摘要
智能视频监控在人们生产生活中的应用越来越广,区域入侵检测是智能视频监控的一个重要方面。主要研究了基于目标跟踪的区域入侵检测方法。首先采用混合高斯模型进行运动物体检测,在实际验证中获得了较好的前景检测效果;进一步采用了Mean-Shift算法对从前景检测图像中提取出的运动物体进行实时跟踪,能够实时标记出运动物体并绘制出运动轨迹。采用OpenCV的运动物体跟踪算法框架,在Qt软件开发平台上进行了算法验证,获得了较好的实验效果,充分验证了算法的跟踪效果和实时性,具有实际工程指导价值。
It is more and more widely that intelligent video surveillance is used in people＇s production and life.As we know,regional intrusion detection is a very important aspect of intelligent video surveillance.This paper mainly studies the regional intrusion detection method by using target tracking.Firstly,the moving object is detected by using Gaussian mixture model.In the practical test,foreground detection effect is good.Secondly,the Mean-Shift algorithm is used to track the moving objects that fetched from the foreground image,also with the motion trajectory marked in real time.Finally,the algorithm is validated based on the moving object tracking algorithm framework of OpenCV on Qt platform.The experimental results are very good.It reveal that the method has perfect tracking ability and realtime,and suggest high practical engineering guidance value.
出处
《电子测量技术》
2015年第2期51-54,63,共5页
Electronic Measurement Technology
