[1]潘 翀,沈鹏飞,张 忠,等.基于无人机巡检图像的绝缘子串实时定位研究[J].电瓷避雷器,2020,(01):234-240.[doi:10.16188/j.isa.1003-8337.2020.01.039]
 PAN Chong,SHEN Pengfei,ZHANG Zhong,et al.Research on Real-Time Positioning of Insulator Strings Based on UAV Inspection Images[J].,2020,(01):234-240.[doi:10.16188/j.isa.1003-8337.2020.01.039]
点击复制

基于无人机巡检图像的绝缘子串实时定位研究()
分享到:

《电瓷避雷器》[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2020年01期
页码:
234-240
栏目:
绝缘子
出版日期:
2020-02-20

文章信息/Info

Title:
Research on Real-Time Positioning of Insulator Strings Based on UAV Inspection Images
作者:
潘 翀 沈鹏飞 张 忠 王 博 朱如桂 张 颖
(国网安徽省电力公司马鞍山供电公司, 安徽 马鞍山 243000)
Author(s):
PAN Chong SHEN Pengfei ZHANG Zhong WANG Bo ZHU Rugui ZHANG Ying
(StateGrid Anhui Electric Power Company Maanshan Power Supply Company,Maanshan 243000, China)
关键词:
绝缘子串 散射变换 实时定位 卷积神经网络 SSD网络框架
Keywords:
insulator string scattering transformation real-time location CNN SSD network framework
DOI:
10.16188/j.isa.1003-8337.2020.01.039
摘要:
针对无人机巡检输电线路图像中存在绝缘子串定位难点,笔者在分析散射变换原理和卷积神经网络(CNN)的基础上,通过低通滤波器作散射系数处理,结合Gram矩阵法来降低绝缘子串背景信息的噪声干扰,以增强低频系数的边缘纹理特征,结合SSD网络框架实现了CNN对绝缘子串实时定位计算的高效性。实验结果表明:该方法在保证实时计算的前提下,与传统SSD网络框架相比,召回率和交并比分别提升了1.04%和1.38%。
Abstract:
In view of the difficulties in locating insulator strings in transmission line images by UAV inspection, this paper analyzes the principle of scattering transformation and convolution neural network(CNN), processes the scattering coefficient through low-pass filter, and combines Gram matrix method to reduce the noise interference of insulator string background information, so as to enhance the edge ripple of low-frequency coefficient. Combined with the SSD network framework, the efficiency of CNN in real-time positioning of insulator strings is achieved. Test results show that compared with the traditional SSD network framework, the proposed method can improve the recall rate and the delivery-union ratio by 1.04% and 1.38% respectively under the premise of guaranteeing real-time computation.

参考文献/References:

[1] 黄成才,李永刚.基于复合绝缘子污闪模型的闪络特性分析[J].电瓷避雷器,2014(4):31-36. HUANG Chengcai, LI Yonggang. Flashover characteristics analysis based on contamination flashover model of composite insulators[J]. Insulators and Surge Arresters, 2014(4): 31-36.
[2] 张志劲,张东东,袁超,蒋兴良,刘小欢,胡建林.污秽成分对XP-160绝缘子串交流闪络特性的影响[J].高电压技术,2014, 40(7):1970-1976. ZHANG Zhijin, ZHANG Dongdong, YUAN Chao, JIANG Xingliang, LIU Xiaohuan, HU Jianlin.Effect of contamination component on AC flashover performance of insulator string XP-160[J]. High Voltage Engineering, 2014, 40(7): 1970-1976.
[3] 翟永杰,王迪,伍洋,程海燕.基于骨架提取的航拍绝缘子图像分步识别方法[J].华北电力大学学报:自然科学版,2015(3):105-110. ZHAI Yongjie, WANG Di, WU Yang, CHENG Haiyan. Step recognition method of aerial photo insulator image based on skeleton extraction[J]. Journal of North China Electric Power University, 2015(3): 105-110.
[4] 王俊波,余红波,陈贤熙,张伟忠.非接触式超声波检测10 kV线路绝缘子事故隐患的应用探讨[J].电瓷避雷器,2014(3):27-31. WANG Junbo, YU Hongbo, CHEN Xianxi, ZHANG Weizhong. Discussion on the application of non-contact ultrasonic insulator accident detection on 10 kV line[J]. Insulators and Surge Arresters, 2014(3): 27-31.
[5] 谷波.红外紫外检测技术在特高压输电线路中的运用[J].低碳世界,2017(12):43-44. GU Bo. The application of valley wave infrared ultraviolet detection technology to UHV transmission lines[J]. Low Carbon World, 2017(12): 43-44.
[6] 张德钦,庞玉海,钟宇,高冰洁.基于无人机平台的复合绝缘子检测方法[J].自动化技术与应用,2018,37(03):111-115. ZHANG Deqin, PANG Yuhai, ZHONG Yu, GAO Bingjie. Composite insulator detection method based on UAV platform [J].Automation technology and application, 2018,37(03): 111-115.
[7] 彭向阳,钟清,饶章权,杨必胜,陈驰,苏林晓.基于无人机紫外检测的输电线路电晕放电缺陷智能诊断技术[J].高电压技术,2014, 40(8):2292-2298. PENG Xiangyang, ZHONG Qing, RAO Zhangquan, YANG Bisheng, CHEN Chi, SU Linxiao. Intelligent diagnostic techniques of transmission lines corona discharge defect based on ultraviolet detection from unmanned aerial vehicle[J]. High Voltage Engineering, 2014, 40(8): 2292-2298.
[8] 何思远,刘刚,王玲,唐延东.基于无人机的输电线路设备识别方法研究[J].红外与激光工程,2013, 42(7):1940-1944. HE Siyuan, LIU Gang, WANG Ling, TANG Yandong. Research on identification method of transmission line equipment based on UAV[J]. Infrared and Laser Engineering, 2013, 42(7): 1940-1944.
[9] 何洪英,姚建刚,王玲.一种基于Bayes估计的小波自适应绝缘子红外图像去噪方法[J].电工技术学报,2006(1):37-41. HE Hongying, YAO Jiangang, WANG Ling. A wavelet adaptive infrared image denoising method for insulators based on bayes estimation[J]. Transactions of China Electrotechnical Society, 2006(1): 37-41.
[10] 卢航,姚建刚,付鹏.基于总体最小二乘的Shearlet自适应零值绝缘子红外图像去噪[J].红外技术,2015, 37(10):842-846. LU Hang, YAO Jiangang, FU Peng. Smalllet adaptive zero-value insulator infrared image denoising based on total least squares[J]. Infrared Technology, 2015, 37(10): 842-846.
[11] 柯洪昌,付浩海,孔德刚.基于改进的Hough变换的绝缘子检测方法[J].科技创新与应用,2017(31):6. KE Hongchang, FU Haohai, KONG Degang.Insulator detection method based on improved Hough transform[J].Innovation and application of science and technology, 2017(31): 6.
[12] LECOCHE F, GABORIT G, GILLETTE L A, GRAU A, DUVILLARET L. Contactless, real time, and vectorial inspection of a multiwire power cable voltage using an Electro-Optic technique[J]. IEEE Sensors Journal, 2014, 14(8): 2881-2888.
[13] 高强,孟格格.基于卷积神经网络的绝缘子故障识别算法研究[J].电测与仪表,2017, 54(21):30-36. GAO Qiang, MENG Gege. Insulator fault recognition algorithm based on convolution neural network[J]. Electrical Measurement & Instrumentation, 2017, 54(21): 30-36.
[14] 刘莉莉,田丽媛.散射变换理论研究进展[J].成都大学学报:自然科学版,2014, 33(2):145-147. LIU Lili, TIAN Liyuan. Progress in the study of scattering transformation theory[J]. Journal of Chengdu University(Natural Science), 2014, 33(2): 145-147.
[15] 俞翔,朱岱寅,张劲东,蒋锐.基于设计结构化Gram矩阵的ISAR运动补偿方法[J].电子学报,2014, 42(3):452-461. YU Xiang, ZHU Daiyin, ZHANG Jindong, JIANG Rui. ISAR motion compensation method based on structured gram matrix design[J]. Journal of Electronics, 2014, 42(3): 452-461.
[16] 陈金辉,叶西宁.行人检测中非极大值抑制算法的改进[J].华东理工大学学报:自然科学版,2015, 41(3):371-378. CHEN Jinhui, YEXining. Improvement of non-maximum suppression algorithm in pedestrian detection[J]. Journal of East China University of Science and Technology(Natural Science Edition), 2015, 41(3): 371-378.

备注/Memo

备注/Memo:
收稿日期:2018-05-15作者简介:潘翀(1980—),男,工程师,主要研究方向:电力系统自动化。
更新日期/Last Update: 2020-02-20