Attention module is included in the feature extraction network to limit the interference of complicated background, enabling the model to extract more effective feature information. Then, the spatial pyramid pooling is enhanced by the dilated pooling layer, which effectively increases the receptive field and improves the effect of feature fusion. Finally, the Mish function is utilized as the activation function of the path enhancement network to limit the loss of low-level information. The experimental results show that the mean average precision of YOLOv4-MP reaches 93.60%, which is 6.37% higher than that of the YOLOv4 algorithm. Compared with the commonly used detection algorithms, YOLOv4-MP has better detection performance and has high application value for the detection of insulator burst defects.

Key words: object detection, insulator detection, convolutional neural network, feature fusion, transmission line inspection

中图分类号: 

  • TP391.4
[1] SANYAL S, ASLAM F, KIM T, et al. Deterioration of porcelain insulators utilized in overhead transmission lines: a review [J]. Transactions on Electrical and Electronic Materials, 2020, 21(1): 16-21.
[2] MIAO X, LIU X, CHEN J, et al. Insulator detection in aerial images for transmission line inspection using single shot multibox detector [J]. IEEE Access, 2019, 7: 9945-9956.
[3] CHEN M H, TIAN Y N, XING S Y, et al. Environment perception technologies for power transmission line inspection robots [J]. Journal of Sensors, 2021, 2021(2): 1-16.
[4] 杨罡, 孙昌雯, 王大伟, 等. 基于无人机前端和SSD算法的输电线路部件检测模型对比研究[J]. 太原理工大学学报, 2020, 51(2): 212-219.
YANG G, SUN C W, WANG D W, et al. Comparative study of transmission line component detection models based on UAV front end and SSD algorithm [J]. Journal of Taiyuan University of Technology, 2020, 51(2): 212-219.
[5] LIU X, MIAO X, JIANG H, et al. Data analysis in visual power line inspection: an in-depth review of deep learning for component detection and fault diagnosis [J]. Annual Reviews in Control, 2020, 50: 253-277.
[6] TAN P, LI X F, XU J M, et al. Catenary insulator defect detection based on contour features and gray similarity matching [J]. Journal of Zhejiang University - Science A:Applied Physics & Engineering, 2020, 21(1): 64-73.
[7] 闫丽梅, 刘永强, 徐建军, 等. 基于Grabcut分割和填充物面积判别的复合绝缘子断串诊断[J]. 电力系统保护与控制, 2021, 49(22): 114-119.
YAN L M, LIU Y Q, XU J J, et al. Broken string diagnosis of composite insulator based on Grabcut segmentation and filler area discrimination [J]. Power System Protection and Control, 2021, 49(22): 114-119.
[8] ZHAI Y, CHEN R, YANG Q, et al. Insulator fault detection based on spatial morphological features of aerial images [J]. IEEE Access, 2018, 6: 35316-35326.
[9] 王银立, 闫斌. 基于视觉的绝缘子“掉串”缺陷的检测与定位[J]. 计算机工程与设计, 2014, 35(2): 583-587.
WANG Y L, YAN B. Vision based detection and location for cracked insulator [J]. Computer Engineering and Design, 2014, 35(2): 583-587.
[10] 姜云土, 韩军, 丁建, 等. 基于多特征融合的玻璃绝缘子识别及自爆缺陷的诊断[J]. 中国电力, 2017, 50(5): 52-58.
JIANG Y T, HAN J, DING J, et al. The identification and diagnosis of self-blast defects of glass insulators based on multi-feature fusion [J]. Electric Power, 2017, 50(5): 52-58.
[11] ZHAI Y, WANG D, ZHANG M, et al. Fault detection of insulator based on saliency and adaptive morphology [J]. Multimedia Tools and Applications, 2017, 76(9): 12051-12064.
[12] 黄剑航, 王振友. 基于特征融合的深度学习目标检测算法研究[J]. 广东工业大学学报, 2021, 38(4): 52-58.
HUANG J H, WANG Z Y. A Research on deep learning object detection algorithm based on feature fusion [J]. Journal of Guangdong University of Technology, 2021, 38(4): 52-58.
[13] ZHAO W, XU M, CHENG X, 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: 1-8.
