Heuristic Application System on Pose Detection of Elderly Activity Using Machine Learning in Real-Time

Ariyani, Sofia and Yuniarno, Eko Mulyanto and Purnomo, Mauridhi Hery (2022) Heuristic Application System on Pose Detection of Elderly Activity Using Machine Learning in Real-Time. In: IEEE International Conference on Computational Intelligence and Virtual Environments for Measurement Systems and Applications (CIVEMSA 2022), June 15-17, 2022, Chemnitz, Germany.

Text IEEE_CIVEMSA_2022_Program+Final book.pdf Download (3MB)

Abstract

Many developed countries have a population of elderly people that is greater than the population of young workers. So. To meet the demand for labor there is a decline because many are retiring. To meet the future demand that many elderly people need a quality care service so that the elderly who experience physical and cognitive decline can be well protected.Great potential study and evaluation of elderly movement activity for healthcare. The algorithm of pose estimation takes advantage of recording video have tracked elderly movement automatically using camera devices and computer vision. Monitoring and measuring elderly movement activity in real-time more easily accessible with this view of technology offers a clear and exciting potential as motor assessment by the doctor at the patient at home. The perpetrator can send video recording directly in the field by combining expertise and perspective as from physical therapy insight into the application of pose estimation in human health, especially the elderly. This is focusing in a safe and comfortable way. These models use CNN and LSTM for classified labeling landmark point detection results with high performance 97,3 accuracy average and in real-time have range 30 FPS. So the heuristic application system can be recommended for monitoring the use of the camera with all its limitations

Item Type:	Conference or Workshop Item (Paper)
Subjects:	600 Technology and Applied Science > 620 Engineering
Divisions:	Faculty of Engineering > Department of Electrical Engineering (S1)
Depositing User:	Berlian Eka Kurnia
Date Deposited:	14 Sep 2024 01:22
Last Modified:	14 Sep 2024 01:22
URI:	http://repository.unmuhjember.ac.id/id/eprint/23401

Actions (login required)

View Item