基于端到端算法的绝缘子检测技术研究

doi:10.3969/j.issn.1674-1951.2021.02.005

华电技术 ›› 2021, Vol. 43 ›› Issue (2): 28-33.doi: 10.3969/j.issn.1674-1951.2021.02.005

基于端到端算法的绝缘子检测技术研究

肖新帅(), 田秀霞^*(), 徐曼

上海电力大学计算机科学与技术学院,上海 200090

收稿日期:2020-07-30 修回日期:2021-02-02 出版日期:2021-02-25 发布日期:2021-03-05
通讯作者: 田秀霞
作者简介:肖新帅（1996—）,男,山东聊城人,在读硕士研究生,从事目标检测方面的研究（E-mail： 1519116524@qq.com）。|徐曼（1975—）,女,上海人,讲师,工学硕士,从事计算机视觉方面的研究。
基金资助:
国家自然科学基金重点项目(61532021);国家自然科学基金面上项目(61772327)

Research on insulator detection technology based on end-to-end algorithm

XIAO Xinshuai(), TIAN Xiuxia^*(), XU Man

School of Computer Science and Technology,Shanghai University of Electric Power, Shanghai 200090,China

Received:2020-07-30 Revised:2021-02-02 Online:2021-02-25 Published:2021-03-05
Contact: TIAN Xiuxia

摘要/Abstract

摘要：

绝缘子是电力系统中十分重要的电工原件,因此研究绝缘子目标的检测尤为重要。传统的识别方法难以充分利用图像的信息且准确率较低,而深度学习在图像识别与图像检测中取得了良好的效果。介绍了端到端的深度学习目标检测方法（YOLOv1,SSD,YOLOv2）,利用自制的绝缘子数据集进行试验并对检测结果进行了对比,结果表明：利用端到端的深度学习算法能够完成对绝缘子的识别和定位,在保持绝缘子检测性能的情况下,可以提升检测速度,满足实时电力巡检的需要。

关键词: 绝缘子检测, 图像识别, 图像检测, 深度学习, 数据集, 端到端检测算法, 目标检测算法

Abstract:

Insulators are important electrical components in power systems, so it is important to study the target detection of insulators. Traditional recognition methods are of low utilization rate of image information and low accuracy. With the development of deep learning, good recognition results have been achieved in image identification and image detection. End-to-end deep learning target detection methods （YOLOv1,SSD,YOLOv2）are used in testing a custom dataset of an insulator and the results are compared. The experimental results show that the end-to-end deep learning algorithm can identify and locate the insulator. Maintaining the current detection performance, the method can improve the detection speed for insulators and meet the requirement of real-time power inspection.

Key words: insulator detection, image identification, image detection, deep learning, dataset, end-to-end detection algorithm, target detection algorithm

中图分类号:

TP399：TM755

肖新帅, 田秀霞, 徐曼. 基于端到端算法的绝缘子检测技术研究[J]. 华电技术, 2021, 43(2): 28-33.

XIAO Xinshuai, TIAN Xiuxia, XU Man. Research on insulator detection technology based on end-to-end algorithm[J]. Huadian Technology, 2021, 43(2): 28-33.

