Electronic Nose System Research Articles

E-Nose System Based on Fourier Series for Gases Identification and Concentration Estimation From Food Spoilage

This work presents an electronic nose (EN)-based gas identification and concentration estimation method for the detection of food spoilage. The response data of sensors were acquired through a commercial gas sensor array and data acquisition circuit board and transformed into pictures with the form of the Fourier series. A convolutional neural network (CNN) model was used to identify the pictures from the conversion of sensor data, thus achieving the purpose of identifying the gases (C2H5OH, NH3, and H2S). In order to solve the problem of sample imbalance and to improve the generalization performance of classification models, the synthetic minority oversampling technique (SMOTE) and dropout technique were employed. Fivefold cross-validation was used to evaluate the performance of the model, of which the gas identification accuracy rate reached 96.67%. Moreover, a gas concentration regression model with the advantages of simplicity and strong interpretability was further proposed to estimate the concentrations of C2H5OH, NH3, and H2S. The mean absolute errors and coefficients of determination for the concentration estimation of C2H5OH, NH3, and H2S are (3.71 ppm, 0.968), (0.50 ppm, 0.968), and (0.13 ppm, 0.99), respectively. Furthermore, we used our model to evaluate the freshness of kiwifruit, pork, and beef, and it showed satisfactory predictive performance. The method proposed in this work realizes high-precision detection of gases from food spoilage and has a good application prospect in the rapid judgment of food freshness on the EN system.

Aroma modeling and quality evaluation of spearmint (Mentha spicata subsp. spicata) using electronic nose technology coupled with artificial intelligence algorithms

The present study aimed to evaluate the capability of an electronic nose system coupled with chemometrics tools as a fast, real-time, and non-destructive technology for spearmint grading. The developed e-nose system was utilized to measure 46 samples' volatile organic components (VOCs) procured from different geographical origins. PCA was performed for original data clustering and dimensionality reduction of the aroma data. PLS-DA and fuzzy clustering were performed for differentiating the samples. Fuzzy clustering presented a more reasonable performance due to the overlapping nature of the flowering and vegetative stages clusters. An artificial neural network model was utilized as a supervised pattern recognition algorithm to map the aroma data of the samples to their quality grade based on their essential oils percentage. The developed system predicted the sample’s grade with high correlation coefficients of prediction values (Rp2 = 0.97 and RMSEp= 0.075) and acceptable sensitivity (97.7%) and specificity (86.4%).

A Comprehensive Evaluation Model for Optimizing the Sensor Array of Electronic Nose

As the acquisition unit of gas information, the sensor array directly determines the overall performance of the electronic nose system (E-nose). This paper proposed a new method for optimizing the sensor array. Firstly, four evaluation indicators (sensitivity, selectivity, correlation, and repeatability) were selected to evaluate the sensor array. Subsequently, different evaluation indicators were assigned different weight values according to their contributions to the overall performance of the E-nose. Finally, a comprehensive evaluation model was established based on the EWM-TOPSIS algorithm to optimize the sensor array. In order to verify the effectiveness of the as-proposed model, it was applied to the optimization of the E-nose sensor array composed of 10 gas sensors, and the influence of the sensor array optimization on the gas recognition ability of the E-nose was investigated. The experimental results showed that the optimized sensor array can identify the CO-CH4 gas mixtures with an accuracy of 96.5%, which a significant improvement compared with the accuracy of 78.3% before the sensor array optimization.

Development of a Nondestructive Moldy Coffee Beans Detection System Based on Electronic Nose

Coffee drinks prepared from moldy coffee beans can adversely affect human health. No convenient screening method for detecting the smell of stale coffee beans exists. Accordingly, this study developed an electronic nose (E-nose) system for detecting the smell of coffee beans. This system comprises an environmental control system, a sensor array, and a data signal readout system. The system can distinguish various degrees of mold on coffee beans through the recognition of the smell of the coffee beans. In this study, we established a standard operating procedure to collect gas samples from coffee beans in a temperature- and humidity-controlled environment and recorded changes in the signals by using the sensor array after introducing the target gas. Features were first extracted from the collected data, then dimensionality reduction methods such as principal component analysis (PCA) and linear discriminant analysis (LDA) were applied to these features. Thus, their complexity was reduced, and the noise was eliminated. K-nearest neighbor (KNN) and support vector machine (SVM) were adopted as classification algorithms, and the classification accuracy of the proposed system reached 91.77%.

Multivariate Regression in Conjunction with GA-BP for Optimization of Data Processing of Trace NO Gas Flow in Active Pumping Electronic Nose.

