Qualitative Discrimination Research Articles

Study on the Identification and Detection of Walnut Quality Based on Terahertz Imaging.

Walnuts have rich nutritional value and are favored by the majority of consumers. As walnuts are shelled nuts, they are prone to suffer from defects such as mildew during storage. The fullness and mildew of the fruit impose effects on the quality of the walnuts. Therefore, it is of great economic significance to carry out a study on the rapid, non-destructive detection of walnut quality. Terahertz spectroscopy, with wavelengths between infrared and electromagnetic waves, has unique detection advantages. In this paper, the rapid and nondestructive detection of walnut mildew and fullness based on terahertz spectroscopy is carried out using the emerging terahertz transmission spectroscopy imaging technology. First, the normal walnuts and mildewed walnuts are identified and analyzed. At the same time, the image processing is carried out on the physical samples with different kernel sizes to calculate the fullness of the walnut kernels. The THz image of the walnuts is collected to extract the spectral information in different regions of interest. Four kinds of time domain signals in different regions of interest are extracted, and three qualitative discrimination models are established, including the support vector machine (SVM), random forest (RF), and k-nearest neighbor (KNN) algorithms. In addition, in order to realize the visual expression of walnut fullness, the terahertz images of the walnut are segmented with a binarization threshold, and the walnut fullness is calculated by the proportion of the shell and kernel pixels. In the frequency domain signal, the amplitude intensity from high to low is the mildew sample, walnut kernel, and walnut shell, and the distinction between walnut kernel, shell samples, and mildew samples is high. The overall identification accuracy of the aforementioned three models is 90.83%, 97.38%, and 97.87%, respectively. Among them, KNN has the best qualitative discrimination effect. In a single category, the recognition accuracy of the model for the walnut kernel, walnut shell, mildew sample, and reference group (background) reaches 94%, 100%, 97.43%, and 100%, respectively. The terahertz transmission images of the five categories of walnut samples with different kernel sizes are processed to visualize the detection of kernel fullness inside walnuts, and the errors are less than 5% compared to the actual fullness of walnuts. This study illustrates that terahertz spectroscopy detection can achieve the detection of walnut mildew, and terahertz imaging technology can realize the visual expression and fullness calculation of walnut kernels. Terahertz spectroscopy and imaging provides a non-destructive detection method for walnut quality, which can provide a reference for the quality detection of other dried nuts with shells, thus having significant practical value.

Olive Fruit Selection through AI Algorithms and RGB Imaging.

(1) Background: Extra virgin olive oil production is strictly influenced by the quality of fruits. The optical selection allows for obtaining high quality oils starting from batches with different qualitative characteristics. This study aims to test a CNN algorithm in order to assess its potential for olive classification into several quality classes for industrial purposes, specifically its potential integration and sorting performance evaluation. (2) Methods: The acquired samples were all subjected to visual analysis by a trained operator for the distinction of the products in five classes related to the state of external veraison and the presence of visible defects. The olive samples were placed at a regular distance and in a fixed position on a conveyor belt that moved at a constant speed of 1 cm/s. The images of the olives were taken every 15 s with a compact industrial RGB camera mounted on the main frame in aluminum to allow overlapping of the images, and to avoid loss of information. (3) Results: The modelling approaches used, all based on AI techniques, showed excellent results for both RGB datasets. (4) Conclusions: The presented approach regarding the qualitative discrimination of olive fruits shows its potential for both sorting machine performance evaluation and for future implementation on machines used for industrial sorting processes.

Morphological markers of ear print in Iraqi population

The outer ear is an emerging biometric that has drawn the attention of the research community for more than a decade. Empirical studies confirm, that the shape of the outer ear is unique, and it is also assumed to be stable throughout a human life time. An ear print is a two-dimensional depiction of the external ear pieces that have got into contact with a surface. In 1965, the first criminal ear print identification was made in Switzerland. The variation in shape and size of ear features makes it richer biometric tool than face. Material and method: The study performed a qualitative discrimination of the external ear shape of 143 individuals, including lobe size, lobe attachment, helix rolling, ear protrusion, and antihelix folding, by using ear photographs. Results: Six landmarks were evaluated for auricle morphometry in both sexes. There was variation in the morphological features of the external ear of the sampled subjects. The external ear features vary in both sexes. All variables were statistically significant (p ≤ 0.05). The triangle-shaped ear was prevalent in females and males had an oval shape. ear print

Front-face synchronous fluorescence spectroscopy: a rapid and non-destructive authentication method for Arabica coffee adulterated with maize and soybean flours.

This article describes a novel front-face synchronous fluorescence spectroscopy (FFSFS) method for the fast and non-invasive authentication of ground roasted Arabica coffee adulterated with roasted maize and soybean flours. The detection was based on the different composition of fluorescent Maillard reaction products and caffeine in roasted coffee and cereal flours. For each roasted maize or soybean adulterant flour (5–40 wt%), principal component analysis coupled with linear discriminant analysis (PCA–LDA) was used for qualitative discrimination. Quantitative prediction models were constructed based on the combination of unfolded total synchronous fluorescence spectra and partial least square regression (PLSR), followed by fivefold cross-validation and external validation. The PLSR models produced suitable results, with the determination coefficient of prediction (Rp2) > 0.9, root mean square error of prediction (RMSEP) < 5%, relative error of prediction (REP) < 25% and residual predictive deviation (RPD) > 3. The limits of detection (LOD) were both 10% for roasted maize and soybean flours. Most relative errors for the prediction of simulated blind samples were between -30% and + 30%. The benefits of this strategy are simplicity, rapidity, and non-destructive detection. However, owing to the high similarity between roasted coffee and roasted cereal flours and the influence of the roasting degree on fluorescent Maillard reaction products, its application is limited to the preliminary screening of roasted coffee with the same roasting degree, adulterated with relatively large amounts of roasted cereal flours which are roasted to analogous color to the coffee.Supplementary InformationThe online version contains supplementary material available at 10.1007/s00003-022-01396-8.

