Show simple item record

dc.contributor.advisor雷蕴奇
dc.contributor.author谢毅雄
dc.date.accessioned2016-01-13T09:08:45Z
dc.date.available2016-01-13T09:08:45Z
dc.date.issued2013-12-13 10:17:09.0
dc.identifier.urihttps://dspace.xmu.edu.cn/handle/2288/79519
dc.description.abstract随着社会和科技的不断进步和发展,监控系统越来越多地进入到我们的生活。然而,大多数的监控系统是非智能化的,它们需要耗费大量的人力物力去干预和监督,因此,监控系统的智能化需求越来越迫切。计算机视觉技术为监控系统的智能化提供了一种有效的方法。 监控系统的智能化需要稳健而有效的行为识别和事件发现技术。本文主要针对智能监控中对人体行为的识别和事件发现,同时构建了一个并行化的系统,对视频中的单人或多人的行为做出快速实时的分类和识别,并对此进行分析和理解。 首先,本文对视频中运动人体目标的检测方法进行了研究,对现在比较流行的背景建模方法和阴影去除方法进行了对比,采用了效果较好的VIBE背景建模算法和基于...
dc.description.abstractWith the continuous progress and development of society and technology, supervising system is more and more coming into our life. However, most of the monitoring system is not intelligent,they need to spend a lot of manpower and material resources to intervene and supervision. Therefore, the need of intelligent control system becomes more and more urgent. Computer vision technology provides an eff...
dc.language.isozh_CN
dc.relation.urihttps://catalog.xmu.edu.cn/opac/openlink.php?strText=38585&doctype=ALL&strSearchType=callno
dc.source.urihttps://etd.xmu.edu.cn/detail.asp?serial=39354
dc.subject多分类器
dc.subjectReliefF特征选择
dc.subject形式语言与自动机
dc.subjectMultiple classfifer
dc.subjectReliefF feature selection
dc.subjectformal language and automata
dc.title并行环境下的人体行为识别和事件发现技术的研究
dc.title.alternativeResearch on the techonology of human behavior recognition and event discovery in the parallel environment
dc.typethesis
dc.date.replied2013-06-04
dc.description.note学位:工程硕士
dc.description.note院系专业:信息科学与技术学院计算机科学系_计算机技术
dc.description.note学号:23020101153024


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record