基于运动区域检测的视频缩减技术
首发时间:2011-11-24
摘要:在视频监控领域,随着大量廉价摄像头的部署,每天产生的巨量视频给后续的视频阅览,存储造成困难。本文提出了一种基于背景差分和运动区域检测的可复原的视频缩减技术,该技术一方面检测并挑拣视频含有"有用信息"(运动区域)的视频帧组成缩减视频,另一方面记录下没有运动信息的视频片段的起始和结束时间以及在此期间的公共背景,在复原阶段根据这些信息复原出原监控视频。?
For information in English, please click here
Mixture of Gaussian model based video condensation
Abstract:With the huge amount of cameras are being deployed in cities, a large amount of video data have caused reviewing those video to be a very difficult task. In this paper, based on the mixture of gaussian model background subtraction and moving area detection, a novel method to condense the surveillance video data is proposed. By scaning moving areas in the video and remembering the start and end time of useless video chips, the algorithm could also restore the video to it's original length.
Keywords: video condensation video summarization background subtraction video surveillance
基金:
论文图表:
引用
No.****
同行评议
共计0人参与
勘误表
基于运动区域检测的视频缩减技术
评论
全部评论0/1000