Target Tracking Algorithm Based on HOG Feature and Sparse Representation

CORC > 兰州理工大学 > 兰州理工大学

	Target Tracking Algorithm Based on HOG Feature and Sparse Representation
	Li, Ming; Fang, Qingsong
	2016
关键词	visual tracking HOG feature sparse representation classifier
卷号	80
页码	411-416
英文摘要	In this paper, we propose a novel algorithm to deal with the problem of visual tracking in some challenging situations, which is based on HOG feature and sparse representation. First of all, describe target according to the HOG feature; secondly, construct the appearance model of target with the sparse representation, and then predict the target position on the basis of the particle filter method. At last, apply Naive Bayes classifier to track target. The experiment results show that the proposed algorithm is superior in accuracy than the classical tracking algorithm and has better robustness in the scene that contains the target posture changes, illumination variations and occlusion.
会议录	PROCEEDINGS OF THE 2016 5TH INTERNATIONAL CONFERENCE ON ADVANCED MATERIALS AND COMPUTER SCIENCE
会议录出版者	ATLANTIS PRESS
会议录出版地	29 AVENUE LAVMIERE, PARIS, 75019, FRANCE
语种	英语
WOS研究方向	Computer Science ; Materials Science
WOS记录号	WOS:000388289100087
内容类型	会议论文
源URL	[http://119.78.100.223/handle/2XXMBERH/36383]
专题	兰州理工大学
通讯作者	Li, Ming
作者单位	LanZhou Univ Technol, Sch Comp & Commun, Lanzhou 730050, Peoples R China
推荐引用方式 GB/T 7714	Li, Ming,Fang, Qingsong. Target Tracking Algorithm Based on HOG Feature and Sparse Representation[C]. 见:.