摘要
本文对视频序列中的运动目标检测方法进行了研究,提出了一种基于背景差分的运动目标检测算法。首先通过建立混合高斯模型,获取背景图像,并对背景进行实时更新。然后通过背景差分法确定运动目标区域,再进行图像处理,得到较为精确的运动目标。最后利用八连通域算法,对多个目标进行分割。实验证明,该方法能够快速、准确的检测出多个运动目标。
According to the result of moving object detection research on video,a novel moving object detection algorithm based on background difference is presented.Firstly,obtain the background image by rebuilding mixture Gaussian model , and update the back- ground real-timely.Secondly,a moving object region by background difference algorithm is obtained,then,image processing made the moving object areas more accurate.At last,the moving objects are segmented by selecting a eight connected domain algorithm.Experi- ment results demonstrate that the algorithm can detect the moving objects quickly and exactly.
出处
《微计算机信息》
2009年第21期257-258,282,共3页
Control & Automation
基金
基金申请人:黄山
项目名称:智能化视频交通事件检测系统
基金颁发部门:四川省经济委员会(07XM020)
关键词
运动目标检测
混合高斯模型
背景差分
八连通域
moving object detection
mixture Gaussian model
background difference
eight connected domain
