Search

We report on the development of a simple and cost-effective potentiometric sensor array that is based on manual “drawing” on the polymeric support with the pencils composed of graphite and different types of zeolites. The sensor array demonstrates distinct sensitivity towards a variety of inorganic ions in aqueous media. This multisensor system has been successfully applied to quantitative analysis of 100 real-life surface waters sampled in Mahananda and Hooghly rivers in the West Bengal state (India). Partial least squares regression has been utilized to relate responses of the sensors to the values of different water quality parameters. It has been found that the developed sensor array, or electronic tongue, is capable of quantifying total hardness, total alkalinity, and calcium content in the samples, with the mean relative errors below 18%.

The objective of this paper is to develop a microcontroller-based, real-time data acquisition, and remotely operated system, also called the Remote Sensing System for a Standalone E-Nose System (RSENS). The Volatile Organic Compounds (VOC) content of a provided sample is detected using Quartz Crystal Microbalance (QCM) sensors and stored. The data obtained is analyzed in the RSENS using its GUI interface attached to it. This paper investigates remote control of the RSENS where the system operation and the data can be accessed remotely by an Android application through a common cloud server that connects both the android application and the RSENS. Moreover, the analyzed data is also visualized on the remote Android platform. The RSENS can analyze and store the data in the cloud server wherein real-time monitoring application can retrieve it from the server and perform additional plots and apply other segregation algorithms, as well as modify the generated data.

E-nose technology has been successfully employed in determination of rancidity in nutraceutical-rich drop cookies. The cookies were formulated with extracts obtained by supercritical carbon dioxide extractions and post-extraction sample matrices of black pepper and small cardamom. Rancidity in cookies was estimated by comparing the odor profiles of stored cookies with the profiles of deliberately rancid cookies (training sets), using e-nose. PCA plots along with spoilage indices obtained from Mahalanobis distances calculated from e-nose responses revealed that the extracts and post-extraction sample matrices of these spices performed as natural antioxidants in preventing rancidity in cookies. Both the extracts enhanced the shelf lives of cookies by at least 120 days; post-extraction matrices of black pepper and small cardamom enhanced the same by 80 and 40 days, respectively. These findings were further affirmed by biochemical analyses and linear regression equations were generated for prediction of FFA content, PV, and MDA values from the spoilage indices of each type of cookie. Thus, spoilage indices can circumvent the requirement of conducting cumbersome biochemical assays for estimation of rancidity in cookies. Similar applications of e-nose technology and spoilage indices can be envisaged for other bakery products also.

Tea industries enjoy a significant position in the socio-economic ladder for any demographics, especially in India who is the largest producer as well as consumer of the agro-product. While tea ranks only next to water in the pedigree of globally consumed beverages, the imperative fermentation stage in the processing of tea leaves is conventionally monitored through olfactory perception of tea tasters. Recent advances in the field of machine olfaction have witnessed the advent of electronic nose prototypes, which provide a scientific validation to the organoleptic estimations disseminated by the tasters. However, fermentation is a continuous process requiring constant monitoring whose successful completion relies heavily on identification of distinct aroma peaks emanated at optimum instants. Since the fermentation process is integral to the final quality, it is deemed beneficial if the optimum fermentation period can be predicted at an earlier stage. Such preemptive information can mitigate constant monitoring requirements and momentary concentration lapses. Recognizing the time series nature of the data generated during the fermentation process with an electronic nose prototype, we have implemented a recurrent Elman network to predict the optimum fermentation period for different black tea samples. The results showed that the prescribed network could predict the optimum period with confidence at the halfway of the process. The minimal error between the predicted and the actual fermentation period at the halfway point suggests that the proposed model can well be integrated with an electronic nose dedicated for monitoring the fermentation process.

The effect of aging on quality of jasmine (Jasminum Sambac) concrete in terms of aroma was studied with the intention of estimating the shelf life of this highly priced export oriented commodity. A handheld electronic nose was developed and customized to find out aroma intensity of fresh and aged commercial concrete samples. The same fresh and aged samples were also analyzed with the help of a gas chromatography–mass spectrometry (GC–MS). Between first and final experiment, the samples were kept in air-tight containers. The electronic nose showed very high sensitivity toward the concrete’s aroma molecules and could detect aroma volatiles from concrete samples at a minimum sample weight of 0.05 g. The e-nose also showed the decrease in aroma intensity in the samples with aging. The GC-MS results established a strong correlation with the e-nose observations. The aroma intensity of concrete was observed to be decreased almost linearly over a period of one year. Electronic nose has been found to be a potent tool for rapid prediction of concrete age.

Mango possesses an elegant smell and attracts consumers due to its high nutritive and medicinal benefits. A proper monitoring of maturity stages of mango is required to increase the shelf life and to boost consumer acceptability. This paper aims to discriminate the maturity stages of three cultivars of mango based on the aroma using quartz crystal microbalance (QCM) based E-nose, which can discriminate the ripening stages.

