DocumentCode :
2851107
Title :
Study on fall detection based on intelligent video analysis
Author :
Ngo, Y.T. ; Nguyen, Hien ; Pham, Thuy V.
Author_Institution :
Electron. & Telecomm. Dept., Duc Minh Coll. of Econ. & Technol., Danang, Vietnam
fYear :
2012
fDate :
10-12 Oct. 2012
Firstpage :
114
Lastpage :
117
Abstract :
In this paper, a fall detection algorithm has been built using intelligent analysis of captured video signal. Five geometrical features are extracted from input video signal and are recognized by a trained feed-forward neural network. Experimental results on our self-built database show that the proposed fall detection system can detect fall events with quite high precision under different falling conditions.
Keywords :
feature extraction; feedforward neural nets; image segmentation; learning (artificial intelligence); video signal processing; visual databases; feature recognition; feedforward neural network training; geometrical feature extraction; input video signal; intelligent video analysis-based fall detection algorithm; self-built database; Databases; Training; Vectors; fall detection; feature extraction; neural network; recognition; segmentation;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Advanced Technologies for Communications (ATC), 2012 International Conference on
Conference_Location :
Hanoi
ISSN :
2162-1020
Print_ISBN :
978-1-4673-4351-0
Type :
conf
DOI :
10.1109/ATC.2012.6404242
Filename :
6404242
Link To Document :
بازگشت