图/表 10

图1

图2

图3

图4

图5

图6

表1

图7

图8

表2

参考文献 17

[1]	GIRSHICK R, DONAHUE J, DARRELL T, et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2014:580-587.
[2]	GIRSHICK R. Fast R-CNN[C]// Proceedings of the IEEE International Conference on Computer Vision, 2015:1440-1448.
[3]	REN S, HE K, GIRSHICK R, et al. Faster R-CNN: Towards real-time object detection with region proposal networks[J]. EEE Transactions on Pattern Analysis and Machine Intelligence, 2015,39(6):91-95.
[4]	JOSEPH R, SANTOSH D, ROSS G, et al. You only look once: Unified, real-time object detection[J]. Computer Vision and Pattern Recognition, 2016,2:779-788.
[5]	LIU W, ANGUELOV D, ERHAN D, et al. SSD: Single shot multi box detector[C]// European Conference on Computer Vision, 2016:267-279.
[6]	REDMON J, FARHADI A. YOLO9000: Better, faster, stronger[C]// IEEE Conference on Computer Vision & Pattern Recognition, 2017:6517-6525.
[7]	苑津莎, 崔克彬, 李宝树. 基于 ASIFT 算法的绝缘子视频图像的识别与定位[J]. 电测与仪表, 2015,7(7):106-112.
	YUAN Jinsha, CUI Kebin, LI Baoshu. Recognition and location of insulator video image based on ASIFT algorithm[J]. Electrical Measurement and Instrumentation, 2015,7(7):106-112.
[8]	DALAL N, TRIGGS B. Histograms of oriented gradients for human detection[J]. Computer Vision and Pattern Recognition, 2005,1:886-893.
[9]	GIRSHICK R, IANDOLA F, DARRELL T, et al. Deformable part models are convolutional neural networks[C]// 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2015.
[10]	FELZENSZWALB P F, GIRSHICK R, MCALLESTER D, et al. Object detection with discriminatively trained part-based models[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010,32(9):1627-1645. doi: 10.1109/TPAMI.2009.167 pmid: 20634557
[11]	HE K, ZHANG X, REN S, et al. Spatial pyramid pooling in deep convolutional networks for visual recognition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015,37(9):1904-1916. doi: 10.1109/TPAMI.2015.2389824 pmid: 26353135
[12]	王万国, 田兵, 刘越. 基于 RCNN 的无人机巡检图像电力小部件识别研究[J]. 地球信息科学学报, 2017,19(2):256-263.
	WANG Wanguo, TIAN Bing, LIU Yue. Research on recognition of power components of UAV inspection images based on RCNN[J]. Journal of Earth Information Science, 2017,19(2):256-263.
[13]	LI Jianan, WEI Yunchao, LIANG Xiaodan, et al. Attentive contexts for object detection[J]. IEEE Transactions on Multimedia, 2017,19(5):944-954. doi: 10.1109/TMM.2016.2642789
[14]	聂凡杰. 基于端到端的深度学习目标检测算法研究[D]. 北京:北京邮电大学, 2018.
[15]	JAN H, RODRIGO B, BERNT S. A convent for non-maximum suppression[C]// International Conference on Learning Representations, 2016:68-81.
[16]	陈景文, 周鑫, 张蓉. 基于 U-net 网络的航拍绝缘子检测[J]. 陕西科技大学学报, 2018,23(4):13-15.
	CHEN Jingwen, ZHOU Xin, ZHANG Rong. U-net network-based aerial insulator testing[J]. Journal of Shanxi University of Science and Technology, 2018,23(4):13-15.
[17]	XIAN Tao, ZHANG Dapeng, WANG Zihao, 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.

检测框架	训练日期	mAP	F_s
YOLOv1	2007年	63.4	45
SSD	2007年	74.3	46
YOLOv2	2007年	78.6	40

检测框架	检测时间/s	阈值（I_U）
YOLOv1	1.376	0.36
SSD	1.786	0.65
YOLOv2	2.234	0.82

基于端到端算法的绝缘子检测技术研究

Research on insulator detection technology based on end-to-end algorithm

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 10

参考文献 17

相关文章 3

编辑推荐

Metrics

本文评价

[1]	吕政权, 李朝阳, 王海峰, 陈怡君, 彭道刚. 基于GRU-CNN的综合能源网络安全攻击检测方法[J]. 华电技术, 2021, 43(2): 9-14.
[2]	李小斌, 张红娜, 曲凯阳, 李凤臣 . 核能集中供热系统优越性分析[J]. 华电技术, 2020, 42(11): 69-82.
[3]	马青;管保安;张扬;李乳演. 基于图像识别技术的电力系统在线监测[J]. 华电技术, 2011, 33(11): 31-34.