The paper describes a wide-range practical application of the potentiometric multisensor system (MS) (1) for integral safety evaluation of a variety of natural waters at multiple locations, under various climatic conditions and anthropogenic stress and (2) for close to real consistency evaluation of waste water purification processes at urban water treatment plants. In total, 25 natural surface water samples were collected around St. Petersburg (Russia), analyzed as is, and after ultrasonic treatment. Toxicity of the samples was evaluated using bioassay and MS. Relative errors of toxicity assessment with MS in these samples were below 20%. The system was also applied for fast determination of integral water quality using chemical oxygen demand (COD) values in 20 samples of water from river and ponds in Kolkata (India) and performed with an acceptable precision of 20% to 22% in this task. Furthermore, the MS was applied for fast simultaneous evaluation of COD, biochemical oxygen demand, inorganic phosphorous, ammonia, and nitrate nitrogen at two waste water treatment plants (over 320 samples). Reasonable precision (within 25%) of such analysis is acceptable for rapid water safety evaluation and enables fast control of the purification process. MS proved to be a practicable analytical instrument for various real-world tasks related to water safety monitoring.

The extract of Jasmine called concrete is an important floricultural export commodity of India. The major quality attribute of both jasmine and concrete is fragrance. The fragrance profile of jasmine flower has been studied using a custom built handheld e-Nose (HEN). Three species of jasmine, J. Sambac , J. Grandiflorum , and J. Auriculatum were applied to e-Nose one by one. The principal component analysis of the responses showed that distinct clusters were formed for different species. Another study was made on the mostly cultivated J. Sambac species to find out a relationship between concrete quality with various flowering stages as well as with seasonal variation. Experiments were carried out over 200 samples collected from various gardens of South India. Multivariate sensor data were computed in a single dimensional intensity value termed as fragrance index (FI). Flowering stages were classified by Industry experts in terms of bud opening indicator (BOI). HEN was trained to formulate a scale of BOI in the range of 0–2 based on the FI. Best quality of concrete is found to be obtained from jasmine flower with BOI level of 1.5. Jasmine concrete, produced at laboratory from flowers harvested at different time-of-the-day, i.e., with different BOI levels were analysed using GC-MS to check the volatile emission patterns. Results showed that concrete produced with BOI level 1.5 emitted maximum aroma generating volatiles. These observations showed that e-Nose can be a potential gadget for fragrance profiling as well as selection of best quality flowers for concrete extraction.

When partaking of any food or beverage, we actually perceive its flavour, which is the combined effect of aroma and taste. This parallel processing of signals from our nose and tongue is possible due to the presence of large and complex biological neural networks in our brain. Artificial perception of odor and taste is attempted by the scientists using similar techniques, and is implemented with an array of sensors and an array of electrodes known as electronic nose and tongue, respectively. Fusion of electronic nose and tongue at various levels provides improved results for various applications This review article discusses different techniques of fusing electronic nose and tongue information as well as applications of the combined system in different fields.

Expeller-pressed virgin coconut oil, known to have disagreeable rancid-acid odor due to the presence of octanoic acid, was subjected to gamma (γ) irradiation for removal of the same. Sensory evaluation in tandem with electronic nose (Heracles and ENOVISION) analyses established that the oil irradiated at 4.2 kGy had no rancid-acid odor (absence of octanoic acid) at the end of 28 days of storage at 23 ± 2 °C. Therefore, 4.2 kGy was established as the “appropriate irradiation dose” for removal of rancid-acid odor of virgin coconut oil. A “deodorization index” with respect to rancid-acid odor was also developed using electronic nose for confirmation of this “appropriate dose” and thereby provided possible means of preventing detrimental effects of high dose of γ-irradiation in oils. GC–MS analysis of hydrocarbons extracted from oil irradiated at 4.2 kGy by supercritical carbon dioxide extraction revealed radiolysis of octanoic acid resulting in elimination of rancid-acid odor. Further, γ-irradiation at 4.2 kGy did not affect the antioxidant potency of coconut oil rendering it a safe and healthy alternative to conventional deodorization techniques for oils. Development of methodology for assessment of rancid-acid odor of coconut oil using electronic nose has been reported for the first time in this work.

In the food and beverage industries, there is a growing need for a systematic approach to maintain a certain standard of quality of the product. Quality of a product must ensure consumer expectations regarding the analytical and sensory quality of food and beverages. In order to achieve a guaranteed sustainable market success for beverage industries, control of quality standards throughout all stages of the manufacturing process (raw materials, processing, transport, and finished product) are to be maintained. Sensory quality evaluation is a practiced method for beverage industries where expert panelists ensure the quality by tasting the product which is highly subjective and non-repeatable also. Another method for quality estimation is analytical chemical testing process which is expensive and depends on special laboratory and operator. Electronic nose and electronic tongue systems are the sensor systems which can detect quality of beverages by detecting aroma and taste, respectively. This chapter is intended to explain the application of these sensors for quality evaluation of beverages. Over last two decades these electronic sensors have grown considerably to detect several parameters related to quality issues like age of sample, contamination level, impurities present if any, that is, overall instrumental quality evaluation are possible using these sensors which are repeatable also.

