Activity recognition for indoor fall detection using convolutional neural network

Authors: Adhikari, K., Bouchachia, H. and Nait-Charif, H.

http://eprints.bournemouth.ac.uk/29421/

Start date: 8 May 2017

Falls are a major health problem in the elderly population.

Therefore, a dedicated monitoring system is highly desirable to improve independent living. This paper presents a video based fall detection system in an indoor environment using convolution neural network.

Identifying human poses is important in detecting fall events as specific ”change of pose” defines a fall. Knowledge of series of poses is a key detecting fall or non-fall events. A lying pose which may be considered as an after-fall pose is different from other normal activities such as standing, sitting, bending or crawling.This paper uses Convolutional Neural Networks (CNN) to recognise different poses. Using Kinect, the following image combinations are explored: RGB, Depth, RGB-D and background subtracted RGBD.

We have constructed our own dataset by recording different activities performed by different people in different indoor set-ups. Our results suggest that combining background subtracted RGB and Depth with CNN gives the best possible solution for monitoring indoor video based falls.

This source preferred by Hamid Bouchachia and Hammadi Nait-Charif

This data was imported from Scopus:

Authors: Adhikari, K., Bouchachia, H. and Nait-Charif, H.

http://eprints.bournemouth.ac.uk/29421/

Journal: Proceedings of the 15th IAPR International Conference on Machine Vision Applications, MVA 2017

Pages: 81-84

ISBN: 9784901122160

DOI: 10.23919/MVA.2017.7986795

© 2017 MVA Organization All Rights Reserved. Falls are a major health problem in the elderly population. Therefore, a dedicated monitoring system is highly desirable to improve independent living. This paper presents a video based fall detection system in an indoor environment using convolution neural network. Identifying human poses is important in detecting fall events as specific 'change of pose' defines a fall. Knowledge of series of poses is a key to detecting fall or non-fall events. A lying pose which may be considered as an after-fall pose is different from other normal activities such as lying/sleeping on the sofa or crawling. This paper uses Convolutional Neural Networks (CNN) to recognize different poses. Using Kinect, the following image combinations are explored: RGB, Depth, RGB-D and background subtracted RGB-D. We have constructed our own dataset by recording different activities performed by different people in different indoor set-ups. Our results suggest that combining RGB background subtracted and Depth with CNN gives the best possible solution for monitoring indoor video based falls.

The data on this page was last updated at 04:40 on November 22, 2017.