Cargando…

A Lightweight Human Fall Detection Network

The rising issue of an aging population has intensified the focus on the health concerns of the elderly. Among these concerns, falls have emerged as a predominant health threat for this demographic. The YOLOv5 family represents the forefront of techniques for human fall detection. However, this algo...

Descripción completa

Detalles Bibliográficos
Autores principales: Kan, Xi, Zhu, Shenghao, Zhang, Yonghong, Qian, Chengshan
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10674212/
https://www.ncbi.nlm.nih.gov/pubmed/38005456
http://dx.doi.org/10.3390/s23229069
_version_ 1785140775204421632
author Kan, Xi
Zhu, Shenghao
Zhang, Yonghong
Qian, Chengshan
author_facet Kan, Xi
Zhu, Shenghao
Zhang, Yonghong
Qian, Chengshan
author_sort Kan, Xi
collection PubMed
description The rising issue of an aging population has intensified the focus on the health concerns of the elderly. Among these concerns, falls have emerged as a predominant health threat for this demographic. The YOLOv5 family represents the forefront of techniques for human fall detection. However, this algorithm, although advanced, grapples with issues such as computational demands, challenges in hardware integration, and vulnerability to occlusions in the designated target group. To address these limitations, we introduce a pioneering lightweight approach named CGNS-YOLO for human fall detection. Our method incorporates both the GSConv module and the GDCN module to reconfigure the neck network of YOLOv5s. The objective behind this modification is to diminish the model size, curtail floating-point computations during feature channel fusion, and bolster feature extraction efficacy, thereby enhancing hardware adaptability. We also integrate a normalization-based attention module (NAM) into the framework, which concentrates on salient fall-related data and deemphasizes less pertinent information. This strategic refinement augments the algorithm’s precision. By embedding the SCYLLA Intersection over Union (SIoU) loss function, our model benefits from faster convergence and heightened detection precision. We evaluated our model using the Multicam dataset and the Le2i Fall Detection dataset. Our findings indicate a 1.2% enhancement in detection accuracy compared with the conventional YOLOv5s framework. Notably, our model realized a 20.3% decrease in parameter tally and a 29.6% drop in floating-point operations. A comprehensive instance analysis and comparative assessments underscore the method’s superiority and efficacy.
format Online
Article
Text
id pubmed-10674212
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-106742122023-11-09 A Lightweight Human Fall Detection Network Kan, Xi Zhu, Shenghao Zhang, Yonghong Qian, Chengshan Sensors (Basel) Article The rising issue of an aging population has intensified the focus on the health concerns of the elderly. Among these concerns, falls have emerged as a predominant health threat for this demographic. The YOLOv5 family represents the forefront of techniques for human fall detection. However, this algorithm, although advanced, grapples with issues such as computational demands, challenges in hardware integration, and vulnerability to occlusions in the designated target group. To address these limitations, we introduce a pioneering lightweight approach named CGNS-YOLO for human fall detection. Our method incorporates both the GSConv module and the GDCN module to reconfigure the neck network of YOLOv5s. The objective behind this modification is to diminish the model size, curtail floating-point computations during feature channel fusion, and bolster feature extraction efficacy, thereby enhancing hardware adaptability. We also integrate a normalization-based attention module (NAM) into the framework, which concentrates on salient fall-related data and deemphasizes less pertinent information. This strategic refinement augments the algorithm’s precision. By embedding the SCYLLA Intersection over Union (SIoU) loss function, our model benefits from faster convergence and heightened detection precision. We evaluated our model using the Multicam dataset and the Le2i Fall Detection dataset. Our findings indicate a 1.2% enhancement in detection accuracy compared with the conventional YOLOv5s framework. Notably, our model realized a 20.3% decrease in parameter tally and a 29.6% drop in floating-point operations. A comprehensive instance analysis and comparative assessments underscore the method’s superiority and efficacy. MDPI 2023-11-09 /pmc/articles/PMC10674212/ /pubmed/38005456 http://dx.doi.org/10.3390/s23229069 Text en © 2023 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Kan, Xi
Zhu, Shenghao
Zhang, Yonghong
Qian, Chengshan
A Lightweight Human Fall Detection Network
title A Lightweight Human Fall Detection Network
title_full A Lightweight Human Fall Detection Network
title_fullStr A Lightweight Human Fall Detection Network
title_full_unstemmed A Lightweight Human Fall Detection Network
title_short A Lightweight Human Fall Detection Network
title_sort lightweight human fall detection network
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10674212/
https://www.ncbi.nlm.nih.gov/pubmed/38005456
http://dx.doi.org/10.3390/s23229069
work_keys_str_mv AT kanxi alightweighthumanfalldetectionnetwork
AT zhushenghao alightweighthumanfalldetectionnetwork
AT zhangyonghong alightweighthumanfalldetectionnetwork
AT qianchengshan alightweighthumanfalldetectionnetwork
AT kanxi lightweighthumanfalldetectionnetwork
AT zhushenghao lightweighthumanfalldetectionnetwork
AT zhangyonghong lightweighthumanfalldetectionnetwork
AT qianchengshan lightweighthumanfalldetectionnetwork