SIA OpenIR  > 工艺装备与智能机器人研究室
基于多传感器信息融合的绝缘子串自动识别方法
Alternative TitleThe Method to Automatic Recognition of Insulator Based on Multi-Sensor Information Fusion
刘澈1,2; 姜勇1; 姚森敬3; 刘国伟3
Department工艺装备与智能机器人研究室
Source Publication机械设计与制造
ISSN1001-3997
2019
Issue4Pages:220-224
Contribution Rank1
Funding Organization国家自然科学基金资助项目(51535008) ; 南方电网公司重点项目(SZ-KJXM-20150901)
Keyword绝缘子串 激光雷达 相机 特征融合 支持向量机 证据理论
Abstract绝缘子串的自主识别是变电站带电水冲洗机器人执行绝缘子串冲洗任务的重要前提。针对变电站内电力设备复杂的架设结构和室外光照条件的复杂多变问题,提出一种基于SVM与证据相容度改进D-S证据理论方法,融合相机光学图像特征和激光点云图像特征的变电站支柱绝缘子串自主识别方法。通过构建光学图像中特征曲线的灰度周期性特征矩阵和灰度阈值特征矩阵,激光点云图像中特征曲线的深度重复性特征矩阵和宽度特征矩阵,将四种特征矩阵作为特征证据,通过SVM与证据相容度方法改进的D-S证据理论进行融合,得到结果,实现对三种不同形状支柱绝缘子串的自主识别。实验结果表明,该方法可以克服图像逆光和复杂背景等干扰,准确识别出绝缘子串的类型。
Other AbstractAutomatic recognition of isolator string is the basis on Live Washing Robot to complete the washing task. Aiming at the environment of substation,the condition of illumination is variable and the installation of substation is complicated. A new method to automatically recognize isolator by D-S evidence theory based on SVM and evidence compatibility. The information fusion system is composed of industrial camera and laser radar. By the optical images,the gray characteristic curve is acquired to extract gray repetitive matrix and gray threshold matrix. By the optical depth images,the depth characteristic curve is acquired to extract depth repetitive matrix and width matrix. The features of every matrix are fussed by the improvement Dempster-Shafer evidence theory to complete automatic recognition of three types isolators. The experimental results show that the new method can overcome the backlight environment and has the ability to recognize isolator accuracy.
Language中文
Document Type期刊论文
Identifierhttp://ir.sia.cn/handle/173321/24595
Collection工艺装备与智能机器人研究室
Corresponding Author刘澈
Affiliation1.中国科学院沈阳自动化研究所机器人国家重点实验室
2.中国科学院大学
3.深圳供电局有限公司
Recommended Citation
GB/T 7714
刘澈,姜勇,姚森敬,等. 基于多传感器信息融合的绝缘子串自动识别方法[J]. 机械设计与制造,2019(4):220-224.
APA 刘澈,姜勇,姚森敬,&刘国伟.(2019).基于多传感器信息融合的绝缘子串自动识别方法.机械设计与制造(4),220-224.
MLA 刘澈,et al."基于多传感器信息融合的绝缘子串自动识别方法".机械设计与制造 .4(2019):220-224.
Files in This Item: Download All
File Name/Size DocType Version Access License
基于多传感器信息融合的绝缘子串自动识别方(504KB)期刊论文出版稿开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[刘澈]'s Articles
[姜勇]'s Articles
[姚森敬]'s Articles
Baidu academic
Similar articles in Baidu academic
[刘澈]'s Articles
[姜勇]'s Articles
[姚森敬]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[刘澈]'s Articles
[姜勇]'s Articles
[姚森敬]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: 基于多传感器信息融合的绝缘子串自动识别方法.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.