基于深度学习的室内场景人员检测方法研究.pdf

《基于深度学习的室内场景人员检测方法研究.pdf》由会员上传分享，免费在线阅读，更多相关内容在学术论文-天天文库。

1、分类号TP957学号14063014UDC00493密级公开工学硕士学位论文基于深度学习的室内场景人员检测方法研究硕士生姓名李玉乐学科专业计算机科学与技术研究方向计算机视觉指导教师窦勇研究员国防科学技术大学研究生院二〇一七年一月PeopleDetectioninIndoorSceneBasedonDeepLearningMethodCandidate:LiYuleAdvisor:ProfessorDouYongAdissertationSubmittedinpartialfulfillmentoftherequireme

2、ntsforthedegreeofMasterofEngineeringinComputerScienceandTechnologyGraduateSchoolofNationalUniversityofDefenseTechnologyChangsha,Hunan,P.R.ChinaJanuary3,2017国防科学技术大学研究生院硕士学位论文目录摘要.....................................................................................

3、...................................iABSTRACT...............................................................................................................iii第一章绪论....................................................................................................

4、........11.1研究背景...................................................................................................11.2研究现状...................................................................................................21.2.1有监督学习的目标检测方法.......................

5、.................................21.2.2基于无监督的目标检测方法........................................................71.3本文工作...................................................................................................81.3.1基于卷积网络和递归网络的室内场景人员检测.......................

6、.81.3.2基于区域特征和局部特征融合的人员检测................................81.3.3基于视频的无监督室内场景人员检测........................................81.4论文结构...................................................................................................8第二章基于卷积网络和递归网络的室内场景人员检测.............

7、.......................102.1引言...........................................................................................................102.1.1相关工作........................................................................................102.2基于卷积网络和递归网络的模型........

8、...................................................122.2.1模型整体框架................................................................................122.2.2LSTM介绍..