In black tea manufacturing, the fermentation process plays a significant role to determine the quality of finished tea. During this process, complex chain of biochemical reactions takes place and green tea leaves transform from grassy to sweet floral aroma. Monitoring and detection of optimum fermentation time are necessary, as both under and over fermented tea produce inferior grades of tea. This paper presents a method of real-time monitoring to detect the optimum fermentation time of black tea using an electronic nose consisting of eight quartz crystal microbalance (QCM) sensors. The sensor is coated with glucose derivative coating materials viz. maltose (MAL), maltodextrin (MDEX), β-cyclodextrin (β-CD), d -glucose ( d -GLU) and polyethylene glycols (PEG) with different molecular weights, 200, 1500, 4000, and 6000. Optimum fermentation times were determined for twelve black tea samples, and the results show good agreement with the estimations of the ultra-violet-visible (UV) spectrophotometer based reference method.

The present review evaluates the key modules of the electronic nose, a biomimetic system, with specific examples of applications to industrial emissions monitoring and measurement. Regulations concerning the odor control are becoming very strict, due to ever mounting environmental pollution and its subsequent consequences and it is advantageous to employ real time measurement system. In this perspective, systems like the electronic nose are an improved substitute for assessing the complex industrial emissions over other analytical techniques (odorant concentration measurement) and olfactometry (odor concentration measurement). Compared to tools like gas chromatography, electronic nose systems are easy to develop, are non-destructive and useful for both laboratory and on field purposes. Although there has been immense development of more sensitive and selective sensor arrays and advanced data mining techniques, there have been limited reports on the application of electronic nose for the measurement of industrial emissions. The current study sheds light on the practical applicability of electronic nose for the effective industrial odor and gaseous emissions measurement. The applications categorization is based on gaseous pollutants released from the industries. Calibration and calibration transfer methodologies have been discussed to enhance the applicability of electronic nose system. Further, industrial gas grab sampling technique is reviewed. Lastly, the electronic mucosa system, which has the ability to overcome the flaws of electronic nose system, has been examined. The review ends with the concluding remarks describing the pros and cons of artificial olfaction technique for the industrial applications.

A novel incremental algorithm based on fuzzy-c-means (FCM) method is proposed and implemented to effectively cluster data obtained from an electronic nose for black tea quality evaluation. The algorithm segregates data generated with the electronic nose from different batches of black tea into clusters with similar features, without requiring to access previously collected data. This feature of appending information exclusively from fresh data points entitles the algorithm to overcome catastrophic interference phenomenon common to conventional pattern recognition techniques.

Industrial odor concentration measurement in continuous mode is a challenging task using olfactometers, as it's expensive and requires human involvement for a prolonged time. This chapter presents the development of an indigenous metal oxide sensor-based electronic nose system for measurement of industrial odor in ou/m3. The results of electronic nose and field olfactometer were correlated using multilinear regression and partial least square regression techniques. The results showed satisfactory prediction by both the models, with RMSE (6.70, and 4.02), RAE (0.29 and 0.16), and NAE (0.89 and 0.96), respectively, for MLR and PLS. The results indicated better performance of PLS compared to MLR. The objective of the present work is to train and employ artificial olfaction system for continuous measurement of obnoxious emissions emitted from industries bypassing involvement of olfactometer.

The human perception process related to quality evaluation of food or beverages can be broadly divided into two processes – sensation and perception. While the process of sensation is responsible for collection of huge amount of data by the different sensory organs, the perception process interprets the data with a fusion process in the brain. In this paper, we describe a fusion model to combine the senses of smell and taste for quality assessment of black tea using two instruments – electronic nose and electronic tongue. We propose an artificial perception model based on multi sensor data fusion to analyze the sensory information for assessing tea quality and to correlate the same with human perception. Bayesian technique is employed for multi sensor data fusion and is tested on the combined data obtained using electronic nose and tongue. Experimental results show that the artificial perception improves when two sensory systems are fused together (Classification error 8%) compared with individual system (Classification error 30%).

Development of a methodology for rapid assessment of shelf-life of fried potato wedges were carried out using electronic noses (e-noses) with metal oxide gas sensors. Fuzzy logic analysis was applied for the first time for screening the sensors and it was found that four sensors were more specific for detecting volatile organic compounds from fried potato wedges. Data obtained from these screened sensors concluded that fried potato wedges had shelf-life of 3 days when stored in inert atmosphere of nitrogen in Ziploc pouches at 23 ± 2 °C. Mahalanobis distance method was adopted for quantifying the extent of spoilage and was correlated to peroxide values and free fatty acid content obtained by biochemical assays. Since, this methodology accurately and rapidly predicted the shelf-life of fried potato wedges; it could be also applied for fast and reliable estimation of shelf-life of various fried food products.

