High-resolution time-frequency distributions for fall detection

Moeness G. Amin; Yimin D. Zhang; Boualem Boashash

doi:10.1117/12.2180139

21 May 2015 High-resolution time-frequency distributions for fall detection

Moeness G. Amin, Yimin D. Zhang, Boualem Boashash

Proceedings Volume 9461, Radar Sensor Technology XIX; and Active and Passive Signatures VI; 94611B (2015) https://doi.org/10.1117/12.2180139
Event: SPIE Defense + Security, 2015, Baltimore, MD, United States

Abstract

In this paper, we examine the role of high-resolution time-frequency distributions (TFDs) of radar micro-Doppler signatures for fall detection. The work supports the recent and rising interest in using emerging radar technology for elderly care and assisted living. Spectrograms have been the de facto joint-variable signal representation, depicting the signal power in both time and frequency. Although there have been major advances in designing quadratic TFDs which are superior to spectrograms in terms of detailing the local signal behavior, the contributions of these distributions in the area of human motion classifications and their offerings in enhanced feature extractions have not yet been properly evaluated. The main purpose of this paper is to show the effect of using high-resolution TFD kernels, in lieu of spectrogram, on fall detection. We focus on the extended modified B-distribution (EMBD) and exploit the level of details it provides as compared with the coarse and smoothed time-frequency signatures offered by spectrograms.

Citation Download Citation

Moeness G. Amin, Yimin D. Zhang, and Boualem Boashash "High-resolution time-frequency distributions for fall detection", Proc. SPIE 9461, Radar Sensor Technology XIX; and Active and Passive Signatures VI, 94611B (21 May 2015); https://doi.org/10.1117/12.2180139

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available