Exhaled nitric oxide trace gas at the ppb level is a biomarker of human airway inflammation. To detect this, we developed a method for the collection of active pumping electronic nose bionic chamber gas. An optimization algorithm based on multivariate regression (MR) and genetic algorithm-back propagation (GA-BP) was proposed to improve the accuracy of trace-level gas detection. An electronic nose was used to detect NO gas at the ppb level by substituting breathing gas with a sample gas. The impact of the pump suction flow capacity variation on the response of the electronic nose system was determined using an ANOVA. Further, the optimization algorithm based on MR and GA-BP was studied for flow correction. The results of this study demonstrate an increase in the detection accuracy of the system by more than twofold, from 17.40%FS before correction to 6.86%FS after correction. The findings of this research lay the technical groundwork for the practical application of electronic nose systems in the daily monitoring of FeNO.

Smooth Discriminant Analysis Combined with an Electronic Nose System to Classify the Gas Information of Beer

Smooth Discriminant Analysis Combined with an Electronic Nose System to Classify the Gas Information of Beer

PRECONDITIONS FOR THE CREATION OF A MEAT FRESHNESS CONTROL AND IDENTIFICATION SYSTEM

The relevance of creating a comprehensive system for meat control and identification to determine its freshness level has been demonstrated in the study. The drawbacks of traditional organoleptic and laboratory methods commonly used for meat inspection were analyzed. The authors presented the advantages and challenges of employing an electronic nose. A design for a meat control and identification system was proposed, which includes an Arduino Uno microcontroller, Raspberry Pi, USB to TTL adapter, gas sensors, color sensor, thermal camera, and image sensor. The proposed implementation of the electronic nose system on a single-board computer demonstrates its success in controlling and identifying meat freshness. A matrix of semiconductor gas sensors, TGS2602, MQ137, and MQ138, is formed as olfactory sensors, and TCS3200 is used as an RGB vision sensor, enabling the identification of the smell and color of different degrees of meat freshness. To obtain clear output differences from the gas sensors that react to the freshness level of meat, the baseline method is proposed for use. Therefore, a system enhanced with neural network capabilities will replace traditional devices for identifying meat freshness.

Ultra-Low-Power E-Nose System Based on Multi-Micro-LED-Integrated, Nanostructured Gas Sensors and Deep Learning.

As interests in air quality monitoring related to environmental pollution and industrial safety increase, demands for gas sensors are rapidly increasing. Among various gas sensor types, the semiconductor metal oxide (SMO)-type sensor has advantages of high sensitivity, low cost, mass production, and small size but suffers from poor selectivity. To solve this problem, electronic nose (e-nose) systems using a gas sensor array and pattern recognition are widely used. However, as the number of sensors in the e-nose system increases, total power consumption also increases. In this study, an ultra-low-power e-nose system was developed using ultraviolet (UV) micro-LED (μLED) gas sensors and a convolutional neural network (CNN). A monolithic photoactivated gas sensor was developed by depositing a nanocolumnar In2O3 film coated with plasmonic metal nanoparticles (NPs) directly on the μLED. The e-nose system consists of two different μLED sensors with silver and gold NP coating, and the total power consumption was measured as 0.38 mW, which is one-hundredth of the conventional heater-based e-nose system. Responses to various target gases measured by multi-μLED gas sensors were analyzed by pattern recognition and used as the training data for the CNN algorithm. As a result, a real-time, highly selective e-nose system with a gas classification accuracy of 99.32% and a gas concentration regression error (mean absolute) of 13.82% for five different gases (air, ethanol, NO2, acetone, methanol) was developed. The μLED-based e-nose system can be stably battery-driven for a long period and is expected to be widely used in environmental internet of things (IoT) applications.

A Low Cost Electronic Nose System for Classification of Gayo Arabica Coffee Roasting Levels Using Stepwise Linear Discriminant and K-Nearest Neighbor

A low-cost electronic nose (E-Nose) system using metal oxide sensor (MOS) was developed for Gayo arabica coffee roasting level. The developed electronic nose was designed to have a simple, rapid detection, as wells as provides reliable results. The E-Nose system is equipped with MOS sensors, sensor chamber, microcontroller, computer, and data acquisition system. The level of coffee roasting was monitored by read the data from the sensors continuously in real time every second. The sensor signals were recorded in Excel file using data acquisition system and analysed using both stepwise linear discrimination and k-nearest neighbor classifiers. A high percentage (91.67%) of accuracy was obtained using stepwise linear discrimination method. Furthermore, k-nearest neighbor classifier using city block distance demonstrated higher accuracy than stepwise linear discrimination classifier. The results showed that the electronic nose system has a potential for assessing Gayo arabica coffee roasting level. The study confirmed that the proposed electronic nose equipped with at least two MOS sensors was suitable for monitoring the level of coffee roasting level. The result could be used for evaluating other varieties of roasted coffee.

