[1]黄 清,张 伟,程 敏,等.一种基于几何约束的跌落式熔断器图像识别方法[J].南京理工大学学报(自然科学版),2019,43(02):217.[doi:10.14177/j.cnki.32-1397n.2019.43.02.014]
 Huang Qing,Zhang Wei,Cheng Min,et al.Image recognition for drop-out fuse based on geometric constraint[J].Journal of Nanjing University of Science and Technology,2019,43(02):217.[doi:10.14177/j.cnki.32-1397n.2019.43.02.014]
点击复制

一种基于几何约束的跌落式熔断器图像识别方法()
分享到:

《南京理工大学学报》(自然科学版)[ISSN:1005-9830/CN:32-1397/N]

卷:
43卷
期数:
2019年02期
页码:
217
栏目:
出版日期:
2019-04-26

文章信息/Info

Title:
Image recognition for drop-out fuse based on geometric constraint
文章编号:
1005-9830(2019)02-0217-07
作者:
黄 清1张 伟2程 敏3苏鹏飞4丁棋炳4郭 毓4
1.国网江苏省电力公司,江苏 南京 210024; 2.国网江苏省电力公司 常州供电公司,江苏 常州 213003; 3.亿嘉和科技股份有限公司,江苏 南京 210012; 4.南京理工大学 自动化学院,江苏 南京 210094
Author(s):
Huang Qing1Zhang Wei2Cheng Min3Su Pengfei4Ding Qibing4Guo Yu4
1.State Grid Jiangsu Electric Power Company,Nanjing 210024,China; 2.Changzhou Power Supply Company,State Grid Jiangsu Electric Power Company,Changzhou 213003,China; 3.Yijiahe Technology Co Ltd,Nanjing 210012,China; 4.School of Automation,Nanjing Univer
关键词:
配电线路维护 跌落式熔断器 几何约束 概率Hough变换 椭圆拟合
Keywords:
distribution line maintenance drop-out fuses geometric constraint probabilistic Hough transform ellipse fitting
分类号:
TP391.4
DOI:
10.14177/j.cnki.32-1397n.2019.43.02.014
摘要:
针对配电线路维护机器人跌落式熔断器的熔丝管识别问题,提出了一种基于几何约束的直线拟合法。采用概率Hough变换,结合直线平行和距离约束等条件,提高了熔丝管管体的识别率。针对操作环识别困难的问题,利用熔丝管管体和操作环的几何位置关系,提出了基于位置和几何尺寸约束的椭圆拟合法识别操作环。为消除复杂背景中其他物体对熔丝管识别的干扰,提出了基于熔丝管管体和操作环互相约束的熔丝管识别方法。实验结果表明,所提方法可以识别不同场景中的熔丝管,识别率高且鲁棒性好。
Abstract:
Aiming at the recognition problem of fuse tube of the drop-out fuse for the distribution line maintenance robot,a straight-line fitting approach based on geometric constraint is proposed here. The accuracy of the fuse tube body recognition is improved by using the probabilistic Hough transform combined with the lines parallelism and distance constraints. In order to recognize the operation ring correctly,a novel ellipse fitting method based on position and geometry size constraint is proposed by means of the geometrical position relation between the fuse tube body and the operating ring. To remove the interference caused by other objects in the complex background,a recognition method for the fuse tube is presented based on the mutually constrain condition of the fuse tube body and the operating ring. Experimental results show that the proposed method can identify fuse tubes in different scenes with high accuracy and good robustness.

