无人值守变电站监控视频异常模式识别方法

为提高变电站视频监控的智能化水平,提出了一种识别无人值守变电站环境监控视频中异常模式的方法.对变电站环境监控中的运动目标进行分类(涉及到人、动物、普通火焰(红黄颜色火焰)、白色火焰、白炽灯),提取多种特征,基于混淆矩阵产生层次化分类器结构,以支持向量机(SVM)作为基本的两类分类器,对于分类精度不理想的SVM,通过Ad...

A recognition method of abnormal patterns for video surveillance in unmanned substation

In order to improve the intelligent level of monitoring and timely detect abnormalities,an identification method of abnormal patterns of surveillance video in unmanned substation environment was proposed in this paper.Object classification was implemented in substation environment video surveillance(related to people,animals,ordinary flames(red and yellow flames),white flames,incandescent lamps).Multiple features were extracted.Hierarchical classifier structure was generated from the confusion matrix.Support vector machine(SVM) was used as the basic binary classifier.AdaBoost algorithm applies weighted voting on SVM whose classification accuracy was not ideal.Simulation experiment using actual video data was implemented,and experimental results show that the proposed method can get a better recognition for people,animals and flame and eliminate interferences such as the impact of incandescent lamps.It can provide the necessary conditions for fire alarm and steal precaution in unattended substation.