Narrow Results

Volatile organic compounds (VOCs) are ubiquitous in the surroundings, originating from both industrial and natural sources. VOCs directly impact the quality of both indoor and outdoor air and play a significant role in processes such as fruit ripening and the body’s metabolism. VOC monitoring has seen significant growth recently, with an emphasis on developing low-cost, portable sensors capable of both vapor discrimination and concentration measurements. VOC sensing remains challenging, mainly because these compounds are nonreactive, appear in low concentrations and share similar chemical structures that results in poor sensor selectivity. Therefore, individual gas sensors struggle to selectively detect target VOCs in the presence of interferences. Electronic noses overcome these limitations by employing machine learning for pattern recognition from arrays of gas sensors. Here, an electronic nose fabricated with four types of functionalized gold nanoparticles demonstrates rapid detection and quantification of eight types of VOCs at four concentration levels. A robust two-step machine learning pipeline is implemented for classification followed by regression analysis for concentration prediction. Random Forest and support vector machine classifiers show excellent results of 100% accuracy for VOC discrimination, independent of measured concentration levels. Each Random Forest regression analysis exhibits high R² and low RMSE with an average of 0.999 and 0.002, respectively. These results demonstrate the ability of gold nanoparticle gas sensor arrays for rapid detection and quantification.

The aim of this study is to test the freshness of horse mackerels by using a low cost electronic nose system composed of eight different metal oxide sensors. The process of freshness evaluation covers a scala of seven different classes corresponding to 1, 3, 5, 7, 9, 11, and 13 storage days. These seven classes are categorized according to six different classifiers in the proposed binary decision tree structure. Classifiers at each particular node of the tree are individually trained with the training dataset. To increase success in determining the level of fish freshness, one of the k-Nearest Neighbors (k-NN), Support Vector Machines (SVM), Linear Discriminant Analysis (LDA) and Bayes methods is selected for every classifier and the feature spaces change in every node. The significance of this study among the others in the literature is that this proposed decision tree structure has never been applied to determine fish freshness before. Because the freshness of fish is observed under actual market storage conditions, the classification is more difficult. The results show that the electronic nose designed with the proposed decision tree structure is able to determine the freshness of horse mackerels with 85.71% accuracy for the test data obtained one year after the training process. Also, the performances of the proposed methods are compared against conventional methods such as Bayes, k-NN, and LDA.

Roasting process needs to be monitored and carefully controlled because it plays as the most important stage for determining flavor quality on the final product in the secondary coffee processing. Common quality monitoring method by applying parameters namely roasting time, roasting temperature and grain color may have disadvantages especially for nonuniform quality of green beans and stirring mechanism of regular roasters; therefore, an alternative quality monitoring model is necessary. Because emitted vapor during roasting may represent the occurred reaction stage, it is possible to indicate the roast degree of the coffee grain. This study evaluated the application of an electronic nose based on semiconductor sensor array for quality monitoring of coffee roasting. The electronic nose designed with gas sensor array was integrated to a mini batch coffee roaster. Data including sensor array response, vapor humidity and temperature were recorded in line to the roasting process and compared to the coffee grain color measured with a universal colorimeter. The experiment showed that the gas sensors respond to the emitted coffee flavors, from which a logarithmic profile as a function of grain color was obtained with the highest slope of $-$1.6V/color difference of roasted coffee. Aroma patterns obtained from sensor responses were then analyzed with principle component analysis (PCA), by which a distinctive profile between initial and final phases of roasting is obtained. Although the corresponding analysis is still unable to distinguish the levels of light, medium and dark (LMD); high sensor responses indicate a further benefit of this system for developing an analogous quality monitoring system.

Drought stress is one of the most significant abiotic stresses, adversely affecting the economy by tumbling or even eliminating agricultural productivity, development, and yield. Adverse effects of water scarcity can be reduced if certain precautionary actions could be taken in advance. Therefore, monitoring and early detection of the drought can be helpful for preparing a well-developed response plan. In response to the stresses, an intricate response system of the plants is involved which emits a range of Volatile Organic Compounds (VOCs) from different parts, such as flowers, leaves, roots and stems. These VOCs can be used as fingerprints for categorizing stressed and nonstressed plants. This paper addresses the optimization of an array of gas sensors used in an in-situ stress diagnosis system, and the data obtained after optimization have been used for the detection of induced stresses in plants by recording the VOCs emitted by the plants. The flow characteristics of the gas chamber were modeled using the Finite Element method before fabrication. The temperature modulation of the gas sensors used in the designed electronic nose system was accomplished. The optimization of the sensor array was performed using the radar plot and Wilks’ Lambda technique. The optimum operating temperatures for each gas sensor were selected using a radar plot. Furthermore, the number of the sensors in the sensor array was reduced by choosing the sensors having higher discriminant ability using the Wilks’ Lambda optimization technique. Twelve healthy Khasi Mandarin Orange saplings were considered for the investigation. Three different levels of water stresses are induced in the plants artificially for the experiments. The overall response and the optimized response of the developed electronic nose system are compared using Linear Discriminant Analysis (LDA) and bootstrap ensemble K-Nearest Neighbors (KNN) classifier. The Leaf Relative Water Content (LRWC) of the leaves is also measured concurrently to confirm the stress induction in the plants.

