High-Performance Lightweight Fall Detection with an Improved YOLOv5s Algorithm
The aging population has drastically increased in the past two decades, stimulating the development of devices for healthcare and medical purposes. As one of the leading potential risks, the injuries caused by accidental falls at home are hazardous to the health (and even lifespan) of elderly people...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-08-01
|
| Series: | Machines |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-1702/11/8/818 |
| _version_ | 1797584006611468288 |
|---|---|
| author | Yuanpeng Wang Zhaozhan Chi Meng Liu Guangxian Li Songlin Ding |
| author_facet | Yuanpeng Wang Zhaozhan Chi Meng Liu Guangxian Li Songlin Ding |
| author_sort | Yuanpeng Wang |
| collection | DOAJ |
| description | The aging population has drastically increased in the past two decades, stimulating the development of devices for healthcare and medical purposes. As one of the leading potential risks, the injuries caused by accidental falls at home are hazardous to the health (and even lifespan) of elderly people. In this paper, an improved YOLOv5s algorithm is proposed, aiming to improve the efficiency and accuracy of lightweight fall detection via the following modifications that elevate its accuracy and speed: first, a k-means++ clustering algorithm was applied to increase the accuracy of the anchor boxes; the backbone network was replaced with a lightweight ShuffleNetV2 network to embed simplified devices with limited computing ability; an SE attention mechanism module was added to the last layer of the backbone to improve the feature extraction capability; the GIOU loss function was replaced by a SIOU loss function to increase the accuracy of detection and the training speed. The results of testing show that the mAP of the improved algorithm was improved by 3.5%, the model size was reduced by 75%, and the time consumed for computation was reduced by 79.4% compared with the conventional YOLOv5s. The algorithm proposed in this paper has higher detection accuracy and detection speed. It is suitable for deployment in embedded devices with limited performance and with lower cost. |
| first_indexed | 2024-03-10T23:47:07Z |
| format | Article |
| id | doaj.art-043e4c2db8db45689991917393f3978d |
| institution | Directory Open Access Journal |
| issn | 2075-1702 |
| language | English |
| last_indexed | 2024-03-10T23:47:07Z |
| publishDate | 2023-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Machines |
| spelling | doaj.art-043e4c2db8db45689991917393f3978d2023-11-19T01:57:07ZengMDPI AGMachines2075-17022023-08-0111881810.3390/machines11080818High-Performance Lightweight Fall Detection with an Improved YOLOv5s AlgorithmYuanpeng Wang0Zhaozhan Chi1Meng Liu2Guangxian Li3Songlin Ding4School of Computer, Electronics and Information, Guangxi University, Nanning 530004, ChinaSchool of Mechanical Engineering, Guangxi University, Nanning 530004, ChinaSchool of Mechanical Engineering, Guangxi University, Nanning 530004, ChinaSchool of Mechanical Engineering, Guangxi University, Nanning 530004, ChinaSchool of Engineering, RMIT University, Melbourne 3000, AustraliaThe aging population has drastically increased in the past two decades, stimulating the development of devices for healthcare and medical purposes. As one of the leading potential risks, the injuries caused by accidental falls at home are hazardous to the health (and even lifespan) of elderly people. In this paper, an improved YOLOv5s algorithm is proposed, aiming to improve the efficiency and accuracy of lightweight fall detection via the following modifications that elevate its accuracy and speed: first, a k-means++ clustering algorithm was applied to increase the accuracy of the anchor boxes; the backbone network was replaced with a lightweight ShuffleNetV2 network to embed simplified devices with limited computing ability; an SE attention mechanism module was added to the last layer of the backbone to improve the feature extraction capability; the GIOU loss function was replaced by a SIOU loss function to increase the accuracy of detection and the training speed. The results of testing show that the mAP of the improved algorithm was improved by 3.5%, the model size was reduced by 75%, and the time consumed for computation was reduced by 79.4% compared with the conventional YOLOv5s. The algorithm proposed in this paper has higher detection accuracy and detection speed. It is suitable for deployment in embedded devices with limited performance and with lower cost.https://www.mdpi.com/2075-1702/11/8/818fall detectionk-means++ShufflenetV2SE attention mechanismSIOU loss function |
| spellingShingle | Yuanpeng Wang Zhaozhan Chi Meng Liu Guangxian Li Songlin Ding High-Performance Lightweight Fall Detection with an Improved YOLOv5s Algorithm Machines fall detection k-means++ ShufflenetV2 SE attention mechanism SIOU loss function |
| title | High-Performance Lightweight Fall Detection with an Improved YOLOv5s Algorithm |
| title_full | High-Performance Lightweight Fall Detection with an Improved YOLOv5s Algorithm |
| title_fullStr | High-Performance Lightweight Fall Detection with an Improved YOLOv5s Algorithm |
| title_full_unstemmed | High-Performance Lightweight Fall Detection with an Improved YOLOv5s Algorithm |
| title_short | High-Performance Lightweight Fall Detection with an Improved YOLOv5s Algorithm |
| title_sort | high performance lightweight fall detection with an improved yolov5s algorithm |
| topic | fall detection k-means++ ShufflenetV2 SE attention mechanism SIOU loss function |
| url | https://www.mdpi.com/2075-1702/11/8/818 |
| work_keys_str_mv | AT yuanpengwang highperformancelightweightfalldetectionwithanimprovedyolov5salgorithm AT zhaozhanchi highperformancelightweightfalldetectionwithanimprovedyolov5salgorithm AT mengliu highperformancelightweightfalldetectionwithanimprovedyolov5salgorithm AT guangxianli highperformancelightweightfalldetectionwithanimprovedyolov5salgorithm AT songlinding highperformancelightweightfalldetectionwithanimprovedyolov5salgorithm |