In many applications of electronic nose, the instrument is trained with data generated by human experts prior to its deployment in the fields. Quite often, these data are conflicting and inaccurate and thus the performance of an electronic nose is degraded. Moreover, degradation of its performance may also be due to the presence of redundant features or sensors in the array. While deploying an electronic nose for a specific application, it is observed that some of the sensors may not be required and only a subset of the sensor array contributes to the decision, which implies that optimization of the sensor array is also very important. To obtain a consistent and precise data set, both the conflicting data and irrelevant features must be removed. The rough set theory is capable of dealing with such an imprecise, inconsistent data set and in this paper, the rough-set based algorithm has been applied to remove the conflicting training patterns and optimize the sensor array in an electronic nose instrument used for sensing aroma of black tea samples.

The obnoxious odors generated from pulp and paper industries have been the cause of nuisance since the instigation of these industries. The objective of the study was to develop a metal oxide sensor based electronic nose for rapid measurement of odorant concentration and associated odor intensity of major reduced sulfur compounds emitted from different sources of these pulp and paper mills. The gas samples collected from the surroundings of major source points of industry were exposed to sensor array of the electronic nose and the change in voltage was measured and taken to PC through data acquisition cards. The same sets of samples were also tested with gas chromatography. The results of electronic nose and GC-FPD were correlated using response surface methodology to know the odorant concentration. The model fed with unknown industrial samples had more than 95% prediction capability. To determine odor intensity by electronic nose firstly a collective index was generated using SVD based 2-norm method (e-nose index) proportional to the sensors response relative to reference gas, methyl mercaptan. Secondly the e-nose index was associated with human expert evaluations. The training of the electronic nose enabled it to predict odorant concentration found at the industrial site and associated odor intensity in methyl mercaptan equivalent. The overall results of the experiments carried out suggest the potential of electronic nose as a device for on or off line measurement of odorant concentration and odor intensity.

Cross-perception is a perceptual phenomenon that demonstrates an interaction between two or more different sensory perception where the stimulus of one sensory system guides the instinctive response of another sense. In this work, a cross-perception model imitating multiple human’s perception by multi-sensor data fusion was proposed for the instrumental study. “Electronic nose” and “electronic tongue” were employed for detection of aroma and taste respectively of black tea samples. The data collected from two different sensory systems were pre-processed with suitable pre-processing technique and merged prior to further use. Two cross-perception variables i.e. cross correlation factor of aroma on taste and vice versa were assigned using principal component analysis and multiple linear regression. KNN classifier was comparatively used for classification of the conventional fusion model as well as cross-perception multi sensor fusion model. Results indicated that the cross-perception multi-sensors data fusion demonstrated noticeable superiority to the conventional fusion methodologies.

Automation for tea quality estimation is a challenging task and with the advent of electronic nose and electronic tongue systems this problem becomes quite addressable by instrumental means. Electronic nose judges tea sample based on aroma of the sample and based on taste tea quality can be classified using electronic tongue. For estimation of flavour of tea rather the overall quality of tea can be estimated if these two sensory responses can be fused. In this work, we have attempted to fuse these two sensory features using fuzzy fusion technique. A general fuzzy rule base is developed from the transient datasets obtained from electronic nose and tongue separately. The fuzzy system model can give accurate prediction with much simpler model than neural network. But both the system has certain advantages. In order to develop better classifier fuzzy neural network (FNN) model is also developed. Moreover the model works with transient responses and no data compression technique is employed. It is found that the combined sensor signature regarding tea quality estimation is quiet improved compared to individual sensor systems for all three classifiers and among these FNN is the best suited model for tea classification.

With the advent of electronic nose and electronic tongue, instrumental quality estimation of black tea has become possible. Electronic nose detects tea samples on the basis of aroma and taste can be detected by electronic tongue. But to develop human like quality judgment system, both the sensor responses must be considered. The objective of this work is to find optimized dataset by combining the sensor responses of e-nose and e-tongue so that the improved classification of tea can be done. Like individual sensor systems the merged sensor response should be preprocessed with suitable preprocessing method before feeding the data set to the classifier. In order to find the optimized dataset from both the systems, several multivariate preprocessing techniques are explored in this study. The quantitative measurements of preprocessing techniques applied to individual sensor system as well as fused dataset are estimated first with cluster separability measure (SI) and Dunn index (DI) calculation. Then each preprocessed dataset are tested with neural network based classifier. For getting optimized merged sensor response two methods of data merging have been attempted in this work. One by directly merging the raw sensor responses and then tested with preprocessing techniques and another is to find the optimized response from individual sensor systems and then to merge those optimized responses. The results show that the second method of fusion gives much better performance than direct merging of sensor responses.

