Employment of an electronic tongue combined with deep learning and transfer learning for discriminating the storage time of Pu-erh tea

Abstract Pu-erh tea is a famous Chinese fermented tea, and its quality and flavor are closely related to the storage time used for its fermentation. This paper puts forward one method to discriminate the age of Pu-erh tea by employing a voltammetric electronic tongue (VE-Tongue) combined with deep learning and transfer learning techniques. To make the deep learning model suitable for processing VE-Tongue signals, a one-dimensional convolutional neural network (1-D CNN) was developed to automatically perform feature extraction and classification. Transfer learning (TL) was introduced to train the model to reduce the training complexity and enhance the generalization capability of the CNN. The performance of the proposed model was further compared with that of traditional machine learning methods such as the backpropagation neural network, support vector machine and extreme learning machine. The results showed that the proposed model exhibited better performance in classifying Pu-erh tea than other methods. Its accuracy for the test set, precision, recall and F1 score was 98.80%, 98.2%, 98%, and 0.98, respectively. This study found that the VE-Tongue combined with deep learning and TL algorithms could be a sensitive, reliable and effective detection method for identifying the amount of storage time of Pu-erh tea, which could further expand its applications to other related fields.

[1]  Benachir Bouchikhi,et al.  Classification of Honey According to Geographical and Botanical Origins and Detection of Its Adulteration Using Voltammetric Electronic Tongue , 2016, Food Analytical Methods.

[2]  Z. Jane Wang,et al.  DT-LET: Deep Transfer Learning by Exploring where to Transfer , 2018, Neurocomputing.

[3]  Buket D. Barkana,et al.  Utilizing CNNs and transfer learning of pre-trained models for age range classification from unconstrained face images , 2019, Image Vis. Comput..

[4]  Esmail Sohouli,et al.  Fabrication of an electrochemical mesalazine sensor based on ZIF-67 , 2020 .

[5]  Kuang-Hua Chang,et al.  A hand-held electronic tongue based on fluorometry for taste assessment of tea. , 2010, Biosensors & bioelectronics.

[6]  Qingrui Shi,et al.  Classification of Pericarpium Citri Reticulatae of Different Ages by Using a Voltammetric Electronic Tongue System , 2018, International Journal of Electrochemical Science.

[7]  Jiewen Zhao,et al.  Identification of the green tea grade level using electronic tongue and pattern recognition , 2008 .

[8]  Dazhong Wu,et al.  Deep learning for smart manufacturing: Methods and applications , 2018, Journal of Manufacturing Systems.

[9]  Maria Luz Rodriguez-Mendez,et al.  Beer discrimination using a portable electronic tongue based on screen-printed electrodes , 2015 .

[10]  Zhenbo Wei,et al.  Application of the voltammetric electronic tongue based on nanocomposite modified electrodes for identifying rice wines of different geographical origins. , 2019, Analytica chimica acta.

[11]  Vincenzo Chiofalo,et al.  Fusion of electronic nose, electronic tongue and computer vision for animal source food authentication and quality assessment – A review , 2017 .

[12]  Pete Warden,et al.  Speech Commands: A Dataset for Limited-Vocabulary Speech Recognition , 2018, ArXiv.

[13]  Lin Lin,et al.  A novel gas turbine fault diagnosis method based on transfer learning with CNN , 2019, Measurement.

[14]  Zhongwen Xie,et al.  Metabolomics based on UHPLC-Orbitrap-MS and GNPS molecular networking reveals effects of time scale and environment of storage on the metabolites and taste quality of raw Pu-erh tea. , 2019, Journal of agricultural and food chemistry.

[15]  Jie Zhao,et al.  Development of a LeNet-5 Gas Identification CNN Structure for Electronic Noses , 2019, Sensors.

[16]  C. Lee,et al.  Application of metabolomics in the analysis of manufacturing type of pu-erh tea and composition changes with different postfermentation year. , 2010, Journal of agricultural and food chemistry.

[17]  M Suchetha,et al.  PSO optimized 1-D CNN-SVM architecture for real-time detection and classification applications , 2019, Comput. Biol. Medicine.

[18]  A. V.DavidSánchez,et al.  Advanced support vector machines and kernel methods , 2003, Neurocomputing.

