Open AccessHuman Fall Detection Using Video Surveillance
Author(s) -
G Kavya,
Sunil Kumar C T,
C. Dhanush,
J Kruthika
Publication year - 2021
Publication title -
acs journal for science and engineering
Language(s) - English
Resource type - Journals
ISSN - 2582-9610
DOI - 10.34293/acsjse.v1i1.1
Subject(s) - computer science , background subtraction , artificial intelligence , convolutional neural network , alarm , computer vision , video monitoring , deep learning , object detection , pattern recognition (psychology) , real time computing , engineering , pixel , aerospace engineering
Fall is one of the biggest challenge in elderly people, pregnant and small children’s, who stays alone in home. Sometimes this fall leads to severe injuries and even to death. Detecting the fall is very much important for elderly people. Convolutional Neural Network (CNN) is an deep learning algorithm used for image processing. In this paper, we present a video-based fall detection using CNN, this CNN will perform background subtraction and captures only foreground objects to detect the human movements and detect if fall happens. Firstly, camera will be capturing all the movements of the person. Our proposed model will detect the fall and finally an alarm is raised and email is sent to a given particular caretaker and family member. Our experimental results show the best performance of the proposed model.