Gas identification based on committee machine for microelectronic gas sensor

Minghua Shi, Amine Bermak, Sofiane Brahim Belhouari, Philip C.H. Chan

Research output: Contribution to journalArticle

20 Citations (Scopus)

Abstract

Gas identification represents a big challenge for pattern recognition systems due to several particular problems such as nonselectivity and drift. The purpose of this paper is twofold: 1) to compare the accuracy of a range of advanced and classical pattern recognition algorithms for gas identification for the in-house sensor array signals and 2) to propose a gas identification ensemble machine (GIEM), which combines various gas identification algorithms, to obtain a unified decision with improved accuracy. An integrated sensor array has been designed with the aim of identifying combustion gases. The classification accuracy of different density models is compared with several neural network architectures. On the gas sensors data used in this paper, Gaussian mixture models achieved the best performance with higher than 94% accuracy. A committee machine is implemented by assembling the outputs of these gas identification algorithms through advanced voting machines using a weighting and classification confidence function. Experiments on real sensors' data proved the effectiveness of the system with an improved accuracy over the individual classifiers. An average performance of 97% was achieved using the proposed committee machine.

Original languageEnglish
Pages (from-to)1786-1793
Number of pages8
JournalIEEE Transactions on Instrumentation and Measurement
Volume55
Issue number5
DOIs
Publication statusPublished - Oct 2006
Externally publishedYes

Fingerprint

Chemical sensors
microelectronics
Microelectronics
sensors
Gases
gases
Sensor arrays
Voting machines
pattern recognition
Pattern recognition systems
voting
Network architecture
Pattern recognition
classifiers
assembling
Classifiers
confidence
Neural networks
Sensors
output

Keywords

  • Classification
  • Density models
  • Ensemble methods
  • Gas sensor array
  • Neural networks
  • Pattern recognition

ASJC Scopus subject areas

  • Instrumentation
  • Electrical and Electronic Engineering

Cite this

Gas identification based on committee machine for microelectronic gas sensor. / Shi, Minghua; Bermak, Amine; Belhouari, Sofiane Brahim; Chan, Philip C.H.

In: IEEE Transactions on Instrumentation and Measurement, Vol. 55, No. 5, 10.2006, p. 1786-1793.

Research output: Contribution to journalArticle

Shi, Minghua ; Bermak, Amine ; Belhouari, Sofiane Brahim ; Chan, Philip C.H. / Gas identification based on committee machine for microelectronic gas sensor. In: IEEE Transactions on Instrumentation and Measurement. 2006 ; Vol. 55, No. 5. pp. 1786-1793.
@article{9b05b0efa71e421487cd1c70994a528c,
title = "Gas identification based on committee machine for microelectronic gas sensor",
abstract = "Gas identification represents a big challenge for pattern recognition systems due to several particular problems such as nonselectivity and drift. The purpose of this paper is twofold: 1) to compare the accuracy of a range of advanced and classical pattern recognition algorithms for gas identification for the in-house sensor array signals and 2) to propose a gas identification ensemble machine (GIEM), which combines various gas identification algorithms, to obtain a unified decision with improved accuracy. An integrated sensor array has been designed with the aim of identifying combustion gases. The classification accuracy of different density models is compared with several neural network architectures. On the gas sensors data used in this paper, Gaussian mixture models achieved the best performance with higher than 94{\%} accuracy. A committee machine is implemented by assembling the outputs of these gas identification algorithms through advanced voting machines using a weighting and classification confidence function. Experiments on real sensors' data proved the effectiveness of the system with an improved accuracy over the individual classifiers. An average performance of 97{\%} was achieved using the proposed committee machine.",
keywords = "Classification, Density models, Ensemble methods, Gas sensor array, Neural networks, Pattern recognition",
author = "Minghua Shi and Amine Bermak and Belhouari, {Sofiane Brahim} and Chan, {Philip C.H.}",
year = "2006",
month = "10",
doi = "10.1109/TIM.2006.880956",
language = "English",
volume = "55",
pages = "1786--1793",
journal = "IEEE Transactions on Instrumentation and Measurement",
issn = "0018-9456",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "5",

}

TY - JOUR

T1 - Gas identification based on committee machine for microelectronic gas sensor

AU - Shi, Minghua

AU - Bermak, Amine

AU - Belhouari, Sofiane Brahim

AU - Chan, Philip C.H.

PY - 2006/10

Y1 - 2006/10

N2 - Gas identification represents a big challenge for pattern recognition systems due to several particular problems such as nonselectivity and drift. The purpose of this paper is twofold: 1) to compare the accuracy of a range of advanced and classical pattern recognition algorithms for gas identification for the in-house sensor array signals and 2) to propose a gas identification ensemble machine (GIEM), which combines various gas identification algorithms, to obtain a unified decision with improved accuracy. An integrated sensor array has been designed with the aim of identifying combustion gases. The classification accuracy of different density models is compared with several neural network architectures. On the gas sensors data used in this paper, Gaussian mixture models achieved the best performance with higher than 94% accuracy. A committee machine is implemented by assembling the outputs of these gas identification algorithms through advanced voting machines using a weighting and classification confidence function. Experiments on real sensors' data proved the effectiveness of the system with an improved accuracy over the individual classifiers. An average performance of 97% was achieved using the proposed committee machine.

AB - Gas identification represents a big challenge for pattern recognition systems due to several particular problems such as nonselectivity and drift. The purpose of this paper is twofold: 1) to compare the accuracy of a range of advanced and classical pattern recognition algorithms for gas identification for the in-house sensor array signals and 2) to propose a gas identification ensemble machine (GIEM), which combines various gas identification algorithms, to obtain a unified decision with improved accuracy. An integrated sensor array has been designed with the aim of identifying combustion gases. The classification accuracy of different density models is compared with several neural network architectures. On the gas sensors data used in this paper, Gaussian mixture models achieved the best performance with higher than 94% accuracy. A committee machine is implemented by assembling the outputs of these gas identification algorithms through advanced voting machines using a weighting and classification confidence function. Experiments on real sensors' data proved the effectiveness of the system with an improved accuracy over the individual classifiers. An average performance of 97% was achieved using the proposed committee machine.

KW - Classification

KW - Density models

KW - Ensemble methods

KW - Gas sensor array

KW - Neural networks

KW - Pattern recognition

UR - http://www.scopus.com/inward/record.url?scp=33749345821&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33749345821&partnerID=8YFLogxK

U2 - 10.1109/TIM.2006.880956

DO - 10.1109/TIM.2006.880956

M3 - Article

AN - SCOPUS:33749345821

VL - 55

SP - 1786

EP - 1793

JO - IEEE Transactions on Instrumentation and Measurement

JF - IEEE Transactions on Instrumentation and Measurement

SN - 0018-9456

IS - 5

ER -