In the tea industry, experienced tea tasters are employed for evaluation of tea quality and gradation of tea is done on the basis of their scores. This subjective method of assessment has numerous problems like inaccuracy and non-repeatability. Electronic nose and electronic tongue systems are recently being used for measurement of odor and taste of tea samples. As the senses of smell and taste are not independent, and both are interacting, the measured data from the individual systems are combined in this paper for improved estimation of black tea quality. It is found that for the combined system, both the clustering and classification rates improve when compared to the individual systems. With radial basis function neural network, the classification rate increases up to 93%, whereas with the independent systems, the classification rate obtained is 83–84% with electronic nose and 85–86% with electronic tongue.

Commonly used classification algorithms are not capable of incremental learning. When a new pattern is presented to such a computational model, it can either classify the unknown pattern based on its legacy training or declare the pattern as an outlier if such a provision is built into the associated algorithm. In the case of the pattern being an outlier to the existing training model, it is desirable that the same could be seamlessly included in the training model with appropriate class labels so that a universal computational model may be evolved incrementally. To this end, classifiers having the incremental-learning ability can be of great benefit by automatically including the newly presented patterns in the training data set without affecting class integrity of the previously trained system. In the present treatise, an incremental-learning fuzzy model for classification of black tea using electronic nose measurement is proposed. For application in black tea grade discrimination, an attempt has been made to correlate the multisensor aroma pattern of electronic nose with sensory panel (tea tasters) evaluation. However, this problem is associated with 2-D complexities. On one hand, the aroma of tea depends on the agroclimatic condition of a particular location, the specific season of flush, and the clonal variation for the tea plant. On the other hand, the sensory evaluation is completely human dependent that often suffers from subjectivity and nonrepeatability. In our pursuit of developing a universal computational model capable of objectively assigning tea-taster-like scores to tea samples under test, it has been felt that an incremental approach could be extremely beneficial for electronic-nose-based tea quality estimation. To this end, the proposed incremental-learning fuzzy model promises to be a versatile pattern classification algorithm for black tea grade discrimination using electronic nose. The algorithm has been tested in some tea gardens of northeast India, and encouraging results have been obtained.

Black Tea is conventionally tested by human sensory panel called “Tea Tasters”, who assign quality scores to different teas. Electronic nose-based studies aimed at co-relation of sensor array output signature with Tea Tasters’ scores have already been accomplished by the authors with good accuracy [N. Bhattacharyya, R. Bandyopadhyay, M. Bhuyan, B. Tudu, D. Ghosh, A. Jana, Electronic nose for black tea classification and correlation of measurements with “Tea Taster” marks, IEEE Trans. Instrum. Meas. 57 (2008) 1313–1321], where a number of conventional neural network topologies have been explored. One of the principal problems encountered in the above studies is collection of tea samples as tea industries in India are spread over dispersed locations and quality of tea varies considerably on agro-climatic condition, type of plantation, season of flush and method of manufacturing. As a result, the entire dataset is not available at a time and training the conventional neural network models becomes difficult. In this regard, classifiers having the incremental learning ability can be of great benefit by automatically including the newly presented patterns in the training dataset without affecting class integrity of the previously trained system. In the presented paper, the radial basis function (RBF) neural network with the incremental learning feature is used in the pattern recognition algorithm for black tea aroma classification with electronic nose.

Tea is an extensively consumed beverage worldwide with an expanding market. The major quality attributes of tea are flavor, aroma, color, and strength. Out of these, flavor and aroma are the most important attributes. Human experts called ldquotea tastersrdquo conventionally evaluate tea quality, and they usually assign scores to samples of tea that are under evaluation on a scale of 1 to 10, depending on the flavor, the aroma, and the taste of the sample. This paper presents a study where, first, the selection of appropriate sensors was carried out based on sensitivity with the major aroma-producing chemicals of black tea. Then, this sensor array was exposed to black tea samples that were collected from the tea gardens in India, and the computational model has been developed based on artificial neural network methods to correlate the measurements with the tea taster's scores. With unknown tea samples, encouraging results have been obtained with a more than 90% classification rate.

An advanced fuzzy approach of recognition of signal with time for tea classification with responses from electronic nose, electronic tongue and the combined sensor response of electronic nose and electronic tongue is attempted in this paper. In our model, neither priori choice of the suitable features (like peak, mean value, etc.) is considered, nor syntactic primitives as elementary components of signals are selected. Rather a new linguistic classification method named Fuzzy based Response of Signal with Time (FRST) is proposed. The novelty of the work is that instead of considering the complete signal or performing any statistical analysis on the complete signal, the sensor response is processed at the real time and the fuzzy partition is done to the sensor response spaces. At the same time, fuzzy partition is also applied on the time axis. Thus, we can assign an important role to each point of the signal depending on its position.

