一种高速密集视频监控场景背景重构方法  

Background Reconstruction of High Speed Dense Surveillance Scenes

作　　者：王创新[1] 刘仲云[2] 

机构地区：[1]长沙理工大学电气与信息工程学院,长沙410004 [2]长沙理工大学数学与计算科学学院,长沙410004

出　　处：《数据采集与处理》2012年第3期346-352,共7页Journal of Data Acquisition and Processing

基　　金：国家自然科学基金(10771022)资助项目

摘　　要：针对高速密集视频监控序列建立了一种新的简单的背景重构方法。该方法首先基于帧差序列的时空分布特性,利用高阶统计量理论,获取视频序列公共背景区域;然后根据同一背景帧差图像分布特性相似性,去除运动对于背景的干扰,形成路面背景序列值,从而获得路面背景图像;最后利用计分牌监测的自适应背景更新方法进行背景更新。实验结果表明该方法效果理想,为高速路视频背景重建和运动目标检测提供了新的方法。A new and simple technique is proposed to reconstruct the background of high speed dense surveillance sequences. First, the public background of video sequences is extracted by using the space distribution property of a frame difference sequence and the high-order statistics theory. Then, a sequence of road background is formed by removing the background interference caused by motion according to the distribution similarity of a background frame difference image,thus getting a road background image. Finally, the background image is refreshed via an adaptive method of the scoreboard. Results demonstrate the apparent effectiveness of the scheme and it is a new method for reconstructing background and detecting motion target in high Key speed video.

关 键 词：视频监控 背景重构 运动目标 高速密集运动 公共背景区 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]