Home // International Journal On Advances in Intelligent Systems, volume 2, number 1, 2009 // View article

Intelligent Electronic Nose Systems with Metal Oxide Gas Sensors for Fire Detection

Authors:
Michifumi Yoshioka
Toru Fujinaka
Sigeru Omatu

Keywords: electronic nose, neural networks, learning vector quantization, metal oxide gas sensor, smell classification

Abstract:
In this paper, a reliable electronic nose system designed from the combination of various semi-conductor metal oxide gas sensors (MOGS) is applied to the detection of fire resulting from various sources in a kitchen. The time series signals obtained from the same source of fire are highly correlated, and different sources of fire exhibit unique patterns in the time series data. Therefore, the error back-propagation (BP) method can be effectively used for the classification of the tested smell. The accuracy of 99.6% is achieved by using only a single training data set from each source of fire. The accuracy achieved with the k-means algorithm is 98.3%, which also shows the high ability of the EN in detecting the early stage of fire from various sources.

Pages: 268 to 277

Copyright: Copyright (c) to authors, 2009. Used with permission.

Publication date: June 7, 2009

Published in: journal

ISSN: 1942-2679