Activities Recognition and Fall Detection in Continuous Data Streams Using Radar Sensor

Haobo Li; Aman Shrestha; Hadi Heidari; Julien Le Kernec; Francesco Fioranelli

doi:10.1109/imbioc.2019.8777855

Activities Recognition and Fall Detection in Continuous Data Streams Using Radar Sensor

Haobo Li
, Aman Shrestha
, Hadi Heidari
, Julien Le Kernec
, Francesco Fioranelli (Lead / Corresponding author)

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

33 Citations (Scopus)

Abstract

This student paper presents a Quadratic-kernel Support Vector Machine (SVM) based FMCW (Frequency Modulated Continuous Wave) radar system to recognize daily activities and detect fall accidents. Data collected in this work is divided into two different collection modes, namely, snapshots mode (different activities individually collected in isolation) and continuous activity mode (continuous streams of activities collected one after the other). For the continuous activity streams, a sliding window approach with 4s duration and 70% overlapping has achieved 84.7% classification accuracy and subsequent improvement of 2.6% has been proved by using Sequential Forward Selection (SFS) on six participants to identify an optimal feature set. A `tracking' graph has been utilized to verify that the radar system can correctly identify falls as critical events among the other activities.

Original language	English
Title of host publication	IEEE MTT-S 2019 International Microwave Biomedical Conference (IMBioC 2019)
Subtitle of host publication	Proceedings
Publisher	IEEE
Number of pages	4
ISBN (Print)	978-153867395-9
DOIs	https://doi.org/10.1109/imbioc.2019.8777855
Publication status	Published - May 2019
Event	The IEEE MTT-S 2019 International Microwave Biomedical Conference - International conference hotel of Nanjing, Nanjing, China Duration: 6 May 2019 → 8 May 2019

Conference

Conference	The IEEE MTT-S 2019 International Microwave Biomedical Conference
Abbreviated title	IMBioC 2019
Country/Territory	China
City	Nanjing
Period	6/05/19 → 8/05/19

Keywords

Continuous activity streams
Feature selection
Human activity recognition
radar micro-Doppler

Access to Document

10.1109/imbioc.2019.8777855

Cite this

@inproceedings{b7a900161e344e66b157faaefde5f642,

title = "Activities Recognition and Fall Detection in Continuous Data Streams Using Radar Sensor",

abstract = "This student paper presents a Quadratic-kernel Support Vector Machine (SVM) based FMCW (Frequency Modulated Continuous Wave) radar system to recognize daily activities and detect fall accidents. Data collected in this work is divided into two different collection modes, namely, snapshots mode (different activities individually collected in isolation) and continuous activity mode (continuous streams of activities collected one after the other). For the continuous activity streams, a sliding window approach with 4s duration and 70\% overlapping has achieved 84.7\% classification accuracy and subsequent improvement of 2.6\% has been proved by using Sequential Forward Selection (SFS) on six participants to identify an optimal feature set. A `tracking' graph has been utilized to verify that the radar system can correctly identify falls as critical events among the other activities.",

keywords = "Continuous activity streams, Feature selection, Human activity recognition, radar micro-Doppler",

author = "Haobo Li and Aman Shrestha and Hadi Heidari and \{Le Kernec\}, Julien and Francesco Fioranelli",

year = "2019",

month = may,

doi = "10.1109/imbioc.2019.8777855",

language = "English",

isbn = "978-153867395-9",

booktitle = "IEEE MTT-S 2019 International Microwave Biomedical Conference (IMBioC 2019)",

publisher = "IEEE",

note = "The IEEE MTT-S 2019 International Microwave Biomedical Conference, IMBioC 2019 ; Conference date: 06-05-2019 Through 08-05-2019",

}

Li, H, Shrestha, A, Heidari, H, Le Kernec, J & Fioranelli, F 2019, Activities Recognition and Fall Detection in Continuous Data Streams Using Radar Sensor. in IEEE MTT-S 2019 International Microwave Biomedical Conference (IMBioC 2019): Proceedings., 8777855, IEEE, The IEEE MTT-S 2019 International Microwave Biomedical Conference, Nanjing, China, 6/05/19. https://doi.org/10.1109/imbioc.2019.8777855

TY - GEN

T1 - Activities Recognition and Fall Detection in Continuous Data Streams Using Radar Sensor

AU - Li, Haobo

AU - Shrestha, Aman

AU - Heidari, Hadi

AU - Le Kernec, Julien

AU - Fioranelli, Francesco

PY - 2019/5

Y1 - 2019/5

N2 - This student paper presents a Quadratic-kernel Support Vector Machine (SVM) based FMCW (Frequency Modulated Continuous Wave) radar system to recognize daily activities and detect fall accidents. Data collected in this work is divided into two different collection modes, namely, snapshots mode (different activities individually collected in isolation) and continuous activity mode (continuous streams of activities collected one after the other). For the continuous activity streams, a sliding window approach with 4s duration and 70% overlapping has achieved 84.7% classification accuracy and subsequent improvement of 2.6% has been proved by using Sequential Forward Selection (SFS) on six participants to identify an optimal feature set. A `tracking' graph has been utilized to verify that the radar system can correctly identify falls as critical events among the other activities.

AB - This student paper presents a Quadratic-kernel Support Vector Machine (SVM) based FMCW (Frequency Modulated Continuous Wave) radar system to recognize daily activities and detect fall accidents. Data collected in this work is divided into two different collection modes, namely, snapshots mode (different activities individually collected in isolation) and continuous activity mode (continuous streams of activities collected one after the other). For the continuous activity streams, a sliding window approach with 4s duration and 70% overlapping has achieved 84.7% classification accuracy and subsequent improvement of 2.6% has been proved by using Sequential Forward Selection (SFS) on six participants to identify an optimal feature set. A `tracking' graph has been utilized to verify that the radar system can correctly identify falls as critical events among the other activities.

KW - Continuous activity streams

KW - Feature selection

KW - Human activity recognition

KW - radar micro-Doppler

UR - https://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=pure_staging_and_production&SrcAuth=WosAPI&KeyUT=WOS:000676326800075&DestLinkType=FullRecord&DestApp=WOS_CPL

U2 - 10.1109/imbioc.2019.8777855

DO - 10.1109/imbioc.2019.8777855

M3 - Conference contribution

SN - 978-153867395-9

BT - IEEE MTT-S 2019 International Microwave Biomedical Conference (IMBioC 2019)

PB - IEEE

T2 - The IEEE MTT-S 2019 International Microwave Biomedical Conference

Y2 - 6 May 2019 through 8 May 2019

ER -

Activities Recognition and Fall Detection in Continuous Data Streams Using Radar Sensor

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this