智能视频技术在电力系统领域的应用  

Intelligent Video Technology Applications in the Field of Power Systems

作　　者：周封[1] 刘闻博 刘志刚 王丙全[1] 刘健[1] 王晨光[1] 

机构地区：[1]哈尔滨理工大学电气与电子工程学院,黑龙江哈尔滨150080 [2]中国海诚工程科技股份有限公司,上海200031

出　　处：《哈尔滨理工大学学报》2015年第5期14-19,共6页Journal of Harbin University of Science and Technology

基　　金：黑龙江省科技计划(攻关)项目(GZ11A213);黑龙江省教育厅科学技术研究面上项目(12541176)

摘　　要：针对人工监测无法时刻保障电力系统安全稳定运行的问题,依据智能视频技术的原理及发展现状,分析了该技术在电力系统已有的和可能的应用,包括对无人值守变电站的各项检测、对高压开关柜柜内器件的检测以及对野外高压输电线路的各项检测.重点探讨了为实现警戒面检测、遗留物检测、刀闸状态检测、杆塔检测等应用所需的背景差分法,对目标图像的灰度提取来进行输电线路覆冰检测的自适应算法,对于漏油、火焰等矩特征提取的算法,应用Kalman滤波及Mean-shift算法实现人员徘徊、导线舞动、导线弧垂越限等运动目标跟踪检测算法,提出了该领域未来的研究和发展方向.For the problem that manual monitoring is unable to guarantee the safety and stability of power sys- tem all the time, based on the principle and development status of intelligent video technology, analyzing the exist- ing and potential applications of the technology in power system, including the detections of unmanned substations, the detections of components in high tension switch cabinet and the detections of high tension transmission wire op- eration in the countryside. This paper focused on the calculus of differences about alert condition detection, rem- nant detection, disconnector state detection, pole-towers detection, etc, the adaptive algorithms of iced electric transmission wire detecting by extracting the grayscale of target image, the algorithm of extracting leaked oil and flame, and applying Kalman Filter and Mean-shift algorithm to achieve moving target tracking algorithm such as staff hovering, wire waving, wire sagging. Finally, the future direction of the research and development in this field was proposed well.

关 键 词：智能视频 电力系统 目标检测 目标跟踪 

分 类 号：TM7[电气工程—电力系统及自动化]