The Quality Detection and Recognition for Food Seasoning Based on an Artificial Olfactory System

The quality of food seasoning affects the flavor and the safety of the food. Traditional detection methods, such as test-paper, cyclotron or chromatography, have some limits in accuracy, detection range, time consumption, portability and cost. The artificial olfactory system (AOS) can detect and identify the food quality by simulating animal olfactory sensors, which has the advantages of short detection cycle, high sensitivity, and no complicated pre-processing. A sample data source for AOS to detect pepper powder was built by collecting the odor data of the condiment powders under different origins and adulteration conditions using the sensor array of the PEN3 electronic nose system (AIRSENSE, Germany). Then, machine learning algorithms including support vector machine, decision tree and random forest were introduced to form the intelligent models to evaluate the quality of the pepper powder from different origins and with different adulterated substances. The results indicated that the recognition accuracy reached to over 98% for all types of pepper samples with the radial basis function-support vector machine (RBF-SVM) model. The pure pepper could be distinguished from the adulterated samples successfully with the accuracy of near 100% using machine leaning models. Finally, other more machine learning algorithms were compared to recognize the states of different pepper powders. The results showed that the SVM method has the highest accuracy in classifying the adulteration of pepper powder among different machine learning models discussed in the research. It is feasible to detect and recognize the quality of the food seasoning with AOS based on machine learning algorithms and sensor array.

Estimating the Potential of the Electronic Nose System for Monitoring Disturbances in the Functional Obesity of Children in a Hospital: Noninvasive Diagnosis without Biosampling

A portable “diagnostic electronic nose” was used to study the composition of a mixture of volatile compounds emitted by the skin of the forearms in children with obesity and other associated metabolic disorders. Differences and features of the profiles of volatile compounds (primarily organic) of the skin on the left and right forearms in children with obesity and the dependence of the profile on food intake, drugs, concomitant diseases, and body characteristics were evaluated in hospital environment. More than 300 patient profiles were analyzed for 2 years, including the dynamics in profiles during repeated planned hospitalizations. The results suggest that the composition of the gas mixture of compounds emitted by the skin on the forearms can be analyzed and compared without preliminary separation or selection using an array of eight piezoelectric sensors of the “diagnostic electronic nose” device. We measured a correlation between the electronic nose data and the results of standard clinical laboratory tests. The parameters of a sensor array and clinical studies are closely related to each other in the levels of urea, alkaline phosphatase, and glucose. An approach based on scanning a volatile skin profile on the forearm for 3–6 min helps to rank children with obesity into groups associated with the socially significant diseases and conditions: “Normal,” “Obesity with a risk of development of type 2 diabetes mellitus,” and “Type 1 diabetes mellitus.” Using chemometric methods, we optimized the analytical procedure and specified research conditions and control points. The composition of the volatile skin metabolome on the right forearm, scanned on an empty stomach, is the most informative. The accuracy of the difference between the problem groups and the norm is 100%; the accuracy of the rank within pathological groups is 80%. The proportion of false assignments increases in the presence of inflammation in a body of any type. The potentials of the procedure were evaluated in a hospital environment.

Portable electronic nose system with elastic architecture and fault tolerance based on edge computing, ensemble learning, and sensor swarm

The portable electronic nose (E-nose) systems are suffering from the limited computing ability of microcontrollers and can only adopt simple pattern recognition algorithms. The heavy coupling between the sensors greatly limits the anti-fault ability of the system. Herein, a novel ensemble learning framework based on independent artificial neural networks (ANN) is proposed in a portable E-nose system to recognize volatile organic compounds (VOCs). Edge computing is applied to complete data processing and analysis on an ARM Cortex-M3 based microcontroller in an E-nose system. Each sensing unit can recognize VOCs by training independent ANN models. Then an ensemble learning method further organizes the diverse ANN models into an onboard sensor swarm with a significantly enhanced performance. The accuracy of the type classification can reach 81.1%, which is improved by more than 20% compared with the best individual classifier. The R2 score of concentration prediction can reach 84.1%, which is improved by more than 25% compared with the best regressor in the ensemble group. For both type identification and concentration prediction, the tolerance to the sensor faults of swarm-based E-nose is 10 and 18 times greater than that of the conventional array-based sensors, respectively. Our work has designed an elastic architecture based on sensor swarms, providing a novel avenue for the development of E-nose systems with high fault tolerance.