[14] 李鑫, 刘帅男, 杨桢, 等. 基于改进Cascade R-CNN的输电线路多目标检测[J]. 电子测量与仪器学报, 2021, 35(10): 24-32.
LI X, LIU S N, YANG Z, et al. Multi-target detection of transmission lines based on improved cascade R-CNN [J]. Journal of Electronic Measurement and Instrumentation, 2021, 35(10): 24-32.
[15] 赵文清, 张海明, 徐敏夫. 面向改进尺度缩放网络的绝缘子识别[J]. 中国图象图形学报, 2021, 26(11) : 2561-2570.
ZHAO W Q, ZHANG H M, XU M F. Insulator recognition based on an improved scale-transferrable network[J] Journal of Image and Graphics, 2021, 26(11) : 2561-2570.
[16] LIU J, LIU C, WU Y, et al. An improved method based on deep learning for insulator fault detection in diverse aerial images [J]. Energies, 2021, 14(14): 4365.
[17] ZHENG R, ZHU L, HU T, et al. Detection of fault insulator of power transmission line based on region-CNN[C]//2020 35th Youth Academic Annual Conference of Chinese Association of Automation (YAC) . Zhanjiang: IEEE, 2020: 73-76.
[18] LIAO G P, YANG G J, TONG W T, et al. Study on power line insulator defect detection via improved faster region-based convolutional neural network[C]//2019 IEEE 7th International Conference on Computer Science and Network Technology (ICCSNT) . Dalian: IEEE, 2019: 262-266.
[19] 唐小煜, 黄进波, 冯洁文, 等. 基于 U-net 和 YOLOv4 的绝缘子图像分割与缺陷检测[J]. 华南师范大学学报 (自然科学版) , 2020, 52(6): 15-21.
TANG X Y, HUANG J B, FENG J W, et al. Image segmentation and defect detection of insulators based on U-net and YOLOv4 [J]. Journal of South China Normal University (Natural Science Edition) , 2020, 52(6): 15-21.
[20] 党宏社, 薛萌, 郭琴. 基于改进的YOLOv4绝缘子掉片故障检测方法[J]. 电瓷避雷器, 2022, 65(1): 211-218.
DANG H S, XUE M, GUO Q. Insulator dropout fault detection method based on improved YOLOv4 [J]. Insulators and Surge Arrester, 2022, 65(1): 211-218.
[21] ZHANG Q L, YANG Y B. Sa-net: Shuffle attention for deep convolutional neural networks[C]//ICASSP 2021-2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) . [S. l. ]: IEEE, 2021: 2235-2239.
[22] MA N, ZHANG X, ZHENG H T, et al. Shufflenet v2: Practical guidelines for efficient cnn architecture design[C]//Proceedings of the European Conference on Computer Vision (ECCV) . Munich: [s. n. ], 2018: 116-131.
[23] CHEN L C, PAPANDREOU G, KOKKINOS I, et al. Deeplab: semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected crfs [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 40(4): 834-848.
[1] 章云, 王晓东. 基于受限样本的深度学习综述与思考[J]. 广东工业大学学报, 2022, 39(05): 1-8.
[2] 杨积升, 章云, 李东. 点云目标检测残差投票网络[J]. 广东工业大学学报, 2022, 39(01): 56-62.
[3] 张国生, 冯广, 李东. 基于姿态表示的航空影像旋转目标检测网络[J]. 广东工业大学学报, 2021, 38(05): 40-47.
[4] 黄剑航, 王振友. 基于特征融合的深度学习目标检测算法研究[J]. 广东工业大学学报, 2021, 38(04): 52-58.
[5] 马少鹏, 梁路, 滕少华. 一种轻量级的高光谱遥感图像分类方法[J]. 广东工业大学学报, 2021, 38(03): 29-35.
[6] 夏皓, 蔡念, 王平, 王晗. 基于多分辨率学习卷积神经网络的磁共振图像超分辨率重建[J]. 广东工业大学学报, 2020, 37(06): 26-31.
[7] 战荫伟, 朱百万, 杨卓. 车辆颜色和型号识别算法研究与应用[J]. 广东工业大学学报, 2020, 37(04): 9-14.
[8] 曾碧卿, 韩旭丽, 王盛玉, 徐如阳, 周武. 基于双注意力卷积神经网络模型的情感分析研究[J]. 广东工业大学学报, 2019, 36(04): 10-17.
[9] 高俊艳, 刘文印, 杨振国. 结合注意力与特征融合的目标跟踪[J]. 广东工业大学学报, 2019, 36(04): 18-23.
[10] 谢岩, 刘广聪. 基于编解码器模型的车道识别与车辆检测算法[J]. 广东工业大学学报, 2019, 36(04): 36-41.
[11] 杨孟军, 苏成悦, 陈静, 张洁鑫. 基于卷积神经网络的视觉闭环检测研究[J]. 广东工业大学学报, 2018, 35(05): 31-37.
[12] 陈旭, 张军, 陈文伟, 李硕豪. 卷积网络深度学习算法与实例[J]. 广东工业大学学报, 2017, 34(06): 20-26.
[13] 申小敏， 李保俊， 孙旭， 徐维超. 基于卷积神经网络的大规模人脸聚类[J]. 广东工业大学学报, 2016, 33(06): 77-84.
[14] 孙伟, 钟映春, 谭志, 连伟烯. 多特征融合的室内场景分类研究[J]. 广东工业大学学报, 2015, 32(1): 75-79.
[15] 陈世文1, 2, 蔡念2, 肖明明3. 基于高斯混合模型和canny算法的运动目标检测[J]. 广东工业大学学报, 2011, 28(3): 87-91.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
版权所有 © 2017 《广东工业大学学报》编辑部
地址：广州市东风东路729号广东工业大学 邮编：510090
电话：020-37626653,020-37626139,020-37626396
邮箱：xbzrb@gdut.edu.cn
本系统由北京玛格泰克科技发展有限公司设计开发