The novel EOS835 electronic nose and data analysis for evaluating coffee ripening

Abstract We present an analysis of roasted coffee ripening performed by the novel Electronic Olfactory System EOS 835 , manufactured by the Italian company Sacmi Imola s.c.a.r.l., which is based on thin film semiconductor metal oxide gas sensors. We focused our analysis on: (1) exploratory data analysis for systematically investigating the outcomes of different sampling conditions and therefore selecting advantageous settings; (2) feature selection for improving classification performance and ranking the contribution of the different sensors and feature types. Exploratory analysis, via the successive generation of PCA plots, showed that the main factors influencing discrimination between diverse ripening times are headspace generation time (HGT, i.e. time elapsed between vial filling and measurement performing) and sample preparation. A relatively long HGT (18 h) allows to follow the ripening progression of the coffee blend over time and to correctly classify the best coffee ripening (as determined by an expert taster). In forming the feature vector, we added a feature calculated in the phase space to the standard features. Feature selection showed that, the phase space feature consistently lead to improved classification and that, of the three sensor types constituting the array, the two indium–tin oxide sensors perform better for our application.

[1]  H. T. Nagle,et al.  Handbook of Machine Olfaction , 2002 .

[2]  Tetsuo Aishima,et al.  AROMA DISCRIMINATION BY PATTERN RECOGNITION ANALYSIS OF RESPONSES FROM SEMICONDUCTOR GAS SENSOR ARRAY , 1991 .

[3]  J. Bosset,et al.  The electronic nose applied to dairy products: a review , 2003 .

[4]  Anil K. Jain,et al.  Feature Selection: Evaluation, Application, and Small Sample Performance , 1997, IEEE Trans. Pattern Anal. Mach. Intell..

[5]  Pierre Grenier,et al.  Optimisation of electronic nose measurements. Part II : Influence of experimental parameters , 1999 .

[6]  Doron Lancet,et al.  A feature extraction method for chemical sensors in electronic noses , 2003 .

[7]  T. Eklöv,et al.  Selection of variables for interpreting multivariate gas sensor data , 1999 .

[8]  H. T. Nagle,et al.  Transient response analysis of an electronic nose using multi-exponential models , 1999 .

[9]  Udo Weimar,et al.  Odours and flavours identified with hybrid modular sensor systems1 , 1997 .

[10]  Giorgio Sberveglieri,et al.  Mo influence on SnO2 thin films properties , 2002 .

[11]  Giorgio Sberveglieri,et al.  Recent developments in semiconducting thin-film gas sensors , 1995 .

[12]  G. Sberveglieri,et al.  Structural characterization of tin and molybdenum oxides thin films deposited by RGTO , 2001 .

[13]  Giorgio Sberveglieri,et al.  Investigation on the O3 sensitivity properties of WO3 thin films prepared by sol–gel, thermal evaporation and r.f. sputtering techniques , 2000 .

[14]  Sameer Singh,et al.  Fuzzy neural computing of coffee and tainted-water data from an electronic nose , 1996 .

[15]  Y. Jung,et al.  Influence of dc magnetron sputtering parameters on surface morphology of indium tin oxide thin films , 2004 .

[16]  Vijay P. Singh,et al.  Characteristics of indium tin oxide films deposited by r.f. magnetron sputtering , 1995 .

[17]  Véronique Bellon-Maurel,et al.  Optimisation of electronic nose measurements. Part I: Methodology of output feature selection , 1998 .

[18]  Matteo Pardo,et al.  Coffee analysis with an electronic nose , 2002, IEEE Trans. Instrum. Meas..

[19]  Ingemar Lundström,et al.  Data preprocessing enhances the classification of different brands of Espresso coffee with an electronic nose , 2000 .

[20]  J. Gardner,et al.  Application of an electronic nose to the discrimination of coffees , 1992 .

[21]  C. Distante,et al.  On the study of feature extraction methods for an electronic nose , 2002 .

[22]  E. Martinelli,et al.  Feature Extraction of chemical sensors in phase space , 2003 .