Abnormal event detection via covariance matrix for optical flow based feature

doi:10.1007/s11042-017-5309-2

CORC > 兰州理工大学 > 兰州理工大学 > 新能源学院

	Abnormal event detection via covariance matrix for optical flow based feature
	Wang, Tian 4; Qiao, Meina 4; Zhu, Aichun 3; Niu, Yida 4; Li, Ce 2; Snoussi, Hichem 1
刊名	MULTIMEDIA TOOLS AND APPLICATIONS
	2018-07
卷号	77 期号:13 页码:17375-17395
关键词	Global abnormal event Local abnormal event Multi-RoI Covariance matrix Optical flow
ISSN号	1380-7501
DOI	10.1007/s11042-017-5309-2
英文摘要	Abnormal event detection is one of the most important objectives in security surveillance for public scenes. In this paper, a new high-performance algorithm based on spatio-temporal motion information is proposed to detect global abnormal events from the video stream as well as the local abnormal event. We firstly propose a feature descriptor to represent the movement by adopting the covariance matrix coding optical flow and the corresponding partial derivatives of multiple connective frames or the patches of the frames. The covariance matrix of multi-RoI (region of interest) which consists of frames or patches can represent the movement in high accuracy. For public surveillance video, the normal samples are abundant while there are few abnormal samples. Thus the one-class classification method is suitable for handling this problem inherently. The nonlinear one-class support vector machine based on a proposed kernel for Lie group element is applied to detect abnormal events by merely training the normal samples. The computational complexity and time performance of the proposed method is analyzed. The PETS, UMN and UCSD benchmark datasets are employed to verify the advantages of the proposed method for both global abnormal and local abnormal event detection. This method can be used for event detection for a surveillance video and outperforms the state-of-the-art algorithms. Thus it can be adopted to detect the abnormal event in the monitoring video.
资助项目	Fundamental Research Funds for the Central Universities[YWF-14-RSC-102]
WOS研究方向	Computer Science ; Engineering
语种	英语
出版者	SPRINGER
WOS记录号	WOS:000439750300062
状态	已发表
内容类型	期刊论文
源URL	[http://119.78.100.223/handle/2XXMBERH/32595]
专题	新能源学院电气工程与信息工程学院
通讯作者	Zhu, Aichun
作者单位	1.Univ Technol Troyes, Inst Charles Delaunay, LM2S, UMR STMR 6279,CNRS, Troyes, France 2.Lanzhou Univ Technol, Coll Elect & Informat Engn, Lanzhou, Gansu, Peoples R China 3.Nanjing Tech Univ, Sch Comp Sci & Technol, Nanjing, Jiangsu, Peoples R China 4.Beihang Univ, Sch Automat Sci & Elect Engn, Beihang, Peoples R China
推荐引用方式 GB/T 7714	Wang, Tian,Qiao, Meina,Zhu, Aichun,et al. Abnormal event detection via covariance matrix for optical flow based feature[J]. MULTIMEDIA TOOLS AND APPLICATIONS,2018,77(13):17375-17395.
APA	Wang, Tian,Qiao, Meina,Zhu, Aichun,Niu, Yida,Li, Ce,&Snoussi, Hichem.(2018).Abnormal event detection via covariance matrix for optical flow based feature.MULTIMEDIA TOOLS AND APPLICATIONS,77(13),17375-17395.
MLA	Wang, Tian,et al."Abnormal event detection via covariance matrix for optical flow based feature".MULTIMEDIA TOOLS AND APPLICATIONS 77.13(2018):17375-17395.