Detection of Human Activities and Human Fall Recognition Using Deep Learning Techniques

Smartphones are quickly becoming the most important communication device in people's lives in today's world. Human Activity Recognition has grown in popularity as a field of study in a variety of areas, including medical care, tracking, and education. The sensors in smartphones allow us to...

Full description

Saved in:
Bibliographic Details
Published in:International research journal of innovations in engineering and technology 2018-03, Vol.2 (1), p.48
Main Author: Mosiganti, Joseph Prakash
Format: Article
Language:English
Subjects:
Activities of daily living
Deep learning
Health services
Human activity recognition
Machine learning
Moving object recognition
Smartphones
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Smartphones are quickly becoming the most important communication device in people's lives in today's world. Human Activity Recognition has grown in popularity as a field of study in a variety of areas, including medical care, tracking, and education. The sensors in smartphones allow us to use them for a wide range of applications. Healthcare is one of the major domains where human activity recognition is widely used. In this paper, a human activity recognition system has been developed that can detect six activities of daily living (ADL) along with human fall. Human fall occurs due to an accident that can cause serious injuries which may lead to significant medical problems when the issue is not addressed properly. The proposed system uses a variant of deep learning technique to detect human activities and human fall. The accuracy is significantly increased by nearly 4% when compared with previous results.
ISSN:2581-3048