skip to main content
10.1145/3674225.3674307acmotherconferencesArticle/Chapter ViewAbstractPublication PagespeaiConference Proceedingsconference-collections
research-article

Optimizing the YOLO Network for Human Fall Detection

Published: 31 July 2024 Publication History

Abstract

Falls occur frequently in daily life, and with the numerous injuries and safety hazards caused by falls, fall detection has become a study of great importance. If a timely response can be made when a fall event occurs, the injuries caused by falls can be reduced. Given the high interference of background information of the surrounding environment in the monitoring video information of public places and the different scales of human abnormal behavior targets, it is difficult to further improve the accuracy of human fall behavior detection at present. To address the above problems, a fall behavior detection method by improving the YOLOv5 network is designed. The method adds CBAM attention model to the original YOLOv5 backbone network, which makes the network more focused on the channels and spatial locations in the input data that are important for the task through both channel attention and spatial attention, thus improving the performance and generalization ability of the network. Embedding the SE attention module in the network can better adjust the degree of attention of the network to each channel, thus improving the performance of the detection. It reduces the computational complexity of the whole model and improves the accuracy of the model for target localization of abnormal human behavior. Changing the activation function from RELU to Swish helps to alleviate the problem of gradient vanishing, especially in the deep network, which helps to propagate the gradient better and improve the training stability.

References

[1]
Zhenyu Q, **nwen F, Kaixuan C. Research on Pedestrian Fall Detection Based on YOLO [J]. 2022.
[2]
Gomes M E N, Macedo D, Zanchettin C, Multi-human fall detection and localization in videos [J]. Computer Vision and Image Understanding, 2022, 220: 103442.
[3]
Sun G, Wang Z. Fall detection algorithm for the elderly based on human posture estimation [C]//2020 Asia-Pacific Conference on Image Processing, Electronics and Computers (IPEC). IEEE, 2020: 172-176.
[4]
Yan H, Hu B, Chen G, Real-time continuous human rehabilitation action recognition using OpenPose and FCN[C]//2020 3rd International Conference on Advanced Electronic Materials, Computers and Software Engineering (AEMCSE). IEEE, 2020: 239-242.
[5]
Shahzad A, Kim K. FallDroid: An automated smart-phone-based fall detection system using multiple kernel learning [J]. IEEE Transactions on Industrial Informatics, 2018, 15(1): 35-44.
[6]
Uddin M Z, Torresen J. A deep learning-based human activity recognition in darkness [C]//2018 Colour and Visual Computing Symposium (CVCS). IEEE, 2018: 1-5.
[7]
Qiao S, Wang Y, Li J. Real-time human gesture grading based on OpenPose [C]//2017 10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI). IEEE, 2017: 1-6.
[8]
Viswakumar A, Rajagopalan V, Ray T, Human gait analysis using OpenPose [C]//2019 fifth international conference on image information processing (ICIIP). IEEE, 2019: 310-314.
[9]
Shetty A D, Shubha B, Suryanarayana K. Detection and tracking of a human using the infrared thermopile array sensor—“Grid-EYE”[C]//2017 International Conference on Intelligent Computing, Instrumentation and Control Technologies (ICICICT). IEEE, 2017: 1490-1495.
[10]
Ramachandran P, Zoph B, Le Q V. Searching for activation functions [J]. ar**v preprint ar**v:1710.05941, 2017.
[11]
Adhikari K, Bouchachia H, Nait-Charif H. Activity recognition for indoor fall detection using convolutional neural network[C]//2017 Fifteenth IAPR International Conference on Machine Vision Applications (MVA). IEEE, 2017: 81-84.
[12]
Kwolek B, Kepski M. Human fall detection on embedded platform using depth maps and wireless accelerometer[J]. Computer methods and programs in biomedicine, 2014, 117(3): 489-501.
[13]
Lahiri D, Dhiman C, Vishwakarma D K. Abnormal human action recognition using average energy images[C]//2017 Conference on information and communication technology (CICT). IEEE, 2017: 1-5.
[14]
Kwolek B, Kepski M. Improving fall detection by the use of depth sensor and accelerometer[J]. Neurocomputing, 2015, 168: 637-645.

Recommendations

Comments

Information & Contributors

Information

Published In

cover image ACM Other conferences
PEAI '24: Proceedings of the 2024 International Conference on Power Electronics and Artificial Intelligence
January 2024
969 pages
ISBN:9798400716638
DOI:10.1145/3674225
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than the author(s) must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected].

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 31 July 2024

Permissions

Request permissions for this article.

Check for updates

Qualifiers

  • Research-article
  • Research
  • Refereed limited

Conference

PEAI 2024

Contributors

Other Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

  • 0
    Total Citations
  • 58
    Total Downloads
  • Downloads (Last 12 months)58
  • Downloads (Last 6 weeks)13
Reflects downloads up to 17 Feb 2025

Other Metrics

Citations

View Options

Login options

View options

PDF

View or Download as a PDF file.

PDF

eReader

View online with eReader.

eReader

HTML Format

View this article in HTML Format.

HTML Format

Figures

Tables

Media

Share

Share

Share this Publication link

Share on social media