|本期目录/Table of Contents|

[1]陈智羽,闵 锋*.基于改进YOLO V3的接触网绝缘子检测方法[J].武汉工程大学学报,2020,42(04):462-466.[doi:10.19843/j.cnki.CN42-1779/TQ.201906027]
 CHEN Zhiyu,MIN Feng*.Detection Method of Catenary Insulator Based on Improved YOLO V3[J].Journal of Wuhan Institute of Technology,2020,42(04):462-466.[doi:10.19843/j.cnki.CN42-1779/TQ.201906027]
点击复制

基于改进YOLO V3的接触网绝缘子检测方法(/HTML)
分享到:

《武汉工程大学学报》[ISSN:1674-2869/CN:42-1779/TQ]

卷:
42
期数:
2020年04期
页码:
462-466
栏目:
机电与信息工程
出版日期:
2021-01-28

文章信息/Info

Title:
Detection Method of Catenary Insulator Based on Improved YOLO V3
文章编号:
1674 - 2869(2020)04 - 0462 - 05
作者:
陈智羽闵 锋*
武汉工程大学计算机科学与工程学院,湖北 武汉 430205
Author(s):
CHEN ZhiyuMIN Feng*
School of Computer Science and Engineering,Wuhan Institute of Technology,Wuhan 430205,China
关键词:
YOLO V3铁路接触网绝缘子检测残差块
Keywords:
YOLO V3 railway catenary insulator detection residual block
分类号:
TP391.4
DOI:
10.19843/j.cnki.CN42-1779/TQ.201906027
文献标志码:
A
摘要:
为代替人工对4C巡检车拍摄铁路接触网图像进行分析,使检测的速度和准确率达到实用的要求,本文提出一种基于改进YOLO V3的接触网绝缘子检测方法。该方法在YOLO V3的网络结构Darknet-53的第二个残差块和第三个残差块中间再增加一个新的对小目标友好的4倍降采样的残差块,提高对小目标的检测准确率。并根据相似图像中绝缘子的位置大体相同的特点,通过感知哈希算法分类图像,对同类图像采用候选区域扫描策略加快检测速度。实验结果表明改进后的方法对绝缘子检测的准确率从93.6%提升至99.2%,同类图像的检测速度提升了46%。
Abstract:
To replace the manual analysis of railway caternary images taken by 4C inspection vehicles and meet the practical requirements for the time efficiency and accuracy of detection, we proposed a detection algorithm for catenary insulators based on improved You Only Look Once V3(YOLO V3). The YOLO V3 network Darknet-53 was optimized to improve the detection accuracy of small targets by adding an elaborated 4-fold downsampling residual block that is friendly to small targets between the second and the third residual blocks. According to the characteristic that catenary insulators may locate at the same positions in similar images, we divided catenary images into different categories based on the perceptual hash algorithm. Then, for each type of images, a candidate region scanning strategy was used to speed up the detection of insulators. The experimental results show that our approach can improve the accuracy of insulator detection from 93.6% to 99.2%, and the time efficiency is improved by 46%.

参考文献/References:

[1] 铁运[2012]136号,高速铁路供电安全检测监测系统(6C)系统总体技术规范[S]. 北京:中国标准出版社,2012. [2] 张荣,李伟平,莫同. 深度学习研究综述[J]. 信息与控制, 2018, 47(4): 385-397. [3] 华夏,王新晴,王东,等. 基于改进SSD的交通大场景多目标检测[J]. 光学学报, 2018, 38(12):221-231. [4] 王文秀, 傅雨田, 董峰, 等. 基于深度卷积神经网络的红外船只目标检测方法[J]. 光学学报, 2018,436(7):160-166. [5] 罗海波,许凌云,惠斌,等. 基于深度学习的目标跟踪方法研究现状与展望[J]. 红外与激光工程, 2017, 46(5):6-12. [6] 陈希彤,卢涛. 基于全局深度分离卷积残差网络的高效人脸识别算法[J]. 武汉工程大学学报, 2019,41(3):276-282. [7] 单玲玉,闵锋,李延达. 全局阈值与局部阈值相结合的视网膜血管分割方法[J]. 武汉工程大学学报, 2015,37(3):66-71. [8] GIRSHICK R. Fast R-CNN[C] //I EEE International Conference on Computer Vision and Pattern Recognition. Brussels: IEEE, 2015:1440-1448. [9] REN S Q, HE K M, GIRSHICK R, et al. Faster R-CNN: towards real-Ttime object detection with region proposal networks[C]//IEEE Transactions on Pattern Analysis and Machine Intelligence. Boston: IEEE.2017: 1137- 1149. [10] HE K M, GKIOXARI G. DOLLAR P, et al. Mask R-CNN[C]//IEEE Transactions on Pattern Analysis and Machine Intelligence.Boston: IEEE. 2017: 386-397. [11] LIU W, ANGUELOV D, ERHAN D, et al. SSD: single shot multibox detector[C]//European Conference on Computer Vision.Cham:Springer.2016: 21-37. [12] REDMON J,DIVVALA S,GIRSHICK R, et al. You only look once: unified, real-time object detection[C]// Computer Vision and Pattern Recognition.Las Vegas:IEEE 2016: 779-788. [13] REDMON J , FARHADI A . YOLOv3: an incremental improvement[[EB/OL]. [2018-4-8]. https://arxiv.org/abs/1804.02767. [14] REDMON J, FARHADI A. YOLO9000: better, faster, stronger[C]//Computer Vision and Pattern Recognition.Hawaii :IEEE. 2017: 6517-6525. [15] 杨罡,张娜,晋涛,等. 基于YOLO算法的无人机航拍图片绝缘子缺失检测[J]. 科技资讯,2018, 16(24):30-31. [16] 高强,廉启旺. 航拍图像中绝缘子目标检测的研究[J]. 电测与仪表,2019, 56(5):119-123.

相似文献/References:

备注/Memo

备注/Memo:
收稿日期:2019-06-28基金项目:湖北省技术创新重大专项基金(2019AAA045)。作者简介:陈智羽,硕士研究生,E-mail:438824368@qq.com*通讯作者:闵 锋,博士,副教授,E-mail:123018298@qq.com引文格式:陈智羽,闵锋. 基于改进YOLO V3的接触网绝缘子检测方法究[J]. 武汉工程大学学报,2020,42(4):462-466.
更新日期/Last Update: 2020-08-13