An adaptive learning method for the fusion information of electronic nose and hyperspectral system to identify the egg quality

Data fusion technology based on the multi-sensor system can obtain the holistic properties of samples. However, multi-sensor data fusion will bring more redundant information, which will lead to low classification performance. In this work, a multi-data-fusion-attention network (MDFA-Net) is proposed, combined with the electronic nose (e-nose) and hyperspectral system to identify the egg quality. Firstly, the gas information and spectral information of eggs are obtained under different feeding conditions. Secondly, a feature adaptive learning (FAL) unit is designed to select effective information and enhance the ability of feature expression. Thirdly, based on the FAL unit, a decision network is formed to identify the fusion information of e-nose and hyperspectral system. Finally, compared with other deep learning network models, the accuracy of MDFA-Net is 99.88%, the precision is 99.87%, the recall is 99.88%, and the F 1 -score is 99.90%, which shows better classification performance and stability. • E-nose and hyperspectral system are applied to gain the quality information of egg. • An FAL unit is proposed to fuse the gas and spectral information. • Attention mechanism is proposed to adaptively focus on important features. • An MDFA-Net is proposed to achieve the decision of fusion information.

Detection and Extraction of Aroma Characteristics of Fuel Oil Using Gas Sensors Through Electronic-Nose System

Fuel oil is an indispensable strategic source of energy for the fuel of vehicles and other engines. This study aims to detect and differentiate fuel oil using an electronic-nose system. The fuel oil used is Pertalite, Pertamax, and Pertamax Turbo. Four gas sensors are used, namely MQ 4, MQ 7, MQ 9, and MQ 136. The average output voltage values of Pertalite, Pertamax, and Pertamax Turbo fuel oil for MQ 4 sensors are 1.07 V, 1.22 V, and 0.96 V. For MQ 7 sensors, the output voltage values of the samples are 1.44 V, 1.43 V, and 1.37 V, respectively. For the MQ 9 gas sensor, the sample output voltage values are 1.23 V, 1.43 V, and 1.09 V. As for the MQ 136 gas sensor, the output voltage value of each oil is 1.26 V, 1.25 V, and 0.91 V. Sensors that provide the highest response in each sample are MQ 136 sensors. The electronic-nose system can extract characteristics from all three samples using the principal component analysis (PCA) method.

Electronic nose signals-based deep learning models to realize high-precision monitoring of simultaneous saccharification and fermentation of cassava

This study innovatively proposes a high-precision monitoring method for key parameters in the process of ethanol production from simultaneous saccharification and fermentation (SSF) by electronic nose technology combined with recurrent neural network (RNN). A PEN3 electronic nose system was employed to acquire the odor information of the fermented samples, and four deep learning algorithms based on the RNN architecture were employed to design reasonable network structures to realize the deep learning of the electronic nose signal features and model calibration. The results obtained showed that each deep learning model based on the RNN architecture has good generalization performance for the determination of cassava SSF process parameters. Among them, the bidirectional long short-term memory network (BiLSTM) model has the best monitoring effect on ethanol content, with root mean square error of prediction (RMSEP) of 3.7 mg·mL−1 and coefficient of predictive determination (RP2) of 0.98 and the relative percent deviation (RPD) of 8.1. The bidirectional gated recurrent unit (BiGRU) model had the best monitoring effect on glucose content, and its RMSEP, RP2 and RPD were 2.9 mg·mL−1, 0.99 and 9.1, respectively. The overall results reveal that deep learning algorithms have promising application prospects in the feature mining and model calibration of electronic nose signals, which provides an effective analysis tool for in-situ monitoring of electronic nose technology in modern industrial fermentation processes.

An AI-powered Electronic Nose System with Fingerprint Extraction for Aroma Recognition of Coffee Beans.

Aroma and taste have long been considered important indicators of quality coffee. Specialty coffee, that is, coffee from a single estate, farm, or village in a coffee-growing region, in particular, has a unique aroma that reflects the coffee-producing region. In order to enable the traceability of coffee origin, in this study we have developed an e-nose system to discriminate the aroma of freshly roasted coffee in different production regions. In the case study, we employed the e-nose system to experiment with various machine learning models for recognizing several collected coffee beans such as coffees from Yirgacheffe and Kona. Additionally, our contribution also includes the development of a method to create an aromatic digital fingerprint of a specific coffee bean to identify its origin. The experimental results show that the developed e-nose system achieves good recognition performance for coffee aroma recognition. The extracted digital fingerprints have great potential to be stored in an extensible coffee aroma database similar to a comprehensive library of specific coffee bean aroma characteristics, for traceability and reconfirmation of their origin.

