RESEARCH AND DEVELOPMENT FALL DETECTION AND NOTIFICATION WEB APPLICATION FOR THE ELDERLY USING THE MEDIAPIPE FRAMEWORK Articles
Article Sidebar
Main Article Content
Abstract
The purposes of the research were to develop fall detection and notification web application for the elderly using MediaPipe framework, and to evaluate the effectiveness of the application. The scope of development is divided into two parts: Part 1: Camera management system, imports video data using a Real Time Streaming Protocol connection. Part 2: Detection and notification application through the LINE application using MediaPipe Framework's Pose Landmarker Model to detect and track skeletons 33 point. The experimental is six fall patterns: 1) straight face falling forward, 2) straight face falling backward,
3) turning sideways falling to the right, 4) facing forward falling to the right, 5) facing forward. fall to the left, and 6) turn to fall to the left were collected from 10 volunteers, the photos captured and identified 33 important key landmark points were conducted, and approximately 1,000 data records. The results of evaluating the performance of fall detection using Random Forest and K-NN model and testing data sets found that the Random Forest model had the best fall detection results, with an accuracy of 88.00%, and the system is more accurate at detecting falls in a straight-faced position than falls to the left and right side.
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
This article is published and copyrighted by the Science and Technology Journal. South East Bangkok College
References
P. Kulurkar , et al., “AI based elderly fall prediction system using wearable sensors: A smart home-care technology with IOT,” Measurement: Sensors, 100614, pp. 1-11, 2023. https://doi.org/10.1016/
j.measen.2022.100614.
P. Leelatien , et al., “Severity-based fall detection for elderly using WiFi sensing,” The Journal of KMUTNB, vol.34, no.1, pp. 1-12, 2022.
Regional Health Promotion Center 5, “The Effectiveness of Preparation and Fall Prevention Programs in Elderly,” [Online]. Available: https://apps.hpc.go.th/dmkm/web/uploads/2020/036008-2020031
/b4d55b3ad74a37143daf28575f182905.pdf. [Accessed: November 3, 2023].
K. Nuntaboot, “Accidental situation and self-care for prevention of accident among older People,” Journal of Nursing and Health Care, vol.37, no.3, pp. 164-172, 2019.
Regional Health Promotion Center 5, “The Effectiveness of Preparation and Fall Prevention Programs in Elderly,” [Online]. Available: https://apps.hpc.go.th/dmkm/web/uploads/2020/036008-2020031
/b4d55b3ad74a37143daf28575f182905.pdf. [Accessed: November 3, 2023].
Pressac Communications, “Wired or wireless sensors? The advantages and disadvanta-ges of wired and wireless systems,” [Online]. Available: https://www.pressac.com/insights/wired-or-wireless-sensors/#. [Accessed: November 7, 2023].
P. Leelatien , et al., “Severity-based fall detection for elderly using WiFi sensing,” The Journal of KMUTNB, vol.34, no.1, pp. 1-12, 2022.
X. Yang, et al., “WiFi-based multistage fall detection with channel state information,” International Journal of Disributed Sensor Networks, vol.14, no.10, pp 1-10, 2018.
M. Wiboonyasake, “Machine Learning,” [Online]. Available: https://www.a ware.co.th/machine-learning-คืออะไร. [Accessed: September 20, 2023].
Google Inc, “MediaPipe Framework,” [Online]. Available: https://developers.google.com/mediapipe/framework. [Accessed: September 20, 2023].
Google Inc, “Pose Landmark Detection,” [Online]. Available: https://developers.google.com/mediapipe/solutions/vision/pose_landmarker. [Accessed: September 20, 2023].
S. Jeong, et al., "Human-skeleton based Fall-Detection Method using LSTM for Manufacturing Industries," 2019 34th International Technical Conference on Circuits/Systems, Computers and Communications (ITC-CSCC), JeJu, Korea (South), 2019, pp. 1-4, doi: 10.1109/ITC-CSCC.2019.8793342.
C. D. Jawale, et al., “Elderly Care With Fall Detection,” International Conference on Applications of Intelligent Computing in Engineering and Science (AICES-2022), Raipur, India, February 12-13, 2022.
J. Liu and others, “Human skeleton behavior recognition model based on multi-object pose estimation with spatiotemporal semantics,” Machine Vision and Applications, vol.34, 2023. https://doi.org/10.1007/s00138-023-01396-0.
P. Wongkietkajoh, “Pose Estimation for Indoor Fall Detection Using Mediapipe,” Master's thesis, Srinakharinwirot University. 2023. [Online]. Available: http://ir-ithesis.swu.ac.th/dspace/handle/
/2576. [Accessed: January 13, 2024].
Google Inc, “The pose landmarker model tracks 33 body landmark locations,” [Online]. Available: https://developers.google.com/mediapipe/solutions/vision/pose_landmarker. [Accessed: December 20, 2023].
F. Shu, and J. Shu, “Human falling pattern,” Scientific Reports, vol.11, 2021. https://doi.org/10.1038/s41598-021-81115-9.
P. Kulkasem, et al., “Fall Detection System for Monitoring an Elderly Person in Elderly Care Center,” [Online]. Available: https://buuir.buu.ac.th/xmlui/bitstream/handle/1234567890/1949/2560_103.pdf. [Accessed: January 13, 2024].