The memristive neuromorphic computing system (MNCS) can complete related calculations with lower power consumption and higher speed, which has attracted widespread attention. However, due to the limitations of memristor and circuit, the realization of MNCS faces many challenges. In this paper, we propose a heterogeneous deployment strategy for the MNCS and construct a lightweight heterogeneous memristive gas classification neural network (LHM-GSNN) based on the electronic nose (e-nose) application. In addition, the model parameters are quantified by clustering strategy to adapt to the nonideal characteristics of memristor. The experimental results show that the complex structure in the model is visibly simplified, and the number of parameters is correspondingly reduced using the heterogeneous deployment strategy. Furthermore, we also analyze the power consumption of the LHM-GSNN model deployed to the MNCS. This work may provide new solutions for constructing and implementing the MNCS.

Gas detection plays different roles in different environments. Traditional algorithms implemented on electronic nose for gas detection and recognition have high complexity and cannot resist device drift. In response to the above issues, we propose a convolutional neural network based on memristive Stochastic Computing (SC), which combines the characteristics of small devices and low power consumption of memristor devices, as well as the fast and fault-tolerant random calculation speed. It can effectively utilize hardware advantages, recognizing gases by electronic nose. The experimental results show that for two different gas sensor array datasets, the accuracy of the proposed method can achieve the level of 99%. When using memristive SC for deduction, the accuracy decreases by less than 1%, but in drift data, the accuracy can be improved by about 3%. Finally, the improvement in area, power, and energy compared to inference in GPU (NVIDIA Geforce RTX 3060 Laptop) is 1104X, 48X, and 9X, respectively.

The average classification accuracy of an odor classification system is improved using a genetic algorithm to determine optimal parameters for feature extraction. Gaussian windowing functions, called "kernels" are evolved to extract information from the transient response of an array of gas sensors, resulting in a reduced set of extracted features for a linear discriminant pattern classification system. Results show significant improvements are achieved when compared to results obtained using a predetermined and fixed set of four bell-shaped kernels for every sensor. Examination of the evolved kernels reveals the areas of the sensor responses where discriminating information was identified. A novel data migration approach during training helps prevent overtraining, and the fitness measure chosen incorporates adjustments for both population diversity and solution complexity. A variety of adjustable parameters, including the definition of a time-varying dynamic weighting factor, encourage experimentation in order to appropriately tune the sampling methods and fitness measure.

The problem of automating the sensing and classification of odours is one which promises a wide range of industrial applications. During the INTESA project, a prototype electronic nose was developed, using sensors based on novel conducting polymer materials and also more traditional MOS materials. The software component of the prototype processes the transient resistance change signals recorded by the hardware, and classifies the odour sample into one of a number of "odour classes". This paper describes two of the soft computing methods investigated for learning classification rules in this domain. The first method builds on previous work done on the Fril data browser, using clustering, fuzzy matching, Fril rules and evidential logic rules. The second method uses a fuzzy extension of the ID3 decision tree induction method, called "mass assignment tree induction (MATI)". Some of the results of applying these methods to data obtained from the INTESA prototype are presented and discussed.

Resistance noise data from a single gas sensor can be utilized to identify gas mixtures. We calculated the power spectral density. higher order probability densities and the bispectrum function of the recorded noise samples; these functions are sensitive to different natural vapors and can be employed to select a proper detection criterion for gas composites and odors.

Sensing is a key element for any intelligent robotic system. This paper describes the current progress of a project in the Intelligent Robotics Research Center at Monash University that has the aim of developing a synergistic set of sensory systems for a humanoid robot. Currently, sensing modes for colour vision, stereo vision, active range, smell and airflow are being developed in a size and form that is compatible with the humanoid appearance. Essential considerations are sensor calibration and the processing of sensor data to give reliable information about properties of the robot's environment. In order to demonstrate the synergistic use of all of the available sensory modes, a high level supervisory control scheme is being developed for the robot. All time-stamped sensor data together with derived information about the robot's environment are organized in a blackboard system. Control action sequences are then derived from the blackboard data based on a task description. The paper presents details of each of the robot's sensory systems, sensor calibration, and supervisory control. Results are also presented of a demonstration project that involves identifying and selecting mugs containing household chemicals. Proposals for future development of the humanoid robot are also presented.

