Sensing fermentation degree of cocoa (Theobroma cacao L.) beans by machine learning classification models based electronic nose system

Integration of electronic nose, electronic tongue, and colorimeter in combination with chemometrics for monitoring the fermentation process of Tremella fuciformis

This study proposes an efficient method for monitoring the submerged fermentation process of Tremella fuciformis (T. fuciformis) by integrating electronic nose (e-nose), electronic tongue (e-tongue), and colorimeter sensors using a data fusion strategy. Chemometrics was employed to establish qualitative identification and quantitative prediction models. The Pearson correlation analysis was applied to extract features from the e-nose and tongue sensor arrays. The optimal sensor arrays for monitoring the submerged fermentation process of T. fuciformis were obtained, and four different data fusion methods were developed by incorporating the colorimeter data features. To achieve qualitative identification, the physicochemical data and principal component analysis (PCA) results were utilized to determine three stages of the fermentation process. The fusion signal based on full features proved to be the optimal data fusion method, exhibiting the highest accuracy across different models. Notably, random forest (RF) was shown to be the most accurate pattern recognition method in this paper. For quantitative prediction, partial least squares regression (PLSR) and support vector regression (SVR) were employed to predict the sugar content and dry cell weight during fermentation. The best respective predictive R2 values for reducing sugar, tremella polysaccharide and dry cell weight were found to be 0.965, 0.988, and 0.970. Furthermore, due to its ability to capture nonlinear data relationships, SVR had superior performance in prediction modeling than PLSR. The results demonstrated that the combination of electronic sensor fusion signals and chemometrics provided a promising method for effectively monitoring T. fuciformis fermentation.

Technological tools for the measurement of sensory characteristics in food: A review

The use of technological tools, in the food industry, has allowed a quick and reliable identification and measurement of the sensory characteristics of food matrices is of great importance, since they emulate the functioning of the five senses (smell, taste, sight, touch, and hearing). Therefore, industry and academia have been conducting research focused on developing and using these instruments which is evidenced in various studies that have been reported in the scientific literature. In this review, several of these technological tools are documented, such as the e-nose, e-tongue, colorimeter, artificial vision systems, and instruments that allow texture measurement (texture analyzer, electromyography, others). These allow us to carry out processes of analysis, review, and evaluation of food to determine essential characteristics such as quality, composition, maturity, authenticity, and origin. The determination of these characteristics allows the standardization of food matrices, achieving the improvement of existing foods and encouraging the development of new products that satisfy the sensory experiences of the consumer, driving growth in the food sector. However, the tools discussed have some limitations such as acquisition cost, calibration and maintenance cost, and in some cases, they are designed to work with a specific food matrix.

Low-cost electronic-nose (LC-e-nose) systems for the evaluation of plantation and fruit crops: recent advances and future trends

An electronic nose (e-nose) is a device designed to recognize and classify odors. The equipment is built around a series of sensors that detect the presence of odors, especially volatile organic compounds (VOCs), and generate an electric signal (voltage), known as e-nose data, which contains chemical information. In the food business, the use of e-noses for analyses and quality control of fruits and plantation crops has increased in recent years. Their use is particularly relevant due to the lack of non-invasive and inexpensive methods to detect VOCs in crops. However, the majority of reports in the literature involve commercial e-noses, with only a few studies addressing low-cost e-nose (LC-e-nose) devices or providing a data-oriented description to assist researchers in choosing their setup and appropriate statistical methods to analyze crop data. Therefore, the objective of this study is to discuss the hardware of the two most common e-nose sensors: electrochemical (EC) sensors and metal oxide sensors (MOSs), as well as a critical review of the literature reporting MOS-based low-cost e-nose devices used for investigating plantations and fruit crops, including the main features of such devices. Miniaturization of equipment from lab-scale to portable and convenient gear, allowing producers to take it into the field, as shown in many appraised systems, is one of the future advancements in this area. By utilizing the low-cost designs provided in this review, researchers can develop their own devices based on practical demands such as quality control and compare results with those reported in the literature. Overall, this review thoroughly discusses the applications of low-cost e-noses based on MOSs for fruits, tea, and coffee, as well as the key features of their equipment (i.e., advantages and disadvantages) based on their technical parameters (i.e., electronic and physical parts). As a final remark, LC-e-nose technology deserves significant attention as it has the potential to be a valuable quality control tool for emerging countries.

Monitoring the baking quality of Tieguanyin via electronic nose combined with GC-MS

Roasting is extremely important for Tieguanyin oolong tea production because it strongly affects its chemical composition and sensory quality. In addition, there were significant differences in the preference for roasted tea among different people. However, the effect of roasting degree on the aroma characteristics and flavor quality of Tieguanyin tea is still unclear. To further study this, an electronic nose combined with gas chromatography-mass spectrometry (GC-MS) was used to monitor the baking process of Tieguanyin. The physicochemical indexes, sensory quality, and odor characteristics of the tea leaves subjected to different roasting conditions were measured. The increase in the roasting degree caused a decrease in the amount of taste substances such as tea polyphenols, catechins, and amino acids and a sharp increase in the phenol to ammonia ratio. Sensory evaluation results showed that moderate roasting could help improve the quality of the tea leaves. The results obtained using the electronic nose and GC-MS showed that there were substantial differences in the volatile substances, and 103 flavor compounds were highly correlated with the aroma characteristics of roasted tea with different roasting degrees. In addition, the electronic nose combined with various classification models could better distinguish tea leaves with different roasting degrees. Among them, the accuracy of the RF training set and prediction set reached>98.44%. The results of this study will aid in comprehensively monitoring the effects of the baking process on the flavor, chemical composition, and aroma of Tieguanyin as well as in distinguishing Tieguanyin tea leaves with different qualities.

