PRE-PROCESSING AND PATTERN RECOGNITION METHODS FOR ARTIFICIAL OLFACTION SYSTEMS: A REVIEW
Journal Title: Metrology and Measurement Systems - Year 2005, Vol 12, Issue 1
Abstract
Data analysis is a fundamental part of artificial olfaction systems. Nonetheless, while electrical and mechanical components of electronic noses are deeply investigated and optimised for the specific applications, data analysis is sometimes performed without any critical understanding about hypotheses on which it is funded. Several methodologies were imported into this field and utilised to correlate electronic nose data to the information required for classification purposes. Chemometrics and neural networks are the sources of the most popular methods currently in use. In this paper some of the most used methodologies of data analysis are reviewed and their use for electronic nose data is discussed. Although the discussion is given considering chemical sensor arrays, the conclusions are generic and valid for a broader range of sensor systems.
Authors and Affiliations
CORRADO DI NATALE, EUGENIO MARTINELLI, ARNALDO D'AMICO
Keywords
Related Articles
- EP ID EP96834
- DOI -
- Views 64
- Downloads 0
PHOTONICS SOCIETY OF POLAND ESTABLISHED
There is a clear need in this country for strong consolidation and rebuilding of the self-organization of the technical professional communities. They have to accommodate themselves to the dynamic changes and evolutions...
ADVANCES AND GENERIC PROBLEMS IN INSTRUMENT DESIGN METHODOLOGY
The paper is based on a long-term programme of research on mathematical modelling and design of instruments and instrument elements in the Measurement and Instrumentation Centre of City University, London. It considers t...
USING AN OPTICAL FIBRE SENSOR TO MEASURE DISPLACEMENT OF ELECTRODYNAMICALLY FORMED SHEET METAL
The paper discusses the use of reflective fibre-optic sensor for the purposes of measuring motion parameters of sheet metal formed at high speed, subject to percussive Lorentz forces extorted in a pulsed magnetic field....
INTRODUCTION TO THE T AND CHI-SQUARE DISTRIBUTION FOR A MORE ACCURATE EVALUATION OF THE MEASURE OF THE WORD ERROR RATE IN ANALOGTO-DIGITAL CONVERTERS<br />
The word error rate (WER) in an Analog to Digital Converter (ADC) is the probability of receiving an erroneous code for an input, after correction is made for gain, offset, and nonlinearity errors, and a specified allowa...
DETERMINING THE UNCERTAINTY OF FITTING DISCRETE MEASUREMENT DATA TO A NOMINAL SURFACE
Coordinate measurements are a source of digital data in the form of coordinates of measurement points with a discrete distribution on the measured surface. Geometric deviations of freeform surfaces are determined at each...