[19]  Lin Gao,et al.  Quality identification and evaluation of Pu-erh teas of different grade levels and various ages through sensory evaluation and instrumental analysis , 2016 .

[20]  Yuerong Liang,et al.  A study on chemical estimation of pu-erh tea quality , 2005 .

[21]  Yoshua Bengio,et al.  On the Expressive Power of Deep Architectures , 2011, ALT.

[22]  Bo Tan,et al.  Electrocardiogram soft computing using hybrid deep learning CNN-ELM , 2020, Appl. Soft Comput..

[23]  Li Tong,et al.  A visual encoding model based on deep neural networks and transfer learning for brain activity measured by functional magnetic resonance imaging , 2019, Journal of Neuroscience Methods.

[24]  G. Xie,et al.  Characterization of pu-erh tea using chemical and metabolic profiling approaches. , 2009, Journal of agricultural and food chemistry.

[25]  C Di Natale,et al.  Direct and two-stage data analysis procedures based on PCA, PLS-DA and ANN for ISE-based electronic tongue-Effect of supervised feature extraction. , 2005, Talanta.

[26]  Guang-Bin Huang,et al.  What are Extreme Learning Machines? Filling the Gap Between Frank Rosenblatt’s Dream and John von Neumann’s Puzzle , 2015, Cognitive Computation.

[27]  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.

[28]  L. Hwang,et al.  Polyphenol contents of Pu-Erh teas and their abilities to inhibit cholesterol biosynthesis in Hep G2 cell line , 2008 .

[29]  Q. Guo,et al.  The galloyl moiety of green tea catechins is the critical structural feature to inhibit fatty-acid synthase. , 2003, Biochemical pharmacology.

[30]  U. Srinivasulu Reddy,et al.  Crop pest classification based on deep convolutional neural network and transfer learning , 2019, Comput. Electron. Agric..

[31]  Yuerong Liang,et al.  Processing and chemical constituents of Pu-erh tea: A review , 2013 .

[32]  X. Cetó,et al.  Voltammetric electronic tongue for vinegar fingerprinting. , 2020, Talanta.

[33]  Geoffrey E. Hinton,et al.  Learning internal representations by error propagation , 1986 .

[34]  Hua-lin Xie,et al.  An Improved Weighted Partial Least Squares Method Coupled with Near Infrared Spectroscopy for Rapid Determination of Multiple Components and Anti-Oxidant Activity of Pu-Erh Tea , 2018, Molecules.

[35]  Jun Wang,et al.  Classification and prediction of rice wines with different marked ages by using a voltammetric electronic tongue. , 2011, Biosensors & bioelectronics.

[36]  Ana Paula Rocha,et al.  Deep learning in exchange markets , 2019, Information Economics and Policy.

[37]  R. Vig,et al.  Performance Evaluation of a Novel iTongue for Indian Black Tea Discrimination , 2010, IEEE Sensors Journal.

[38]  B.H.M. Sadeghi,et al.  A BP-neural network predictor model for plastic injection molding process , 2000 .

[39]  Bipan Tudu,et al.  Instrumental testing of tea by combining the responses of electronic nose and tongue , 2012 .

[40]  A. H. Kiranmayee,et al.  New data reduction algorithm for voltammetric signals of electronic tongue for discrimination of liquids , 2012 .

[41]  Lin Lu,et al.  Visualized attribute analysis approach for characterization and quantification of rice taste flavor using electronic tongue. , 2016, Analytica chimica acta.

[42]  Feng Ya-dong.  Identification of Pu'er Ripe Teas with Different Origins and Fermentation Years by Surface-Enhanced Raman Spectroscopy , 2013 .

[43]  Mauro Bertotti,et al.  Use of copper and gold electrodes as sensitive elements for fabrication of an electronic tongue: Discrimination of wines and whiskies , 2011 .

[44]  Bipan Tudu,et al.  Classification of black tea liquor using cyclic voltammetry , 2012 .

[45]  Lixiang Duan,et al.  Intelligent Fault Diagnosis of Rolling Element Bearings Based on HHT and CNN , 2018, 2018 Prognostics and System Health Management Conference (PHM-Chongqing).