Human Activity Recognition with Smart Watch based on H-SVM

Tao Tang; Qingxiang Zheng; Shaolin Weng; Ao Peng; Huiru Zheng

Human Activity Recognition with Smart Watch based on H-SVM

Tao Tang, Qingxiang Zheng, Shaolin Weng, Ao Peng, Huiru Zheng

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

11 Downloads (Pure)

Abstract

Activity recognition allows ubiquitous wearable device like smart watch to simplify the study and experiment. It is very convenient and extensibil-ity that we do study with the accelerometer sensor of a smart watch. In this paper, we use Samsung GEAR smart watch to collect data, then extract features, classify with H-SVM (Hierarchical Support Vector Machine) classifier and identify hu-man activities classification. Experiment results show great effect at low sam-pling rate, such as 10 Hz and 5 Hz, which will give us the energy saving. In most cases, the accuracies of activity recognition experiment are above 99%.

Original language	English
Title of host publication	Unknown Host Publication
Publisher	Association for Computing Machinery
Number of pages	8
Publication status	Accepted/In press - 8 Apr 2016
Event	The 5th International Conference on Frontier Computing (FC 2016) - Duration: 8 Apr 2016 → …

Conference

Conference	The 5th International Conference on Frontier Computing (FC 2016)
Period	8/04/16 → …

Keywords

Human Activity Recognition
Smart Watch
H-SVM

Access to Document

FC 2016 paper 35Accepted author manuscript, 374 KB

http://uir.ulster.ac.uk/35147/2/FC_2016_paper_35_acceptance.pdf

Fingerprint Dive into the research topics of 'Human Activity Recognition with Smart Watch based on H-SVM'. Together they form a unique fingerprint.

Cite this

@inproceedings{9e6d6592793f4dc98ae4c24f495f3568,

title = "Human Activity Recognition with Smart Watch based on H-SVM",

abstract = "Activity recognition allows ubiquitous wearable device like smart watch to simplify the study and experiment. It is very convenient and extensibil-ity that we do study with the accelerometer sensor of a smart watch. In this paper, we use Samsung GEAR smart watch to collect data, then extract features, classify with H-SVM (Hierarchical Support Vector Machine) classifier and identify hu-man activities classification. Experiment results show great effect at low sam-pling rate, such as 10 Hz and 5 Hz, which will give us the energy saving. In most cases, the accuracies of activity recognition experiment are above 99%.",

keywords = "Human Activity Recognition, Smart Watch, H-SVM",

author = "Tao Tang and Qingxiang Zheng and Shaolin Weng and Ao Peng and Huiru Zheng",

year = "2016",

month = apr,

day = "8",

language = "English",

booktitle = "Unknown Host Publication",

publisher = "Association for Computing Machinery",

address = "United States",

note = "The 5th International Conference on Frontier Computing (FC 2016) ; Conference date: 08-04-2016",

}

TY - GEN

T1 - Human Activity Recognition with Smart Watch based on H-SVM

AU - Tang, Tao

AU - Zheng, Qingxiang

AU - Weng, Shaolin

AU - Peng, Ao

AU - Zheng, Huiru

PY - 2016/4/8

Y1 - 2016/4/8

N2 - Activity recognition allows ubiquitous wearable device like smart watch to simplify the study and experiment. It is very convenient and extensibil-ity that we do study with the accelerometer sensor of a smart watch. In this paper, we use Samsung GEAR smart watch to collect data, then extract features, classify with H-SVM (Hierarchical Support Vector Machine) classifier and identify hu-man activities classification. Experiment results show great effect at low sam-pling rate, such as 10 Hz and 5 Hz, which will give us the energy saving. In most cases, the accuracies of activity recognition experiment are above 99%.

AB - Activity recognition allows ubiquitous wearable device like smart watch to simplify the study and experiment. It is very convenient and extensibil-ity that we do study with the accelerometer sensor of a smart watch. In this paper, we use Samsung GEAR smart watch to collect data, then extract features, classify with H-SVM (Hierarchical Support Vector Machine) classifier and identify hu-man activities classification. Experiment results show great effect at low sam-pling rate, such as 10 Hz and 5 Hz, which will give us the energy saving. In most cases, the accuracies of activity recognition experiment are above 99%.

KW - Human Activity Recognition

KW - Smart Watch

KW - H-SVM

M3 - Conference contribution

BT - Unknown Host Publication

PB - Association for Computing Machinery

T2 - The 5th International Conference on Frontier Computing (FC 2016)

Y2 - 8 April 2016

ER -