Expressiveness of time domain features for detectingdifferent types of human movements

Author

Rybina, Kateryna ; Ternovoy, Maksym ; Dargie, Waltenegus

Author_Institution

Dept. of Inf.-Telecommun. Networks, Nat. Tech. Univ. Of Ukraine "Kyiv Polytech. Inst.", Kiev, Ukraine

fYear

2010

fDate

23-27 Feb. 2010

Firstpage

98

Lastpage

98

Abstract

The usage of wireless activity recognition systems is significant in many applications. The aim of this paper is to show the expressive power of time domain features based on the case study of correlation coefficient feature for detection of the following types of movement: hopping, running, jumping, balancing and skipping. Time domain analyses were applied to acceleration data and a confusion matrix was established as movement classification method.

Keywords

correlation methods; feature extraction; image recognition; time-domain analysis; balancing; confusion matrix; correlation coefficient feature; hopping; human movement classification method; human movements detection; jumping; running; skipping; time domain features; wireless activity recognition systems; Acceleration; Accelerometers; Biomedical monitoring; Computer networks; Computer science; Computer vision; Conferences; Time domain analysis; Time measurement; Training data; Confusion matrix; Test data; Time domain feature analyses; Training data;

fLanguage

English

Publisher

ieee

Conference_Titel

Modern Problems of Radio Engineering, Telecommunications and Computer Science (TCSET), 2010 International Conference on

Conference_Location

Lviv-Slavske

Print_ISBN

978-966-553-875-2

Electronic_ISBN

978-966-553-901-8

Type

conf

Filename

5446052