This work investigates the potential use of temperature modulation of MOS gas sensors combined with the Hilbert–Huang transform (HHT) as a feature extraction mechanism for MOS-based electronic noses. Five samples each of ethyl acetate, ethanol and isopropanol were prepared. The response of each of four sensors in an array was decomposed using empirical mode decomposition and the marginal Hilbert spectrum was computed. A set of 72 frequency components was extracted from marginal Hilbert spectrum response of each sensor in an array of four sensor to produce a 288 element fingerprint of each sample. The fingerprints were successfully clustered using PCA and classified using a SVM neutral network.

The unique semiconducting, optical and electrochemical properties of radical lanthanide bisphthalocyanines make them ideal materials for sensing applications. A variety of chemical sensors have been developed using rare-earth bisphthalocyanine thin films. In this paper, the characteristics of sensors based on bisphthalocyanines are reviewed. The advantages of these sensors with respect to sensors developed using other metallophthalocyanines are discussed. Resistive sensors based on bisphthalocyanines change their conductivity when exposed to a variety of pollutant gases and volatile organic compounds. Because bisphthalocyanines are intrinsic semiconductors, the conductivity of their thin films is higher than the conductivity of metallophthalocyanine thin films. This facilitates the electrical measurements and enhances the sensitivity of the sensors. Optical sensors have also been developed based on the rich optical properties shown by bisphthalocyanines. Films characterized by a bright green color change to red or to blue upon oxidation or reduction. The changes also affect the charge-transfer band associated to the free radical that bisphthalocyanines show in the near infrared region. This band coincides with telecommunication wavelengths, making possible the fabrication of fiber optic sensors where a phthalocyanine film is deposited at one of the ends of the fiber. Electrochemical sensors have been developed taking advantage of the unique electrochemical behavior associated to the one-electron oxidation and one-electron reduction of the phthalocyanine ring. These reversible processes are extremely sensitive to the nature of the electrolytic solution. This has made possible the development of voltammetric sensors able to produce particular signals when immersed in different liquids. In the last part of the paper, the fundamentals and performance characteristics of electronic noses and electronic tongues based on bisphthalocyanines are described. Such devices have been successfully exploited in quality control, classification, freshness evaluation and authenticity assessment of a variety of food, mainly wines and olive oils.

Chemical sensor layers for environmental applications require optimal selectivity, sensitivity, and long term stability, which can be achieved in artificial matrices. For detecting thiols in air, reversible affinity interactions can be optimized by varying the stoichiometry of molybdenum disulphide nanoparticles to achieve sulphur deficiencies. Generating MoS_1.9 increases the quartz crystal microbalance (QCM) sensor responses towards butane thiol by a factor of three. Artificial recognition sites are accessible by molecular imprinting: acrylate copolymers can be tuned in polarity to interact selectively with atrazine in water leading to detection limits below one ppb with QCM sensors. Finally, sensor arrays coated with six different molecularly imprinted polymers (MIP) correctly reproduce the ethyl acetate concentration of a composter over a period of two weeks validated by GC-MS measurements.

Resistance noise data from a single gas sensor can be utilized to identify gas mixtures. We calculated the power spectral density, higher order probability densities and the bispectrum function of the recorded noise samples; these functions are sensitive to different natural vapors and can be employed to select a proper detection criterion for gas composites and odors.

Volatile organic compounds (VOCs) are ubiquitous in the surroundings, originating from both industrial and natural sources. VOCs directly impact the quality of both indoor and outdoor air and play a significant role in processes such as fruit ripening and the body’s metabolism. VOC monitoring has seen significant growth recently, with an emphasis on developing low-cost, portable sensors capable of both vapor discrimination and concentration measurements. VOC sensing remains challenging, mainly because these compounds are nonreactive, appear in low concentrations and share similar chemical structures that results in poor sensor selectivity. Therefore, individual gas sensors struggle to selectively detect target VOCs in the presence of interferences. Electronic noses overcome these limitations by employing machine learning for pattern recognition from arrays of gas sensors. Here, an electronic nose fabricated with four types of functionalized gold nanoparticles demonstrates rapid detection and quantification of eight types of VOCs at four concentration levels. A robust two-step machine learning pipeline is implemented for classification followed by regression analysis for concentration prediction. Random Forest and support vector machine classifiers show excellent results of 100% accuracy for VOC discrimination, independent of measured concentration levels. Each Random Forest regression analysis exhibits high R² and low RMSE with an average of 0.999 and 0.002, respectively. These results demonstrate the ability of gold nanoparticle gas sensor arrays for rapid detection and quantification.

Metal oxide gas sensors (MOXs) are widely used in olfactory electronic systems for their high sensitivity and low-cost. These sensors modify their conductivity in presence of oxidizing and reducing gases, and their performance is strictly dependent on the measurement technique adopted. In particular, it was already established by many works that a noticeable improvement in selectivity can be obtained by operating MOXs with a variable temperature. In this context, a strong interest in developing simplified models able to predict the sensor response is rising.