Multi-category human motion recognition based on MEMS inertial sensing data

Guangyi Shi; Yuexian Zoui; Yufeng Jin; Yali Zheng; Wen J. Li

doi:10.1109/NEMS.2009.5068625

Multi-category human motion recognition based on MEMS inertial sensing data

Guangyi Shi, Yuexian Zoui^*, Yufeng Jin, Yali Zheng, Wen J. Li^*

^*Corresponding author for this work

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

3 Citations (Scopus)

Abstract

This paper presents multi-category human motion recognition methods based on MEMS inertial sensing data. A Micro Inertial Measurement Unit (pIMU) that is 56mm*23mm*15mm in size was built. This unit consists of three dimensional MEMS accelerometers, gyroscopes, a Bluetooth module and a MCU (Micro Controller Unit), which can record and transfer inertial data to a computer through serial port wirelessly. Five categories of human motion were recorded including walking, running, going upstairs, fall and standing. Fourier transform was used to extract the feature from the human motion data. The concentrated information was finally used to categorize the human motions through CNN (Cascade Neural Network) SVM (Support Vector Machine) and HMM (Hidden Markov Model) respectively. Experimental results showed that for the given 5 human motions, HMM have the best classification result with correct recognition rate range from 90%-100%. © 2009 IEEE.

Original language	English
Title of host publication	4th IEEE International Conference on Nano/Micro Engineered and Molecular Systems, NEMS 2009
Pages	489-493
DOIs	https://doi.org/10.1109/NEMS.2009.5068625
Publication status	Published - 2009
Externally published	Yes
Event	4th IEEE International Conference on Nano/Micro Engineered and Molecular Systems, NEMS 2009 - Shenzhen, China Duration: 5 Jan 2009 → 8 Jan 2009

Conference

Conference	4th IEEE International Conference on Nano/Micro Engineered and Molecular Systems, NEMS 2009
Place	China
City	Shenzhen
Period	5/01/09 → 8/01/09

Research Keywords

μIMU
CNN
HMM
Human motion
MEMS
SVM

Access Output

10.1109/NEMS.2009.5068625

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

@inproceedings{fd54020ceb32495890883958c5b65193,

title = "Multi-category human motion recognition based on MEMS inertial sensing data",

abstract = "This paper presents multi-category human motion recognition methods based on MEMS inertial sensing data. A Micro Inertial Measurement Unit (pIMU) that is 56mm*23mm*15mm in size was built. This unit consists of three dimensional MEMS accelerometers, gyroscopes, a Bluetooth module and a MCU (Micro Controller Unit), which can record and transfer inertial data to a computer through serial port wirelessly. Five categories of human motion were recorded including walking, running, going upstairs, fall and standing. Fourier transform was used to extract the feature from the human motion data. The concentrated information was finally used to categorize the human motions through CNN (Cascade Neural Network) SVM (Support Vector Machine) and HMM (Hidden Markov Model) respectively. Experimental results showed that for the given 5 human motions, HMM have the best classification result with correct recognition rate range from 90%-100%. {\textcopyright} 2009 IEEE.",

keywords = "μIMU, CNN, HMM, Human motion, MEMS, SVM",

author = "Guangyi Shi and Yuexian Zoui and Yufeng Jin and Yali Zheng and Li, {Wen J.}",

year = "2009",

doi = "10.1109/NEMS.2009.5068625",

language = "English",

isbn = "9781424446308",

pages = "489--493",

booktitle = "4th IEEE International Conference on Nano/Micro Engineered and Molecular Systems, NEMS 2009",

note = "4th IEEE International Conference on Nano/Micro Engineered and Molecular Systems, NEMS 2009 ; Conference date: 05-01-2009 Through 08-01-2009",

}

Shi, G, Zoui, Y, Jin, Y, Zheng, Y & Li, WJ 2009, Multi-category human motion recognition based on MEMS inertial sensing data. in 4th IEEE International Conference on Nano/Micro Engineered and Molecular Systems, NEMS 2009., 5068625, pp. 489-493, 4th IEEE International Conference on Nano/Micro Engineered and Molecular Systems, NEMS 2009, Shenzhen, China, 5/01/09. https://doi.org/10.1109/NEMS.2009.5068625

Multi-category human motion recognition based on MEMS inertial sensing data. / Shi, Guangyi; Zoui, Yuexian; Jin, Yufeng et al.
4th IEEE International Conference on Nano/Micro Engineered and Molecular Systems, NEMS 2009. 2009. p. 489-493 5068625.

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

TY - GEN

T1 - Multi-category human motion recognition based on MEMS inertial sensing data

AU - Shi, Guangyi

AU - Zoui, Yuexian

AU - Jin, Yufeng

AU - Zheng, Yali

AU - Li, Wen J.

PY - 2009

Y1 - 2009

N2 - This paper presents multi-category human motion recognition methods based on MEMS inertial sensing data. A Micro Inertial Measurement Unit (pIMU) that is 56mm*23mm*15mm in size was built. This unit consists of three dimensional MEMS accelerometers, gyroscopes, a Bluetooth module and a MCU (Micro Controller Unit), which can record and transfer inertial data to a computer through serial port wirelessly. Five categories of human motion were recorded including walking, running, going upstairs, fall and standing. Fourier transform was used to extract the feature from the human motion data. The concentrated information was finally used to categorize the human motions through CNN (Cascade Neural Network) SVM (Support Vector Machine) and HMM (Hidden Markov Model) respectively. Experimental results showed that for the given 5 human motions, HMM have the best classification result with correct recognition rate range from 90%-100%. © 2009 IEEE.

AB - This paper presents multi-category human motion recognition methods based on MEMS inertial sensing data. A Micro Inertial Measurement Unit (pIMU) that is 56mm*23mm*15mm in size was built. This unit consists of three dimensional MEMS accelerometers, gyroscopes, a Bluetooth module and a MCU (Micro Controller Unit), which can record and transfer inertial data to a computer through serial port wirelessly. Five categories of human motion were recorded including walking, running, going upstairs, fall and standing. Fourier transform was used to extract the feature from the human motion data. The concentrated information was finally used to categorize the human motions through CNN (Cascade Neural Network) SVM (Support Vector Machine) and HMM (Hidden Markov Model) respectively. Experimental results showed that for the given 5 human motions, HMM have the best classification result with correct recognition rate range from 90%-100%. © 2009 IEEE.

KW - μIMU

KW - CNN

KW - HMM

KW - Human motion

KW - MEMS

KW - SVM

UR - http://www.scopus.com/inward/record.url?scp=70349662639&partnerID=8YFLogxK

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-70349662639&origin=recordpage

U2 - 10.1109/NEMS.2009.5068625

DO - 10.1109/NEMS.2009.5068625

M3 - RGC 32 - Refereed conference paper (with host publication)

SN - 9781424446308

SP - 489

EP - 493

BT - 4th IEEE International Conference on Nano/Micro Engineered and Molecular Systems, NEMS 2009

T2 - 4th IEEE International Conference on Nano/Micro Engineered and Molecular Systems, NEMS 2009

Y2 - 5 January 2009 through 8 January 2009

ER -

Multi-category human motion recognition based on MEMS inertial sensing data

Abstract

Conference

Research Keywords

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Cite this