Quality assessment of tea is an extremely difficult task and until now has been carried out by experienced tasters. There are many important quality attributes of tea—aroma, taste, color, and size and, out of these attributes, aroma plays a significant role in determining the price of tea. Here we will focus only on this important attribute of tea—the aroma and the applicability of an electronic nose for its assessment. There have been several attempts for instrumental assessment of tea aroma, but being a natural product with several hundred compounds contributing to the pleasant aroma, there are several challenges to be overcome for the widespread use of such instruments. This chapter presents the research reports on experimentation with using electronic nose to determine tea aroma and a case study.

Black tea fermentation is essentially an oxidation process. After the plucked tea leaves are treated by series of processes called withering (removal of moisture by air flow), pre-conditioning and CTC (essentially maceration and cutting of leaves), the leaves are subjected to the process of fermentation by exposing them to air by laying the cut tea leaves on floor, trough or moving conveyor under controlled temperature, humidity and air-flow conditions. During this process, the leaves change colour from green to coppery brown and the grassy smell gets transformed to floral smell. It is critical that the leaves be allowed to ferment only up to the desired limit and both under and over fermentation result in deteriorated quality of black tea. Out of the two detectable parameters (colour and smell), smell is very important since a strong, very specific fragrance emanates from the leaves once leaves are optimally fermented. A new electronic nose-based approach for monitoring of tea aroma during fermentation is proposed. Two methods namely the 2-Norm method (2NM) and the Mahalanobis distance method (MDM) were tested and the results were correlated with the results of colorimetric tests and human expert evaluation.

Fermentation is an extremely crucial process in black tea manufacturing and is primarily responsible in deciding the final quality of finished tea. During this process, green tea changes its colour from green to coppery brown and grassy smell gets transformed to floral smell. A complex chain of biochemical reactions takes place during the fermentation process and once such changes reach their optimum point, the process should be stopped. The quality of the final product hinges to a large extent on the decision to end fermentation at right time, and either under-fermentation or over-fermentation leads to deterioration of finished tea quality. This paper presents a study on electronic nose-based monitoring of volatile emission pattern during black tea fermentation process and detection of the optimum fermentation time on the basis of peaks in the sensor outputs.

This article describes about an instrument and method for aroma based quality detection of Basmati and other aromatic rice varieties. It comprises few modules such as odour delivery module, sniffing module, water bath module and computing module. Odour handling module helps to deliver odour to the sensor array; a sniffing unit comprising a sensor array module that includes a eight number of metal oxide semiconductor sensors assembled on a printed circuit board, said printed circuit board fitted into a sensor chamber; a water bath module for preparing rice sample, said water bath module including a heater attachment to facilitate cooking; a computing module to quantify the aroma data acquired by sensors; data acquisition module etc. Principal Component Analysis (PCA) implemented for clustering the data sets acquired from sensor array. Also data generated from sensor array was fed to Probabilistic Neural Network (PNN), Back-propagation Multilayer Perceptron (BPMLP) and Linear Discriminant Analysis (LDA) for identification of different rice varieties. Finally, for aroma quantifying, pure-quadratic response surface methodology model used with mean square error (MSE) 0.0028.

The limitations of the classical pattern recognition algorithms may be addressed by an incremental way of learning, through which the existing knowledge base can be expanded from the information gathered solely from new set of samples. In this study, a novel incremental Self Organizing Map (i-SOM) algorithm is proposed and applied on the data generated from an electronic nose for black tea quality evaluation. The algorithm enables data with similar features (data points corresponding to different batches of black tea having similar aroma content) to be clustered together without the necessity of access to previously generated dataset.

The human perception process related to quality evaluation of food or beverages can be broadly divided into two processes -sensation and perception. While the process of sensation is responsible for collection of huge amount of data using the different sensory organs, the perception process interprets the data by a fusion process in the brain. In this paper, we describe a fusion model to combine the senses of smell and taste for quality assessment of black tea using two instruments - electronic nose and electronic tongue. As a first level of analysis, principal component of analysis is used for clustering and fuzzy ART artificial neural network for classification of the samples.

Rancidity is a major problem in food systems which occurs through lipid oxidation and significantly affects quality of foods. Owing to the drawbacks of conventional methods of estimating lipid oxidation in foods, alternative inexpensive technology such as electronic nose (e-nose) technology that can accurately and rapidly determine odor profile of foods, is gaining importance. First section of this chapter presents various applications of e-nose analysis of several food systems. Latter section of this chapter focusses on assessment of shelf-life of cookies formulated with un-encapsulated and encapsulated eugenol-rich clove extracts (obtained by supercritical carbon dioxide extraction from clove buds) as a source of natural antioxidant using ALPHA MOS e-nose system and a customized e-nose system. The e-nose results were further validated by conventional methods of rancidity determination. Shelf-life study for a storage period of 200 days revealed that normal cookies without any antioxidant have shelf-life of 100 days when stored in aluminium laminates at \(23 \pm 2^{\,\circ }\mathrm{C}\), while administration of encapsulated clove extract as antioxidant prevented rancidity in cookies for at least 200 days with a shelf-life lead of 100 days over control cookies. These observations were further affirmed by phytochemical analyses which indicated that the addition of encapsulated clove extract in cookies enhanced its nutraceutical potential and antioxidant activity for at least 200 days. Finally, linear regression equations were developed using which rancidity parameters of conventional biochemical assays of cookies can be directly derived from e-nose sensor responses.

