An Improved SVM-Based Motion Detection Algorithm Using an Accelerometer

doi:10.1007/978-981-10-0740-8_13

	An Improved SVM-Based Motion Detection Algorithm Using an Accelerometer
	Wu, Xu-Kang1,2 ; Yang, Xu-Guang 3; Cai, Zheng-Guo 2; Luo, Shan-Shan 2
	2016
会议录名称	ELECTRONICS, COMMUNICATIONS AND NETWORKS V
卷号	382
页码	109-117
发表状态	已发表
DOI	10.1007/978-981-10-0740-8_13
摘要	This paper presents an effective detection algorithm for wrist motions based on an improved SVM (support vector machine) classifier. Firstly, a novel windowing method is proposed to enhance the consistency of sampled motion. After extracting characteristic features in both time and frequency domains, a feature scaling process is applied and a C-SVC (C-Support Vector Classifier) is trained by a threshold-based grid-search with cross-validation, to achieve higher accuracy and faster convergence. The experiments demonstrate that the proposed algorithm outperforms the traditional SVM in accuracy and convergence.
关键词	Motion detection SVM Grid-search Cross-validation
出版地	SINGAPORE, SINGAPORE
会议地点	Shanghai, China
收录类别	CPCI ; EI
语种	英语
WOS研究方向	Engineering ; Telecommunications
WOS类目	Engineering, Electrical & Electronic ; Telecommunications
WOS记录号	WOS:000386917300013
出版者	SPRINGER-VERLAG SINGAPORE PTE LTD
EI入藏号	20162902618038
EI主题词	Algorithms ; Image retrieval ; Motion analysis ; Signal detection
EI分类号	Information Theory and Signal Processing:716.1 ; Computer Software, Data Handling and Applications:723 ; Data Processing and Image Processing:723.2
WOS关键词	RECOGNITION
原始文献类型	Proceedings Paper
引用统计	正在获取...
文献类型	会议论文
条目标识符	https://kms.shanghaitech.edu.cn/handle/2MSLDSTB/2024
专题	信息科学与技术学院信息科学与技术学院_硕士生
通讯作者	Wu, Xu-Kang
作者单位	1.Shanghaitech Univ, 100 Haike Rd, Shanghai, Peoples R China 2.Shanghai Inst Microsyst & Informat Technol, 865 Changning Rd, Shanghai, Peoples R China 3.Shanghai Internet Things Co Ltd, 1455 Pingcheng Rd, Shanghai, Peoples R China
第一作者单位	上海科技大学
通讯作者单位	上海科技大学
第一作者的第一单位	上海科技大学
推荐引用方式 GB/T 7714	Wu, Xu-Kang,Yang, Xu-Guang,Cai, Zheng-Guo,et al. An Improved SVM-Based Motion Detection Algorithm Using an Accelerometer[C]. SINGAPORE, SINGAPORE:SPRINGER-VERLAG SINGAPORE PTE LTD,2016:109-117.