Vision-based In-room Fall Detection Application

Citation

Cho, Yi Heng and Mat Desa, Shahbe (2022) Vision-based In-room Fall Detection Application. Journal of Logistics, Informatics and Service Science, 9 (4). pp. 51-71. ISSN 2409-2665

[img] Text
Vol.9.No.4.05.pdf - Published Version
Restricted to Repository staff only

Download (825kB)

Abstract

In recent years, fall detection is a hot study topic as video surveillance becomes more universal in public and private spaces. The consequence of falls not only can result in physical damage but as well as psychological issues, especially among the elderly. With the help of the great processing power of a computer, deep learning has come into help to compete with traditional fall detection that required hand-crafted features. This paper presents a fall detection application using a convolutional neural network (CNN) for web-based and mobile-based implementations. A deep network can learn to detect a fall automatically by showing it with an ample amount of examples. A deep learning network is built to detect falls in an image with a CNN structure. A technical representation of the proposed design and methodology is organized and presented in this paper. As a result, a fall detection web application was built with the integration of a fall detection model with 98.37% specificity and 96.47% sensitivity.

Item Type: Article
Uncontrolled Keywords: Fall detection, machine learning, CNN, mobile application, web application
Subjects: Q Science > Q Science (General) > Q300-390 Cybernetics
Divisions: Faculty of Information Science and Technology (FIST)
Depositing User: Ms Nurul Iqtiani Ahmad
Date Deposited: 22 Mar 2023 02:48
Last Modified: 22 Mar 2023 02:48
URII: http://shdl.mmu.edu.my/id/eprint/11262

Downloads

Downloads per month over past year

View ItemEdit (login required)