Large Scale Crowd Analysis Based on Convolutional Neural Network

CASIA OpenIR > 模式识别实验室

	Large Scale Crowd Analysis Based on Convolutional Neural Network
	Lijun, Cao; Kaiqi, Huang
发表期刊	Pattern Recognition
	2015-06-06
卷号	48 期号:10 页码:3016–3024
摘要	Nowadayscrowdsurveillanceisanactiveareaofresearch.Crowdsurveillanceisalwaysaffectedby variousconditions,suchasdifferentscenes,weather,ordensityofcrowd,whichrestrictsthereal application. Thispaperproposesaconvolutionalneuralnetwork(CNN)basedmethodtomonitorthe number ofcrowd flow,suchasthenumberofenteringorleavingpeopleinhighdensitycrowd.Itusesan indirect strategyofcombiningclassification CNNwithregressionCNN,whichismorerobustthanthe direct way.Alargeenoughdatabaseisbuiltwithlotsofrealvideosofpublicgates,andplentyof experimentsshowthattheproposedmethodperformswellundervariousweatherconditionsnomatter either indaytimeoratnight.
关键词	Crowd Analysis Flow Counting Cnn Large Scaledata
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/11816
专题	模式识别实验室
通讯作者	Kaiqi, Huang
作者单位	中国科学院自动化研究所
第一作者单位	中国科学院自动化研究所
通讯作者单位	中国科学院自动化研究所
推荐引用方式 GB/T 7714	Lijun, Cao,Kaiqi, Huang. Large Scale Crowd Analysis Based on Convolutional Neural Network[J]. Pattern Recognition,2015,48(10):3016–3024.
APA	Lijun, Cao,&Kaiqi, Huang.(2015).Large Scale Crowd Analysis Based on Convolutional Neural Network.Pattern Recognition,48(10),3016–3024.
MLA	Lijun, Cao,et al."Large Scale Crowd Analysis Based on Convolutional Neural Network".Pattern Recognition 48.10(2015):3016–3024.

条目包含的文件		下载所有文件
文件名称/大小	文献类型	版本类型	开放类型	使用许可
large scale crowd an（11466KB）	期刊论文	作者接受稿	开放获取	CC BY-NC-SA	浏览下载