道路拥堵视频监控信息智能检测仿真

摘要

对拥挤道路监控视频的信息进行检测,能够有效提升道路的管理质量.对道路拥堵视频监控信息的智能检测,需要计算出道路拥堵视频监控特征值,获取不同特征值变化率指数总和,完成监控信息的检测.传统方法求取道路拥堵视频监控信息特征描述子,提取道路拥堵视频监控信息关键帧,但忽略了不同特征值变化率指数总和的获得,导致检测精度偏低.提出基于改进LBP的道路拥堵视频监控信息智能检测方法.该方法先利用LBP方法对监控下道路状态进行视频序列采集,计算出视频中的道路拥堵视频监控特征值,获取不同特征值变化率指数总和,得到道路边界结构范围,根据不同的道路边界类型选择对应支持向量机进行道路拥堵视频监控信息智能检测.仿真证明,所提方法信息检测精度较高,有效地制止或控制道路拥堵视频监控事故的发生.%Monitoring the information of video in congested road can effectively improve the quality of road management.The intelligent detection of road congestion video information needs to calculate the characteristic value of video monitoring in road congestion.The traditional method ignores the total change rate index of different characteristic value,leading to the low precision of detection.An intelligent detection method of video monitoring information in road congestion based on improved LBP is proposed.Firstly,the LBP method is used to collect video sequence of road condition under monitoring,and the characteristic value of road congestion monitoring video is calculated,then sum of change rate of different characteristic values are obtained.Moreover,the range of road boundary structure is obtained.According to the different types of road boundaries,the corresponding support vector machine is chosen to intelligently detect video surveillance information in congestion road.Simulation results show that the proposed method has high precision of information detection,which can effectively prevent or control the occurrence of monitoring accident of road congestion.