摘要
针对城市安全问题,为实现视频监控的可信预警、辨别潜在的突发事件并及时处理,基于Hadoop搭建了身份识别应急联防移动处置平台,采用HDFS Streaming数据流形式读取视频数据,利用MapReduce框架在计算集群中对视频数据进行分布式分析,使用OpenCV集成的LBPH算法实现视频目标的有效跟踪。平台功能覆盖了样本信息采集、隐患发现、信息发布、应急处置等安全管理应急联防的全过程,可直接服务于应急现场管理,具有极强的适应性和便捷性。
In response to urban security issues,in order to achieve credible warning,identify potential emergencies and deal with them in a timely manner,this paper builds an identity recognition emergency joint defense mobile processing platform based on Hadoop,uses HDFS Streaming to read video data,and uses the MapReduce framework to calculate Distributed analysis of video data in the cluster,using OpenCV integrated LBPH algorithm to achieve effective tracking of video targets.The platform functions cover the entire process of safety management and emergency joint defense,such as sample information collection,hidden danger discovery,information release,and emergency response.It can directly serve the emergency site management,with strong adaptability and convenience.
作者
张立志
李学文
Zhang Lizhi;Li Xuewen(College of Engineering,Xi′an International University,Xi′an 710077,China;Xi′an Xike Security Technology Co.,Ltd.,Xi′an 710054,China;College of Safety Science and Engineering,Xi′an University of Science and Technology,Xi′an 710054,China)
出处
《电子技术应用》
2022年第4期90-93,97,共5页
Application of Electronic Technique
基金
陕西省教育厅专项科研计划项目(18JK1131)。
