The aroma of honey is one of the important parameters in honey grading and that is depended on several factors, such as geographical origin, climate, botanical and environmental conditions. The aim of this study was the development and evaluation of an electronic nose as a new, fast and nondestructive method for detecting adulteration in honey. In this research, the ability of electronic nose as a non-destructive system for detecting honey adulteration with different percentages (pure, 20% syrup, 40% syrup, 60% syrup and 80% syrup) was investigated. The developed electronic nose consists of 8 metal oxide semiconductor sensors (MOS) to detect adultery in honey. After preprocessing the data obtained from the electronic nose the chemometric methods were utilized to classify different type of honey. Principle component analysis (PCA), hierarchical cluster analysis (HCA), linear discriminate analysis (LDA), were used to analyze the data obtained from electronic nose. Based on the results, the detection of adulteration was 98.4% of variance for PCA method, 99% accuracy for HCA method and 100% classification power by LDA method.

Aroma is one of the most important sensory properties of fruits and is particularly sensitive to the changes in fruit compounds. Gases involved in aroma of fruits are produced from the metabolic activities during ripening, harvest, post-harvest and storage stages. Therefore, the emitted aroma of fruits changes during the shelf-life period. The electronic nose (machine olfaction) would simulate the human sense of smell to identify and realize the complex aromas by using an array of chemical sensors. In this research, a low cost electronic nose based on six metal oxide semiconductor (MOS) sensors were designed, developed and implemented and its ability for monitoring changes in aroma fingerprint during ripening of banana was studied. The main components are used in the e-nose system include sampling system, an array of gas sensors, data acquisition system and an appropriate pattern recognition algorithm. Linear Discriminant Analysis (LDA) technique was used for classification of the extracted features of e-nose signals. Based on the results, the classification accuracy of 97/3% was obtained. Results showed the high ability of e-nose for distinguishing between the stages of ripening. It is concluded that the system can be considered as a nondestructive tool for quality control during banana shelf-life.

Honey is a sweet and viscous liquid made by bees from nectar of flowers. The emitted smell by honey depending on flower variety can be different. These factors led to use of machine olfactory system based on metal oxide sensors(MOS) in order to classify different floral origin honeys. Seven samples of different floral origins of honey with a total of 70 samples from each of 10 samples were tested. Principal component analyze (PCA), linear discriminant analyze (LDA) and artificial neural network (ANN) were methods to classify and analyze the extracted features from the machine olfactory system signal that were used. To classify floral origin honey using the machine olfaction, the results was included a 97% variance by PCA, 87.3% and 88.5% accuracy classification, respectively of LDA and ANN. As a conclusion, it was found that the electronic nose could provide good classification among of honeys.

Honey is a supersaturated sugar and viscose solution taken from the nectar of flowers, collected and modified by honeybees. Many producers of honey add some variety of sugars in honey that make difficulties with detection of adulterated and pure honey. Flavor is one of the most important parameters in the classification of honey samples and the smell emitted by the honey depending on the different flowers and constituents that could be different. This causes using an electronic nose system to detect honey adulteration.
Honey samples used in this study were lotus honey that was supplied from a market in Karaj city, Alborz province, Iran. Adulterated honey, along with percentages of fraud (by weight) of zero, 20, 35 and 50 percent, was prepared by mixing sugar syrup. Each group of samples, nine times were tested by the electronic nose system. The proposed system, consists of six metal oxide semiconductor sensors, sensor chamber, sample chamber, data acquisition systems, power supply, electric valves, and pumps. Electronic nose is planned for three-phase system baseline correction, the smell of sample injection and cleaning of the sensor and sample chambers with clean air (Oxygen). Responses of the sensors were collected and stored in 420 seconds by a data acquisition system and LabView ver 2012 software. We used fractional method in this study, in order to improve the quality of the information available and to optimize the array output before passing it on to the pattern recognition system. Linear Discriminant Analysis (LDA), Principal Component Analysis (PCA) and Artificial neural network (ANN) were the methods used for analyzing and recognizing pattern of electronic nose signals. Data processing was carried out using Microsoft Excel, neuralsolution 5 and Unscrambler X 10.3 (CAMO AS, Norway).
PCA PCA reduces the complexity of the data-set and is performed with no information on the classification of samples. It is based on the variance of the data-set. For PCA analysis, overall PC1 and PC2 explained 91% of the total variance among Lotus honey samples and the adulterations (PC1=80% and PC2=11%). The results indicate that it is clearly possible to recognize Lotus honey with adulterant using electronic nose systems.
LDA The LDA method for the detection of adulterated honey samples using leave-one-out validation was estimated. The method is most widely used as a method of classification that maximizes variance between the clusters and minimizes variance of within classes. By applying LDA on the collected data, 100% accurate classification for detecting of honey and their adulterations was obtained. It can also be concluded that this method could recognize adulterated honey samples properly.
ANN The back propagation multilayer perceptron algorithm was used to classify and to detect honey and adulterated types. Performance evaluations of each designed networks were compared by mean square error (MSE) and correlation coefficient (r).The data were divided to three subsets: 60% was used for training, 20% for testing and the remaining 20% were kept for cross validation.After network training and validation using optimized ANN model, i.e. 6-8-4 structure, success rate for 4 outputs (0, 20, 35 and 50% adulterated levels)were found to be 100%.After detecting adulteration, e-nose system accompanied with ANN can accurately classify honey from honey mixtures with fraud materials.
An electronic nose based on six metal oxide semiconductor sensors was used to detect adulterated honey samples. Electronic nose system can successfully classify between original honey and the adulterated one by pattern recognition method. The PCA, LDA and ANN techniques and analyzes of the electronic nose were very useful for evaluating the quality of the lotus honey. The results of these methods were used to classify the fraud in Lotus honey. However, there is a need to do more researches on the detection of adulteration in other agricultural and food products by electronic nose system.

