A Novel Feature Specificity Enhancement for Taste Recognition by Electronic Tongue

An electronic tongue (E-Tongue) is a bionic system that relies on an array of electrode sensors to realize taste perception. Large pulse voltammetry (LAPV) is an important E-Tongue type which generally generates a large amount of response data. Considering that high common-mode characteristics existing in sensor arrays largely depress the recognition performance, we propose an alternative feature extraction method for sensor specificity enhancement, which is called feature specificity enhancement (FSE). Specifically, the proposed FSE method measures sensor specificity on paired sensor responses and utilizes kernel function for nonlinear projection. Meanwhile, kernel extreme learning machine (KELM) is utilized to evaluate the overall performance of recognition. In experimental evaluation, we introduce several feature extraction methods and classifiers for comparison. The results indicate that the proposed feature extraction combined with KELM shows the highest recognition accuracy of 95% on our E-Tongue dataset, which is superior to other methods in both effectiveness and efficiency.

[1]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[2]  Ning Li,et al.  An Optimized Symmetric WENO Method-Based Numerical Simulation of Intense Sound Field Generated by Underwater Plasma Sound Source , 2018, J. Sensors.

[3]  Constantin Apetrei,et al.  Application of voltammetric e-tongue for the detection of ammonia and putrescine in beef products , 2016 .

[4]  Shaoping Deng,et al.  Multifrequency large amplitude pulse voltammetry: A novel electrochemical method for electronic tongue , 2007 .

[5]  Kiyoshi Toko,et al.  Electronic Tongues–A Review , 2013, IEEE Sensors Journal.

[6]  A. Dybko,et al.  Potentiometric electronic tongue based on integrated array of microelectrodes , 2007 .

[7]  Roberto Muñoz,et al.  Voltammetric Electronic Tongue and Support Vector Machines for Identification of Selected Features in Mexican Coffee , 2014, Sensors.

[8]  C. Natale,et al.  Evaluation of Italian wine by the electronic tongue: recognition, quantitative analysis and correlation with human sensory perception , 2003 .

[9]  I. Lundström,et al.  An electronic tongue based on voltammetry , 1997 .

[10]  Stephanus Büttgenbach,et al.  Application of an E-Tongue to the Analysis of Monovarietal and Blends of White Wines , 2011, Sensors.

[11]  Roberto Paolesse,et al.  Metalloporphyrin - Based Electronic Tongue: an Application for the Analysis of Italian White Wines , 2007, Sensors.

[12]  Hongming Zhou,et al.  Extreme Learning Machine for Regression and Multiclass Classification , 2012, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[13]  A. Legin,et al.  Mimicking Daphnia magna bioassay performance by an electronic tongue for urban water quality control. , 2014, Analytica chimica acta.

[14]  A. C. Veloso,et al.  An electronic tongue taste evaluation: identification of goat milk adulteration with bovine milk , 2009 .

[15]  Li Wang,et al.  Discrimination of Rice with Different Pretreatment Methods by Using a Voltammetric Electronic Tongue , 2015, Sensors.

[16]  M Palit,et al.  Classification of Black Tea Taste and Correlation With Tea Taster's Mark Using Voltammetric Electronic Tongue , 2010, IEEE Transactions on Instrumentation and Measurement.

[17]  F. Winquist,et al.  Discrimination of tea by means of a voltammetric electronic tongue and different applied waveforms , 2001 .

[18]  Tao Liu,et al.  An Active Feature Selection Strategy for DWT in Artificial Taste , 2018, J. Sensors.

[19]  Zbigniew Brzozka,et al.  Classification of beverages using a reduced sensor array , 2004 .

[20]  Chee Kheong Siew,et al.  Extreme learning machine: Theory and applications , 2006, Neurocomputing.

[21]  P. K. Dutta,et al.  Comparison of multivariate preprocessing techniques as applied to electronic tongue based pattern classification for black tea. , 2010, Analytica chimica acta.

[22]  Zhenbo Wei,et al.  Tracing floral and geographical origins of honeys by potentiometric and voltammetric electronic tongue , 2014 .

[23]  A. K. Bag,et al.  Monitoring the Fermentation Process and Detection of Optimum Fermentation Time of Black Tea Using an Electronic Tongue , 2015, IEEE Sensors Journal.