Journal of Social Safety Science
Online ISSN : 2187-9842
Print ISSN : 1345-2088
ISSN-L : 1345-2088
Real Time Sound Information Analysis for Machine Recognition of Seismic Indoor Damage Situation
Tadayoshi NAKASHIMAShigeyuki OKADAMasayuki KITAHARA
Author information
JOURNAL FREE ACCESS

2017 Volume 31 Pages 279-286

Details
Abstract

It is one of the ironclad rules of risk management that we take multiple countermeasures for reducing human casualties in dwellings attacked by hazardous earthquakes. As a series of study on seismic indoor safety protection, in this paper we tried to propose a new monitoring system capable of judging some risky events from indoor sound information at earthquake. A recognition function builted-in the system identifies the occurrence and location of events that furniture overturns and pieces of broken dishes scatters as well as the position of inhabitants through a simple sound analysis of frequency characteristics and maximum amplitudes of sound. As a result, we succeeded in recognition with high accuracy.

Content from these authors
© 2017 Institute of Social Safety Science
Previous article Next article
feedback
Top