SCANet: Implementation of Selective Context Adaptation Network in Smart Farming Applications

In the last decade, deep learning has enjoyed its spotlight as the game-changing addition to smart farming and precision agriculture. Such development has been predominantly observed in developed countries, while on the other hand, in developing countries most farmers especially ones with smallholder farms have not enjoyed such wide and deep adoption of this new technologies. In this paper we attempt to improve the image classification part of smart farming and precision agriculture. Agricultural commodities tend to possess certain textural details on their surfaces which we attempt to exploit. In this work, we propose a deep learning based approach called Selective Context Adaptation Network (SCANet). SCANet performs feature enhancement strategy by leveraging level-wise information and employing context selection mechanism. In exploiting contextual correlation feature of the crop images our proposed approach demonstrates the effectiveness of the context selection mechanism. Our proposed scheme achieves 88.72% accuracy and outperforms the existing approaches. Our model is evaluated on the cocoa bean dataset constructed from the real cocoa bean industry scene in Indonesia.

Metabolomics during the spontaneous fermentation in cocoa (Theobroma cacao L.): An exploraty review

Spontaneous fermentation is a process that depends on substrates' physical characteristics, crop variety, and postharvest practices; it induces variations in the metabolites that are responsible for the taste, aroma, and quality. Metabolomics makes it possible to detect key metabolites using chemometrics and makes it possible to establish patterns or identify biomarker behaviors under certain conditions at a given time. Therefore, sensitive and highly efficient analytical techniques allow for studying the metabolomic fingerprint changes during fermentation; which identify and quantify metabolites related to taste and aroma formation of an adequate processing time. This review shows that studying metabolomics in spontaneous fermentation permits the characterization of spontaneous fermentation in different stages. Also, it demonstrates the possibility of modulating the quality of cocoa by improving the spontaneous fermentation time (because of volatile aromatic compounds formation), thus standardizing the process to obtain attributes and quality that will later impact the chocolate quality.

A chemiresistive sensor array based on polyaniline nanocomposites and machine learning classification

The selective detection of ammonia (NH₃), nitrogen dioxide (NO₂), carbon oxides (CO₂ and CO), acetone ((CH₃)₂CO), and toluene (C₆H₅CH₃) is investigated by means of a gas sensor array based on polyaniline nanocomposites. The array composed by seven different conductive sensors with composite sensing layers are measured and analyzed using machine learning. Statistical tools, such as principal component analysis and linear discriminant analysis, are used as dimensionality reduction methods. Five different classification methods, namely k-nearest neighbors algorithm, support vector machine, random forest, decision tree classifier, and Gaussian process classification (GPC) are compared to evaluate the accuracy of target gas determination. We found the Gaussian process classification model trained on features extracted from the data by principal component analysis to be a highly accurate method reach to 99% of the classification of six different gases.

Application of Artificial Intelligence in Food Industry—a Guideline

Artificial intelligence (AI) has embodied the recent technology in the food industry over the past few decades due to the rising of food demands in line with the increasing of the world population. The capability of the said intelligent systems in various tasks such as food quality determination, control tools, classification of food, and prediction purposes has intensified their demand in the food industry. Therefore, this paper reviews those diverse applications in comparing their advantages, limitations, and formulations as a guideline for selecting the most appropriate methods in enhancing future AI- and food industry–related developments. Furthermore, the integration of this system with other devices such as electronic nose, electronic tongue, computer vision system, and near infrared spectroscopy (NIR) is also emphasized, all of which will benefit both the industry players and consumers.

Development of Gas Sensor Array for Methane Reforming Process Monitoring

The article presents a new method of monitoring and assessing the course of the dry methane reforming process with the use of a gas sensor array. Nine commercially available TGS chemical gas sensors were used to construct the array (seven metal oxide sensors and two electrochemical ones). Principal Component Regression (PCR) was used as a calibration method. The developed PCR models were used to determine the quantitative parameters of the methane reforming process: Inlet Molar Ratio (IMR) in the range 0.6-1.5, Outlet Molar Ratio (OMR) in the range 0.6-1.0, and Methane Conversion Level (MCL) in the range 80-95%. The tests were performed on model gas mixtures. The mean error in determining the IMR is 0.096 for the range of molar ratios 0.6-1.5. However, in the case of the process range (0.9-1.1), this error is 0.065, which is about 6.5% of the measured value. For the OMR, an average error of 0.008 was obtained (which gives about 0.8% of the measured value), while for the MCL, the average error was 0.8%. Obtained results are very promising. They show that the use of an array of non-selective chemical sensors together with an appropriately selected mathematical model can be used in the monitoring of commonly used industrial processes.