Cross-Domain Active Learning for Electronic Nose Drift Compensation.

The problem of drift in the electronic nose (E-nose) is an important factor in the distortion of data. The existing active learning methods do not take into account the misalignment of the data feature distribution between different domains due to drift when selecting samples. For this, we proposed a cross-domain active learning (CDAL) method based on the Hellinger distance (HD) and maximum mean difference (MMD). In this framework, we weighted the HD with the MMD as a criterion for sample selection, which can reflect as much drift information as possible with as few labeled samples as possible. Overall, the CDAL framework has the following advantages: (1) CDAL combines active learning and domain adaptation to better assess the interdomain distribution differences and the amount of information contained in the selected samples. (2) The introduction of a Gaussian kernel function mapping aligns the data distribution between domains as closely as possible. (3) The combination of active learning and domain adaptation can significantly suppress the effects of time drift caused by sensor ageing, thus improving the detection accuracy of the electronic nose system for data collected at different times. The results showed that the proposed CDAL method has a better drift compensation effect compared with several recent methodological frameworks.

A Transferable Feature-Based Classifier to Improve Transferability of Electronic Nose Systems

Sensor drift is an important issue in electronic nose systems. When sensor drift occurs, the data distribution changes, which affects classification accuracy. Thus, this letter proposes a feature selection method and a classifier combined with weighted features to address the sensor drift problem. In the proposed method, feature scores used to select better features are calculated based on their separability and transferability in feature selection. In addition, to consider the transferability of each feature, the transferability score of each feature is used as a feature weight. For the classifier, to reduce the distribution difference between two domains, an extreme learning machine is employed, which make the output weights of the target domain similar to those of the source domain while considering the importance of each feature in terms of the feature scores. Experimental results demonstrate that the average classification accuracy on an open-access drifted dataset can be improved using the proposed method.

An Olfactory-Taste Synesthesia Model Combined With Electronic Nose and Electronic Tongue to Identify Flavor Substances

The main task of the electronic nose (e-nose) and electronic tongue (e-tongue) is to identify flavour substances. The human olfactory-taste nerve mechanisms affect the identification of flavour substances by the e-nose and e-tongue systems. Therefore, this study proposed an olfactory-taste synesthesia model built on the human olfactory-taste nerve pathways to process the e-nose and e-tongue data for flavour substance identification. Next, this study simulated the power spectral density and synchronization for the synesthesia model nodes and validated the bionic performance of the synesthesia model. The results suggested that, first, the synesthesia model characterized the dynamic characteristics of the human olfactory-taste nerve conduction pathways. Second, during stimulation, the power spectral density for the synesthesia model nodes exhibited 1/ <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$f$ </tex-math></inline-formula> characteristics. The synesthesia model nodes exhibited synchronization upon stimulation. The 1/ <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$f$ </tex-math></inline-formula> characteristics and synchronization indicated the bionic performance of the synesthesia model. Finally, this study chose beer and apple samples as flavour substances and input the olfactory and taste information for flavour substances into the synesthesia model for identification. Compared with the identification results of a single e-nose, single e-tongue, and the combined electronic nose and tongue systems under the multipattern recognition models, the synesthesia model acquired best accuracy, Kappa coefficients, and F <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sub> -scores in the identification for the beer and apple samples. In conclusion, the effective identification results for flavour substances can be achieved by the synesthesia model.

Preliminary Studies on Detection of Fusarium Basal Rot Infection in Onions and Shallots Using Electronic Nose.

The evaluation of crop health status and early disease detection are critical for implementing a fast response to a pathogen attack, managing crop infection, and minimizing the risk of disease spreading. Fusarium oxysporum f. sp. cepae, which causes fusarium basal rot disease, is considered one of the most harmful pathogens of onion and accounts for considerable crop losses annually. In this work, the capability of the PEN 3 electronic nose system to detect onion and shallot bulbs infected with F. oxysporum f. sp. cepae, to track the progression of fungal infection, and to discriminate between the varying proportions of infected onion bulbs was evaluated. To the best of our knowledge, this is a first report on successful application of an electronic nose to detect fungal infections in post-harvest onion and shallot bulbs. Sensor array responses combined with PCA provided a clear discrimination between non-infected and infected onion and shallot bulbs as well as differentiation between samples with varying proportions of infected bulbs. Classification models based on LDA, SVM, and k-NN algorithms successfully differentiate among various rates of infected bulbs in the samples with accuracy up to 96.9%. Therefore, the electronic nose was proved to be a potentially useful tool for rapid, non-destructive monitoring of the post-harvest crops.