Fall detection based on depth image sequence

  • Liangcan Liao
  • , Guitao Cao*
  • , Wenming Cao
  • *Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Fall is one of the most significant causes of injury in the elderly. In this paper, we propose two types of feature extraction methods to detect when the fall happens based on the Kinect depth image sequences. One is assuming there exists different position lines in the XYZ 3-dimensional space, it will be active when the moving object touches it. The other is mapping the image sequences to single image by Speed-Time Depth Mapping (STDM), and obtaining the 36-dimensiona features in this image. The results of experimental demonstrate the effectiveness of our method.

Original languageEnglish
Title of host publicationWMSCI 2017 - 21st World Multi-Conference on Systemics, Cybernetics and Informatics, Proceedings
EditorsElina Gaile-Sarkane, Nagib C. Callaos, Belkis Sanchez, Shigehiro Hashimoto, Natalja Lace
PublisherInternational Institute of Informatics and Systemics, IIIS
Pages319-324
Number of pages6
ISBN (Electronic)9781941763643
StatePublished - 2017
Event21st World Multi-Conference on Systemics, Cybernetics and Informatics, WMSCI 2017 - Orlando, United States
Duration: 8 Jul 201711 Jul 2017

Publication series

NameWMSCI 2017 - 21st World Multi-Conference on Systemics, Cybernetics and Informatics, Proceedings
Volume2

Conference

Conference21st World Multi-Conference on Systemics, Cybernetics and Informatics, WMSCI 2017
Country/TerritoryUnited States
CityOrlando
Period8/07/1711/07/17

Keywords

  • Depth image
  • Fall detection
  • Feature extraction
  • Kinect
  • SVM

Fingerprint

Dive into the research topics of 'Fall detection based on depth image sequence'. Together they form a unique fingerprint.

Cite this