Video target tracking algorithm based on multi-feature adaptive fusion

doi:10.12733/jcis6243

	Video target tracking algorithm based on multi-feature adaptive fusion
	Wang, Jinhua; Cao, Jie; Li, Wei; Li, Jun
刊名	Journal of Computational Information Systems
	2013-06-01
卷号	9 期号:11 页码:4617-4628
关键词	Linear regression Monte Carlo methods Target tracking Adaptive fusion Feature fusion Object Tracking Observation information Particle filter Statistical linear regressions Unscented particle filters Unscented transformations
ISSN号	15539105
DOI	10.12733/jcis6243
英文摘要	For video target tracking under complex environment, a new Hybrid Iterated Unscented Particle Filter is proposed. New unscented transformations algorithms are optimized using statistical linear regression method, this algorithm not only improve filtering accuracy, also reduce algorithm of time consumption. Object color histogram and SIFT features as a complementary observation information are adaptive fused within the framework of the new algorithm, better overcome the object light mutations and poor tracking stability problems under partial occlusion. Experiments show that this algorithm for object tracking with high-precision and strong robustness under complicated environment. © 2013 by Binary Information Press.
语种	英语
出版者	Binary Information Press, P.O. Box 162, Bethel, CT 06801-0162, United States
内容类型	期刊论文
源URL	[http://ir.lut.edu.cn/handle/2XXMBERH/113205]
专题	电气工程与信息工程学院理学院
作者单位	College of Electrical and Information Engineering, Lanzhou University of Technology, Lanzhou 730050, China
推荐引用方式 GB/T 7714	Wang, Jinhua,Cao, Jie,Li, Wei,et al. Video target tracking algorithm based on multi-feature adaptive fusion[J]. Journal of Computational Information Systems,2013,9(11):4617-4628.
APA	Wang, Jinhua,Cao, Jie,Li, Wei,&Li, Jun.(2013).Video target tracking algorithm based on multi-feature adaptive fusion.Journal of Computational Information Systems,9(11),4617-4628.
MLA	Wang, Jinhua,et al."Video target tracking algorithm based on multi-feature adaptive fusion".Journal of Computational Information Systems 9.11(2013):4617-4628.