Chicken nugget is a popular processed form of chicken meat. Nugget’s special flavor and color are interesting for consumers. Healthy meat is an important quality factor of nuggets. In the production process, the nugget is deeply fried for 1 minute at 185 °C. So, oil quality has an important effect on the final quality and healthy aspects. In this research, for the first time, the meat and oil quality used in chicken nuggets was checked using automatic and non-destructive techniques including an electronic nose (E-nose) and machine vision system. Meat health (healthy and spoiled) and consumed oil (healthy and burnt) were considered quality indicators. After frying, the excess oil was removed, and then the samples were subjected to machine vision and electronic nose systems for quality evaluations. In this research, the optimum of important parameters in the E-nose and the required times for sensing and removing the odor were determined. Also, the sensors' response was investigated. Principal component analysis (PCA), linear discriminant analysis (LDA) and artificial neural network (ANN) were used to analyze and classify the data. The accuracy results of both machine vision and electronic nose systems were similar and healthy meat and healthy oil classes were detected with up to 90 % accuracy. The evaluation results of the E-nose were similar to the machine vision system. Therefore, the combination of the two methods can be used in the measurement of color and smell characteristics.

Rosewater is one of the main products of rosa damascena and is a traditional long-lasting product in the Kashan region that has a global reputation. Regarding the use of rosewater in the treatment of rheumatic, cardiovascular, and also in the baking of different types of sweets and the preparation of ice creams, the main detection or adulteration of the produced rosewater is of particular importance. In this research, the ability to use an olfactory machine system (electronic nose) based on metal oxide semiconductor sensors as a non-destructive tool for detecting different levels of adulteration in rosewater and its authenticity assessment was studied. Principal component analysis (PCA), linear discriminant analysis (LDA), loading analysis, support vector machine (SVM) and decision tree (DT) were the methods used to achieve this goal. Based on the results, the PCA with the two main components of PC1 and PC2 described 92% of the variance of the data set for the used samples. In the sensor array, MQ4, TGS2620 and FIS sensors revealed the highest loading coefficient values and TGS822 and MQ8 sensors devoted the lowest ones. Based on the results of LDA method, the accuracy of the classification was 94%. By use of support vector machine with linear kernel function, in the C-SVM method, training and validation accuracy were obtained 98.75% and 87.5%, respectively. Also, the accuracy of the decision tree method in the classification of samples of rosewater was 80%. Based on the results, the olfactory machine system based on MOS sensors in combination with the pattern recognition methods has the ability to detect adulteration in rosewater and the LDA method has the highest classification accuracy. The research team also suggests using the system ability to detect adulteration in other products with potential for adulteration.

Today, human’s specific attention to food quality has led to the development of fast, easy and non-destructive methods to assess the quality of foodstuffs. Meat is one of the most important foods and its freshness is considered as the most important qualitative feature. Therefore, checking its quality for consumption has great value worthwhile. The main objective of the present study is to investigate the possibility of using electronic nose and artificial neural network methods for detecting the freshness of chicken meat during storage in a refrigerator at 4 ºC. In the used neural network system, the input layer consists of 10 neurons based on the number of sensors and the output layer includes 3 neurons related to classes of different freshness classes of chicken meat. Different classifier networks were designed and after investigation of different network structures; the best structure of the network was obtained with a hidden layer and 6 neurons in that layer. Finally, the optimal network with a general structure of 10-6-3 was created to detect the freshness of chicken meat during different days of storage. The used statistical indices to assess the classifier to evaluate the freshness of chicken meat including accuracy, precision, sensitivity, specificity and area under the curve factors. The values of these indices for classification using selected characteristics are 95.77, 94.7, 92.18, 95.95, and 94.1 respectively. Therefore, given that the main objective of the present study was to develop and implementation of an intelligent diagnosis system of chicken meat freshness using an electronic nose system. The acceptable obtained results of the present study indicate that the proposed applied system based on the electronic nose system and artificial neural networks methods as a smart and reliable method can online classification of chicken meat as fast, easy, economical, non-destructive and with appropriate accuracy.