Electronic nose (e-nose) is a machine olfaction system that has shown significant possibilities as an improved alternative of human taster as olfactory perceptions vary from person to person. In contrast, electronic noses also detect smells with their sensors, but in addition describe those using electronic signals. An efficient e-nose system should analyze and recognize these electronic signals accurately. For this it requires a robust pattern classifier that can perform well on unseen data. This research work shows the efficient prediction of black tea quality by means of modern kernel classifiers using the e-nose signatures. As kernel classifiers, this work investigates the potential of state of the art support vector machine (SVM) classifier and very recently developed nonparallel plane proximal classifier (NPPC) and vector-valued regularized kernel function approximation (VVRKFA) technique of multiclass data classification to build taster-specific computational models. Experimental results show that VVRKFA and one-versus-rest (OVR) SVM models offer high accuracies to predict the considerable variation in tea quality.

Electronic nose systems when deployed in network mesh can effectively provide a low budget and onsite solution for the industrial obnoxious gaseous measurement. For accurate and identical prediction capability by all the electronic nose systems, a reliable calibration transfer model needs to be implemented in order to overcome the inherent sensor array variability. In this work, robust regression (RR) is used for calibration transfer between two electronic nose systems using a Box–Behnken (BB) design. Out of the two electronic nose systems, one was trained using industrial gas samples by four artificial neural network models, for the measurement of obnoxious odours emitted from pulp and paper industries. The emissions constitute mainly of hydrogen sulphide (H 2 S), methyl mercaptan (MM), dimethyl sulphide (DMS) and dimethyl disulphide (DMDS) in different proportions. A Box–Behnken design consisting of 27 experiment sets based on synthetic gas combinations of H 2 S, MM, DMS and DMDS, were conducted for calibration transfer between two identical electronic nose systems. Identical sensors on both the systems were mapped and the prediction models developed using ANN were then transferred to the second system using BB–RR methodology. The results showed successful transmission of prediction models developed for one system to other system, with the mean absolute error between the actual and predicted concentration of analytes in mg L −1 after calibration transfer (on second system) being 0.076, 0.1801, 0.0329, 0.427 for DMS, DMDS, MM, H 2 S respectively.

Electronic nose has wide application in discriminations among food and beverage samples. Electronic nose is an array of gas sensor classifies samples based on their aroma profile. In this work this artificial sensory system is used to classify black tea using the features extracted from sensor response. Gaussian windowing function called `kernel' are used to extract information from the transient response and those are optimized by GA. The number of features being considered for classification was reduced considerably as well as classification performance is much improved than classification by directly using the sensor responses.

The responses generated by a gas sensor array are difficult to classify due to their inherent imprecision, uncertainty and the procedures of computational intelligence are appropriate to deal with such imperfect knowledge. In recent years, rough set theory has attracted more attention of many researchers even though it was proposed in the early 1980’s by Z. Pawlak. The rough set based analysis makes it very convenient for classification of data especially with huge volume of information, as the method is very efficient to find the optimal subset of attributes. In this paper, the rough set based algorithm has been applied to generate representative rules using the datasets obtained from a gas sensor array in an electronic nose instrument, capable of sensing aroma of black tea samples and these rules are used to classify the black tea quality.

Inconsistency in the electronic nose data set may appear due to noise that originates from various sources like electrical equipments, measuring instruments and some times the process itself. The presence of high noise leads to produce data that are of conflicting decision and thus encounters misleading or biased results. Also the performance of the electronic nose depends upon the number of relevant, irredundant features present in the data set. In an electronic nose the features correspond to the sensor array. While deploying an electronic nose for a specific application, it is observed that some of the features (sensors response) may not be required rather than only a subset of the sensor array contributes to the decision, which implies the optimization of sensor array is also important. To obtain a consistent precise data set both the conflicting data and irrelevant features must be removed. The rough set theory proposed by Z. Pawlak, is capable of dealing with such an imprecise, inconsistent data set and in this paper, the rough-set based algorithm has been applied to remove the conflicting training patterns and optimize the sensor array in an electronic nose instrument used for sensing aroma of black tea samples.

