Fall detection using human skeleton features

Heilym Ramirez; Sergio A. Velastin; Ernesto Fabregas; Ignacio Meza; Dimitrios Makris; Gonzalo Farias

doi:10.1049/icp.2021.1465

Fall detection using human skeleton features

Heilym Ramirez, Sergio A. Velastin, Ernesto Fabregas, Ignacio Meza, Dimitrios Makris, Gonzalo Farias

Research output: Contribution to conference › Paper › peer-review

2 Scopus citations

Abstract

Falls are one of the leading causes of death and serious injury in people, especially for the elderly. In addition, falls accidents have a direct financial cost for health systems and, indirectly, for the productivity of society. Among the most important problems in fall detection systems is privacy, limitations of operating devices, and the comparison of machine learning techniques for detection. This article presents a fall detection system by means of a k-Nearest Neighbor (KNN) classifier based on camera-vision using pose detection of the human skeleton for the features extraction. The proposed method is evaluated with UP-FALL dataset, surpassing the results of other fall detection systems that use the same database. This method achieves a 98.84% accuracy and an F₁-Score of 97.41%.

Original language	English
Pages	73-78
Number of pages	6
DOIs	https://doi.org/10.1049/icp.2021.1465
State	Published - 2021
Externally published	Yes
Event	11th International Conference of Pattern Recognition Systems, ICPRS 2021 - Virtual, Online Duration: 17 Mar 2021 → 19 Mar 2021

Conference

Conference	11th International Conference of Pattern Recognition Systems, ICPRS 2021
City	Virtual, Online
Period	17/03/21 → 19/03/21

Keywords

Computer vision
Fall detection
Human skeleton
Pose estimation

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1049/icp.2021.1465

Cite this

@conference{3e8e9ed6f6de4c0d95047ed2e0a42a47,

title = "Fall detection using human skeleton features",

abstract = "Falls are one of the leading causes of death and serious injury in people, especially for the elderly. In addition, falls accidents have a direct financial cost for health systems and, indirectly, for the productivity of society. Among the most important problems in fall detection systems is privacy, limitations of operating devices, and the comparison of machine learning techniques for detection. This article presents a fall detection system by means of a k-Nearest Neighbor (KNN) classifier based on camera-vision using pose detection of the human skeleton for the features extraction. The proposed method is evaluated with UP-FALL dataset, surpassing the results of other fall detection systems that use the same database. This method achieves a 98.84% accuracy and an F1-Score of 97.41%.",

keywords = "Computer vision, Fall detection, Human skeleton, Pose estimation",

author = "Heilym Ramirez and Velastin, {Sergio A.} and Ernesto Fabregas and Ignacio Meza and Dimitrios Makris and Gonzalo Farias",

note = "Publisher Copyright: {\textcopyright} 2021 Institution of Engineering and Technology. All rights reserved.; 11th International Conference of Pattern Recognition Systems, ICPRS 2021 ; Conference date: 17-03-2021 Through 19-03-2021",

year = "2021",

doi = "10.1049/icp.2021.1465",

language = "English",

pages = "73--78",

}

TY - CONF

T1 - Fall detection using human skeleton features

AU - Ramirez, Heilym

AU - Velastin, Sergio A.

AU - Fabregas, Ernesto

AU - Meza, Ignacio

AU - Makris, Dimitrios

AU - Farias, Gonzalo

PY - 2021

Y1 - 2021

N2 - Falls are one of the leading causes of death and serious injury in people, especially for the elderly. In addition, falls accidents have a direct financial cost for health systems and, indirectly, for the productivity of society. Among the most important problems in fall detection systems is privacy, limitations of operating devices, and the comparison of machine learning techniques for detection. This article presents a fall detection system by means of a k-Nearest Neighbor (KNN) classifier based on camera-vision using pose detection of the human skeleton for the features extraction. The proposed method is evaluated with UP-FALL dataset, surpassing the results of other fall detection systems that use the same database. This method achieves a 98.84% accuracy and an F1-Score of 97.41%.

AB - Falls are one of the leading causes of death and serious injury in people, especially for the elderly. In addition, falls accidents have a direct financial cost for health systems and, indirectly, for the productivity of society. Among the most important problems in fall detection systems is privacy, limitations of operating devices, and the comparison of machine learning techniques for detection. This article presents a fall detection system by means of a k-Nearest Neighbor (KNN) classifier based on camera-vision using pose detection of the human skeleton for the features extraction. The proposed method is evaluated with UP-FALL dataset, surpassing the results of other fall detection systems that use the same database. This method achieves a 98.84% accuracy and an F1-Score of 97.41%.

KW - Computer vision

KW - Fall detection

KW - Human skeleton

KW - Pose estimation

UR - http://www.scopus.com/inward/record.url?scp=85118122955&partnerID=8YFLogxK

U2 - 10.1049/icp.2021.1465

DO - 10.1049/icp.2021.1465

M3 - Paper

AN - SCOPUS:85118122955

SP - 73

EP - 78

T2 - 11th International Conference of Pattern Recognition Systems, ICPRS 2021

Y2 - 17 March 2021 through 19 March 2021

ER -

Fall detection using human skeleton features

Abstract

Conference

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this