参考文献/References:

[1] 胡毅,刘凯,彭勇,等. 带电作业关键技术研究进展与趋势[J]. 高电压技术,2014,40(7):1921-1931.
Hu Yi,Liu Kai,Peng Yong,et al. Research status and development trend of live working key technology[J]. High Voltage Engineering,2014,40(7):1921-1931.
[2]李天友,黄超艺,蔡俊宇. 配电带电作业机器人的发展与展望[J]. 供用电,2016,33(11):43-48.
Li Tianyou,Huang Chaoyi,Cai Junyu. Development and prospect of the live-line working robot in distribution network[J]. Distribution & Utilization,2016,33(11):43-48.
[3]Aracil R,Pinto E,Ferre M. Robots for live-power lines:Maintenance and inspection tasks[C]//Proceedingsof the 15th IFAC. Laxenburg,Austria:IFAC Secretariat,2002:13-18.
[4]Park J Y,Cho B H,Byun S H,et al. Development of cleaning robot system for live-line suspension insulator strings[J]. International Journal of Control,Automation and Systems,2009,7(2):211-220.
[5]Song Y,Wang H,Zhang J. A vision-based broken strand detection method for a power-line maintenance robot[J]. IEEE Transactions on Power Delivery,2014,29(5):2154-2161.
[6]高焕兵,田国会. 面向配电系统的带电抢修作业机器人[J]. 山东大学学报(工学版),2015,45(1):45-53.
Gao Huanbing,Tian Guohui. Live-working robot for emergency repair of power distribution system[J]. Journal of Shandong University(Engineering Science),2015,45(1):45-53.
[7]纪良,吴巍,许春山,等. 基于改进椭圆拟合与非线性支持向量机的配电设备螺栓带电检测[J]. 南京理工大学学报,2017,41(6):708-713.
Ji Liang,Wu Wei,Xu Chunshan,et al. Online detection for bolts of power distribution equipments based on improved ellipse fitting and nonlinear support vector machine[J]. Journal of Nanjing University of Science and Technology,2017,41(6):708-713.
[8]张晶晶,韩军,赵亚博,等. 形状感知的绝缘子识别与缺陷诊断[J]. 中国图象图形学报,2014,19(8):1194-1201.
Zhang Jingjing,Han Jun,Zhao Yabo,et al. Insulator recognition and defects detection based on shape perceptual[J]. Journal of Image and Graphics,2014,19(8):1194-1201.
[9]邵剑雄,闫云凤,齐冬莲. 基于霍夫森林的变电站开关设备检测及状态识别[J]. 电力系统自动化,2016,40(11):115-120.
Shao Jianxiong,Yan Yunfeng,Qi Donglian. Substation switch detection and state recognition based on Hough forests[J]. Automation of Electric Power System,2016,40(11):115-120.
[10]乔寅骐,肖健华,黄银和,等. 基于最小二乘修正的随机Hough变换直线检测[J]. 计算机应用,2015,35(11):3312-3315.
Qiao Yinqi,Xiao Jianhua,Huang Yinhe,et al. Randomized Hough transform straight line detection based on least square correction[J]. Journal of Computer Application,2015,35(11):3312-3315.
[11]Eliseo Stefano Maini. Enhanced direct least square fitting of ellipses[J]. International Journal of Pattern Recognition & Artificial Intelligence,2012,20(6):939-953.
[12]熊风光,李希,韩燮. 基于整体最小二乘的椭圆拟合方法[J]. 微电子学与计算机,2017,34(1):102-105.
Xiong Fengguang,Li Xi,Han Xie. A method of ellipse fitting based on total least squares[J]. Microelectronics & Computer,2017,34(1):102-105.

备注/Memo

备注/Memo:
收稿日期:2018-04-11 修回日期:2018-07-25
基金项目:江苏省重点研发计划项目(BE2017161); 国家电网公司科技项目(SGJSCZ00FZJS1701049; SGJSCZ00FZJS1701074; SGJSCZ00FZJS1601242); 江苏高校优势学科建设工程资助项目(AD20540)
作者简介:黄清(1968-),男,本科,高级工程师,主要研究方向:电网设备状态分析及技术管理,E-mail:huangqing68@vip.sina.com; 通讯作者:郭毓(1964-),女,博士,教授,主要研究方向:智能机器人控制,高精度伺服控制等,E-mail:guoyu@njust.edu.cn。
引文格式:黄清,张伟,程敏,等. 一种基于几何约束的跌落式熔断器图像识别方法[J]. 南京理工大学学报,2019,43(2):217-223.
投稿网址:http://zrxuebao.njust.edu.cn
更新日期/Last Update: 2019-04-26