As of today, aroma of rice is measured by an expert sensory panel and they assign scores like ‘+’, ‘++’, ‘+++’ and ‘NA’ for mild, medium, strong and non aromatic varieties of rice respectively. This method of human panel testing is very subjective with numerous problems like inaccuracy, non-repeatability and it is laborious and time consuming also. On the other hand, the analytical instruments, which are used for this purpose are prohibitively expensive and are available in the laboratories only. It is in this pursuit, an electronic nose with an array of gas sensors has been developed for aroma measurement of rice. This user friendly and low cost electronic nose may be extremely useful for rice scientists, researchers and exporters to determine the aroma of aromatic rice. In this paper, we describe the experimental setup and the regression model for classification of rice samples. With unknown rice samples, aroma based classification accuracy by multi-sensor electronic nose using the regression model, has been found to be more than 80%.

Electronic nose (e-nose) is a machine olfaction system and the sensor array is an essential part of the electronic olfaction process. A pattern recognition unit is necessary in electronic nose system to efficiently decide about the output of the test using the responses of all the sensors in the array. The output of a pattern recognition algorithm depends on the quality of the feature set used for training and testing. Relevant and independent feature set improves the performance of a pattern classification algorithm. In some applications of electronic nose, the responses of few sensors are highly corrupted with noise and are either irrelevant or are redundant to the process. These sensors should be identified and eliminated from the sensor system for better accuracy. This paper addresses the selection of sensors in an e-nose system by different feature selection methods and then integrates them to achieve improved classification performance. We have used three types of feature selection methods namely, t-statistics, Fisher's criterion and minimum redundancy maximum relevance (MRMR) technique to select the most informative features. We have tested the proposed method on data obtained from the major aroma producing chemicals of black tea. Multi-class support vector machine (SVM) has been used as a pattern classifier in an electronic nose with black tea samples. The experimental results show that the performance of the e-nose system increased by 6–10% with the use of the proposed combinational feature selection technique.

Electronic nose (e-nose) is a machine olfaction system that has shown significant possibilities as an improved alternative of human taster as olfactory perceptions vary from person to person. In contrast, electronic noses also detect smells with their sensors, but in addition describe those using electronic signals. An efficient e-nose system should analyze and recognize these electronic signals accurately. For this it requires a robust pattern classifier that can perform well on unseen data. This research work shows the efficient prediction of black tea quality using machine learning algorithm with e-nose. This paper investigates the potential of three different types of multi-class support vector machine (SVM) to build taster-specific computational models. Experimental results show that all the three models offer more than 97% accuracies to predict the considerable variation in tea quality.

Quantitative evaluation of aroma of black tea is a difficult task as no single sensor has been developed which can quantify the volatile aroma compounds. As an alternative, the array of sensors has been attempted to solve this problem involving the principle of neural-network for analysis of responses. To develop the array we have used nano-structured ZnO doped with various metals. In this presentation we describe the responses of Geraniol, Trans-2-Hexenal, Linalool Oxide & Linalool, the major aroma chemicals of black tea to doped and undoped ZnO with Pt metal. Nanostructured ZnO doped with and without Pt are synthesized through chemical process. The synthesized materials are characterized by XRD and SEM which confirm their nanostructure and phase purity. The synthesized sensors are heated in the temperature range of 200°C to 400°C and their responses towards volatile biochemical compounds like Geraniol, Trans-2-Hexenal, Linalool Oxide and Linalool are studied.

Cardamom is known as “Queen of Spices”. It is one of the most highly priced spices in the world. The commercial part of the cardamom is the fruit (Capsule) of the plant that is used as a spice and a flavoring agent. The major quality measurement parameter of the cardamom is freshness, size, colour, aroma etc. Aroma is one of the main and crucial quality parameter for cardamom. The present practice of aroma quality estimation is done by GC, GC-MS, where different volatile oil and chemicals qualitative and quantitative tests are done. The present practice is laborious, time consuming and skilled manpower demanding process. In our present study an effort has been made to develop an Electronic Nose for rapid aroma determination of cardamom. Centres for Development of Advanced Computing, (C-DAC), Kolkata has indigenously developed the Electronic Nose (E-Nose) to estimate the quality of food and agro produces. Three-clone specific cardamom samples ware tested using this system as an exploratory study to determine the quality of cardamom and found the system is able to differentiate the samples. The principal component analysis shows distinct three clusters with principal component (PC1) 91.6% and PC2 6.8%. This paper demonstrates the quality estimation of cardamom by E-Nose.

An electronic nose with an array of quartz crystal microbalance (QCM) sensors has been developed to discriminate among different samples of Indian black tea varieties. The important volatile organic components (VOC) responsible for aroma of tea have been considered and the corresponding sensing materials have been identified. Five AT-cut 10MHz Quartz crystal blanks coated with different sensing materials have been used to differentiate the aroma of orthodox and CTC (cut-tearcurl) tea samples. The developed sensors can distinguish not only between the orthodox and CTC tea but distinct clusters are also obtained for the four different tea-samples, as visualized through principal component analysis (PCA). A radial basis function network (RBF) classifier has been used along with 10-fold cross validation technique for classification of data.