Summary
International Technical Conference on Circuits/Systems, Computers and Communications
2016
Session Number:M2-4
Session:
Number:M2-4-6
Motion Estimation-Based Human Falling Detection for Visual Surveillance
Heegwang Kim, Jinho Park, Hasil Park, Joonki Paik ,
pp.197-198
Publication Date:2016/7/10
Online ISSN:2188-5079
DOI:10.34385/proc.61.M2-4-6
PDF download (1.4MB)
Summary:
Detection of a human falling event has attracted increasing attention in a visual surveillance system. This paper presents a novel falling event detection algorithm using motion estimation and an integrated spatiotemporal energy map of the object region. The proposed method first extracts a human region using a background subtraction method. Next, we applied an optical flow algorithm to estimate motion vectors, and the energy map is generated by accumulating the detected human region for a certain period of time. We can then detect the falling event using the k-nearest neighbor (kNN) classification with the previously estimated motion information and energy map.