Validation of the rapid detection approach for enhancing the electronic nose systems performance, using different deep learning models and support vector machines

被引:79
作者
Rodriguez Gamboa, Juan C. [1 ]
da Silva, Adenilton J. [2 ]
Araujo, Ismael C. S. [3 ]
Eva Susana Albarracin, E. [1 ]
Duran, Cristhian M. A. [4 ]
机构
[1] Univ Fed Rural Pernambuco UFRPE, Dept Estat & Informat, Rua Dom Manoel de Medeiros S-N, BR-52171900 Recife, PE, Brazil
[2] Univ Fed Pernambuco UFPE, Ctr Informat, Recife, PE, Brazil
[3] Univ Fed Rural Pernambuco, Dept Comp, Recife, PE, Brazil
[4] Univ Pamplona, Fac Ingn & Arquitectura, GISM Grp, Pamplona, Ndes, Colombia
关键词
Electronic nose; E-Nose; Rapid detection; Datasets; Deep learning; Real-time classification; SENSOR ARRAYS;
D O I
10.1016/j.snb.2020.128921
中图分类号
O65 [分析化学];
学科分类号
070302 ; 081704 ;
摘要
Real-time gas classification is an essential issue and challenge in applications such as food and beverage quality control, accident prevention in industrial environments, for instance. In recent years, the Deep Learning (DL) models have shown great potential to classify and forecast data in diverse problems, even in the electronic nose (E-Nose) field. In this work, a Support Vector Machine (SVM) algorithm and three different DL models were used to validate the rapid detection approach (based on processing an early portion of raw signals and a rising window protocol) over diverse measurement conditions. We performed a set of experiments with five different E-Nose databases, including fifteen datasets to be used with these algorithms. Based on the obtained results, we concluded that the proposed approach has a high potential and reduces the response time for making E-nose forecasts. Because in more than 60 % of the cases, it achieved reliable estimates using only the first 30 % or fewer of measurement data (counted after the gas injection starts). The findings suggest that the rapid detection approach generates reliable forecasting models using different classification methods. Moreover, SVM seems to achieve the best accuracy and better training time.
引用
收藏
页数:7
相关论文
共 21 条
[1]  
Rodríguez MA, 2010, EGA-REV EXPRES GRAF, P10
[2]  
[Anonymous], **DATA OBJECT**, DOI DOI 10.17632/7SPD6FPVYK.1
[3]  
[Anonymous], 2012, P ADV NEUR INF PROC
[4]  
Araujo I.C., 2019, P AN 16 ENC NAC INT, P844
[5]  
Buitinck L., 2013, ABS13090 CORR
[6]   Calibration transfer and drift counteraction in chemical sensor arrays using Direct Standardization [J].
Fonollosa, J. ;
Fernandez, L. ;
Gutierrez-Galvez, A. ;
Huerta, R. ;
Marco, S. .
SENSORS AND ACTUATORS B-CHEMICAL, 2016, 236 :1044-1053
[7]   Data set from chemical sensor array exposed to turbulent gas mixtures [J].
Fonollosa, Jordi ;
Rodriguez-Lujan, Irene ;
Trincavelli, Marco ;
Huerta, Ramon .
DATA IN BRIEF, 2015, 3 :216-220
[8]   Chemical Discrimination in Turbulent Gas Mixtures with MOX Sensors Validated by Gas Chromatography-Mass Spectrometry [J].
Fonollosa, Jordi ;
Rodriguez-Lujan, Irene ;
Trincavelli, Marco ;
Vergara, Alexander ;
Huerta, Ramon .
SENSORS, 2014, 14 (10) :19336-19353
[9]   Deep Residual Learning for Image Recognition [J].
He, Kaiming ;
Zhang, Xiangyu ;
Ren, Shaoqing ;
Sun, Jian .
2016 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2016, :770-778
[10]   Recognition and sensing of organic compounds using analytical methods, chemical sensors, and pattern recognition approaches [J].
Jha, Sunil Kr ;
Yadava, R. D. S. ;
Hayashi, Kenshi ;
Patel, Nilesh .
CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS, 2019, 185 :18-31