Tomato is the second most important crop in the world, which is often consumed freshly. One of the most importance tomato products is paste that considered being the stuffing ingredient in Iranian foods. Therefore, the safety of tomato paste is so important. The aim of this study was diagnosis adulteration in tomato paste. For this purpose, olfactory machine system based on five gas sensors (TGS2600, TGS2620, MQ3, TGS880, TGS2610) was constructed and its potential was evaluated in determining the different levels of pumpkin adulteration in tomato paste (0, 5, 10, 15 and 20%). Pure tomato paste detection from adulteration samples was based on the samples smell in the head space and the receipt of smell by sensors. The principal component analysis (PCA), linear discriminant analysis (LDA), support vector machine (SVM) and partial least squares (PLS) were used to classify and analyze the extracted features of the sensor response. Results of PCA and PLS indicated that 99 and 94 percent of data variance was covered by two main components. TGS2610 and MQ-3 sensors and TGS880 sensor had the highest and lowest application in detection of tomato paste adulteration. The accuracy of the classification by the LDA method was 79.7%. The polynomial function with accuracy of 77.78% of the training and 76.66% validation in the C-SVM method and the radial base function with a precision of 98.8% of the training and 88.14% of Validation in the Nu-SVM method had the highest classification accuracy. In total, the olfactory machine system had acceptable performance in separation of different levels of adulteration.

As the most important source of calories and protein, bread has a special role and importance in the nutrition of the country, and its cheapness has caused it to replace other food items in the diet in recent years. The increase in bread consumption in the low-income and vulnerable groups has been more intense due to the low volume of supply of other food products and the excessive and continuous increase in the price of other alternative products. Flat bread has the highest consumption statistics among other breads in Iran. One of their types is Barbari bread, which is the second most consumed bread after lavash bread in Iran. Therefore, the health and quality of consumed Barbari bread is of particular importance. For this purpose, this study was conducted with the aim of the effect of storage time of Barbari bread based on the characteristic of smell using the olfactory machine system based on eight metal oxide semiconductor sensors.

An odor machine system based on eight MOS sensors was carried out in order to investigate the effect of bread storage time based on odor characteristics. Designed system includes data acquisition system, sensors, shield of sensors, sample container, power supply, connections, electric valves, air pump and air filter. The sensor array was consisted of the 8 MOS sensors that each one reacts to specific volatile compounds. These sensors are widely used in olfactory machines because of their high chemical stability, high durability, low response to moisture and affordable prices. These are the most commonly used sensors in electronic nose system. Sensors are the main components of an electronic nose system therefor it is necessary to select the able sensors to detect differences among samples. In order to carry out the test, the sample was placed in sample container and in the baseline correction step (150 seconds), clean air was passed through the sensors to transmit the response of sensor array to steady state. At the injection step (180 seconds), the sample headspace was transmitted and passed through sensors chamber. Output voltage of each sensor depends on the type of sensor and its sensitivity. At the cleaning step (150 seconds) the clean air was passed through sensors to get the sensor array responsive to a stable state. Also, at this step the pump removed the odor remaining inside the sample container and system is prepared for the next test. The signals obtained from the sensors were recorded and then pre-processed.

The olfactory machine system based on eight metal oxide semiconductor (MOS) sensors was investigated with the PCA pattern recognition method due to the storage time of Barbari bread at four different temperatures. The data obtained from the signals processing with fractional method were used as input of PCA. The results of principal component analysis with two components PC1 and PC2 are 95, 90, 86 and 85%, respectively, for Barbari bread that is stored at room temperature (in the table), refrigerator temperature (4°C), room temperature (in foil) And the temperature of the freezer was placed, it showed. The results obtained from the QDA analysis to determine the quality of Barbari bread at 4 °C for 9 days, at room temperature (in foil) and (on the table) for 5 days and Barbari bread at the freezer temperature of the refrigerator (-18 °C) for 15 days of storage with classification accuracy of 98.52, 96, 100 and 97.35% respectively. The results of LDA analysis for the signals obtained from the olfactory machine, in the classification of the duration of storage of Barbari bread at refrigerator temperature, room temperature (in the table), room temperature (in foil) and refrigerator freezer temperature, respectively, with classification accuracy of 79.26 and 85.33, 78.67 and 75.22 percent were obtained. Also, according to the output obtained from the loading linear graphs and the radar graph, the smell of Barbari bread has the most and the least effect on the MQ9 sensor and the TGS813 sensor, respectively

An olfactory machine system based on eight metal oxide semiconductor (MOS) sensors was investigated for the Barbari bread time retention effect at four different temperatures. The results of principal component analysis with two components PC1 and PC2, for Barbari bread at room temperature (the table), refrigerator temperature, room temperature (in foil) and It showed that they were exposed to freezing temperatures. The results obtained from QDA analysis to detect the quality of Barbari bread at 4°C in room temperature (in foil) and (in the table) and refrigerator freezer temperature respectively. The results of LDA analysis for the classification of Barbari bread at refrigerator temperature, room temperature (in the table), room temperature (in foil) and freezer temperature of the refrigerator. The was obtained. Also, according to the output obtained from the loading linear graphs and the radar graph, the smell of Barbari bread has the most and the least effect on the MQ9 sensor and the TGS813 sensor, respectively.