Neural network based electronic nose for classification of tea aroma

This paper describes an investigation into the performance of a Neural Network (NN) based Electronic Nose (EN) system, which can discriminate the aroma of different tea grades. The EN system comprising of an array of four tin-oxide gas sensors was used to sniff thirteen randomly selected tea grades, which were exemplars of eight categories in terms of aroma profiles. The mean and peak of the transient signals generated by the gas sensors, as a result of aroma sniffing, were treated as the feature vectors for the analysis. Principal Component Analysis (PCA) was used to visualise the different categories of aroma profiles. In addition, K-means and Kohonen’s Self Organising Map (SOM) cluster analysis indicated there were eight clusters in the dataset. Data classification was performed using supervised NN classifiers; namely the Multi-Layer Perceptron (MLP) network, Radial Basis Function (RBF) network, and Constructive Probabilistic Neural Network (CPNN) were used for aroma classification. The results were that the three NNs performed as follows: 90.77, 92.31, and 93.85%, respectively in terms of classification accuracy. Hence the performance of the proposed method of aroma analysis demonstrates that it is possible to use NN based EN to assist with the tea quality monitoring procedure during the tea grading process. In addition the results indicate the possibility for standardization of the tea aroma in numeric terms.

[1]  Teuvo Kohonen,et al.  The self-organizing map , 1990 .

[2]  Anthony J. A. Wright,et al.  Lipid degradation during manufacture of black tea , 1979 .

[3]  J. Gardner,et al.  Electronic noses and their application in the food industry , 1997 .

[4]  Pradip K. Mahanta,et al.  Changes of fatty acid contents, lipoxygenase activities, and volatiles during black tea manufacture , 1993 .

[5]  N. Togari,et al.  Pattern recognition applied to gas chromatographic profiles of volatile components in three tea categories , 1995 .

[6]  G. J. Persley,et al.  Coffee and cocoa. , 1990 .

[7]  T. Yamanishi,et al.  Tea, coffee, cocoa, and other beverages [Flavor analysis] , 1981 .

[8]  Tadakazu Takeo,et al.  Comparison of black tea aromas of orthodox and CTC tea and of black teas made from different varieties , 1983 .

[9]  Ramaswamy Ravichandran.  The impact of pruning and time from pruning on quality and aroma constituents of black tea , 2004 .

[10]  Michael R. Berthold,et al.  Constructive training of probabilistic neural networks , 1998, Neurocomputing.

[11]  Tohru Shimohara,et al.  Quality evaluation on green tea , 1993 .

[12]  Philip O. Owuor.  Changes in quality parameters of commercial black seedling tea due to the time of the year in the Eastern Highlands of Kenya , 1992 .

[13]  Geoffrey E. Hinton,et al.  Learning representations by back-propagating errors , 1986, Nature.

[14]  Julian W. Gardner,et al.  Electronic noses: a review of signal processing techniques , 1999 .

[15]  R. Ravichandran,et al.  The impact of processing techniques on tea volatiles , 1998 .

[16]  John Moody,et al.  Fast Learning in Networks of Locally-Tuned Processing Units , 1989, Neural Computation.

[17]  Julian W. Gardner,et al.  A brief history of electronic noses , 1994 .

[18]  James C. Bezdek,et al.  Pattern Recognition with Fuzzy Objective Function Algorithms , 1981, Advanced Applications in Pattern Recognition.