摘要
针对传统视频监控系统利用人工查看视频,效率低,浪费人力物力的问题,论文设计了一套基于Open CV人脸识别和神经网络预测目标路径的视频监控追踪系统。系统采用Qt框架,利用Qt中QLable显示视频监控信息,利用Open CV技术对视频图像进行面部特征提取与人脸特征识别,结合GIS,通过SQL Sever Data Base管理了道路、监控设备位置等空间位置信息,利用Open Gl展示了空间数据,并通过神经网络预测模型对目标的下一个途经地点进行了预测,实现了系统对监控区域的实时视频监控。测试表明,该系统能够快速高效地进行目标追踪和预测以及辅助生成围堵方案。
Aiming at the problem of traditional video surveillance system which use manual viewing video lead to low efficiency,waste of human and material resources,this paper designs a video surveillance and tracking system based on Open CV face recognition and neural network prediction target path. The system uses the Qt framework to display the video monitoring information in QLable in Qt,uses Open CV technology to carry out face extraction and face recognition of video images,and uses GIS to use the SQL Sever database to manage the spatial location information such as road and monitoring equipment location,and use Open Gl the spatial data is displayed and the next route of the target is predicted by the neural network prediction model,and the real-time video monitoring of the monitoring area is realized. Tests show that the system can quickly and efficiently target tracking and forecasting and assist in generating containment programs.
出处
《城市勘测》
2018年第1期95-97,共3页
Urban Geotechnical Investigation & Surveying
基金
湖北省教育厅科学技术研究项目资助(D20141301)