摘要
针对当前无人值守变电站遥视系统由人工监视的弊端,提出了在无人值守变电站遥视系统中加入视频分析的新构想,以提高监控水平和效果。在比较了当前多种常用视频分析算法优劣的基础上,择优选用了一种基于柯西分布统计模型的运动检测算法,并对模型的建立和维护进行了改进;采用ARM+Linux构成嵌入式系统进行了实验,实验结果表明这种系统具有较好的效果。
Aiming at the defects in current remote supervision of unmanned substations, a new idea by introducing visual analysis into remote supervision of unmanned substations to improve the level of surveillance is presented. By comparing the advantages and disadvantages of some commonly used methods for visual analysis, a motion detection algorithm based on the Cauchy statistical model is selected and the definition and maintenance of the coefficients of the model are improved. The results of experiment using an embedded system constructed by ARM＋ Linux show that the system has fairly good performance.
出处
《电力系统自动化》
EI
CSCD
北大核心
2005年第22期77-80,共4页
Automation of Electric Power Systems
关键词
无人值守变电站
遥视系统
运动检测
背景模型
参数估计
unmanned substation
remote viewing system
motion detecting
background model
parameter estimation