A Variable Selection Method Based on Fast Nondominated Sorting Genetic Algorithm for Qualitative Discrimination of Near Infrared Spectroscopy

A reliable and effective qualitative near-infrared (NIR) spectroscopy discrimination method is critical for excellent model building, yet the performance of models built by these methods is highly dependent on valid feature extraction. The goal of feature selection is to associate the selected variables with the property of interest, which many have done successfully. However, many of selection methods focus only on strong association with the analytes or properties of interest, neglecting correlations between variables. A variable selection method based on a fast nondominated-ranking genetic algorithm (NSGA-II) was proposed in this paper for qualitative discrimination of NIR spectra. The method had two objective functions: (1) maximizing the sum of ratios of interclass variance to intraclass variance, (2) minimizing the sum of correlation coefficients between the selected variables. FT-NIR spectra of a total of 124 tobacco samples from different origins and parts in Guizhou Province, China, were used as the experimental objects, and the part-grade discrimination models of tobacco leaves were established by combining this method with partial least squares-based discriminant analysis (PLS-DA), and compared with PLS-DA model based on the full spectrum. The results showed that the performance of PLS-DA model with the NSGA-II was improved, with a comparable or better correct discrimination rate and reasonable discrimination rate, and could discriminate different parts of the tobacco leaves well. It indicates that the NSGA-II can select a few and effective feature variables to build a high-performance qualitative discrimination model and is proved to be a promising algorithm. In addition, the method is not designed exclusively for spectral data. It is a general strategy that could be used for variable selection for other types of data.

Detection of adulteration in mutton using digital images in time domain combined with deep learning algorithm

A novel method based on digital images in time domain combined with convolutional neural network (CNN) is proposed for discrimination and analysis of the adulterated mutton. For this, 195 sample images during the constant temperature heating process (about 10 min) were combined with CNN for qualitative discrimination and quantitative prediction of adulterated mutton. Furthermore, the hypothesis that temperature disturbance can improve the detection ability of adulterated mutton was confirmed by comparing the model performance of the initial heating stage and the entire heating process. The experimental results show that the performance of the latter was superior to that of the former. The accuracy of the qualitative discriminant model was increased by 7.33%, the R2 and RPD of the quantitative prediction model of the duck/pork in adulterated mutton were increased by 0.08/0.07 and 0.85/0.87 respectively, while the RMSE decreased by 0.01/0.01. Consequently, the proposed method can be used for detecting adulterated mutton effectively and accurately.

Qualitative Analysis of Glass Microfragments Using the Combination of Laser-Induced Breakdown Spectroscopy and Refractive Index Data.

We have successfully demonstrated that although there are significant analytical challenges involved in the qualitative discrimination analysis of sub-mm sized (microfragment) glass samples, the task can be solved with very good accuracy and reliability with the multivariate chemometric evaluation of laser-induced breakdown spectroscopy (LIBS) data or in combination with pre-screening based on refractive index (RI) data. In total, 127 glass samples of four types (fused silica, flint, borosilicate and soda–lime) were involved in the tests. Four multivariate chemometric data evaluation methods (linear discrimination analysis, quadratic discrimination analysis, classification tree and random forest) for LIBS data were evaluated with and without data compression (principal component analysis). Classification tree and random forest methods were found to give the most consistent and most accurate results, with classifications/identifications correct in 92 to 99% of the cases for soda–lime glasses. The developed methods can be used in forensic analysis.

Use of the downscalling method in satellite images for the analysis of heat islands in landfills

Currently, the landfill is the most used method in the world for the treatment and final disposal of municipal solid waste (MSW), especially in first world countries. In Brazil, in 2017, only 36.8% of the municipalities used landfills for the final disposal of MSW, but this number is expected to increase significantly in the next decade. In this way, thermal infrared remote sensing (TIR) is a powerful tool for observation and monitoring the landfill. With the TIR, it is possible to evaluate various processes within the landfill, mainly those linked to biological processes that generate heat as a result. In this way, methods of obtaining the surface temperature of the landfill combined with methods of improving spatial data resolution show great potential for analyzing these data. Thus, this work uses the TIR applied to the study of environmental sanitation problems in the landfill in the city of São Leopoldo, state of Rio Grande do Sul – Brazil. The aim of this work is study the heat islands landfill using LANDSAT-8/SENTINEL-2 thermal images and consolidated downscaling methods to increase the spatial resolution of the temperature image. The results showed an increase of up to 300% in the spatial resolution of the data, allowing better qualitative discrimination of the targets, and thus identification of heat islands without increasing the root mean square error (RMSE) in temperature. In addition, the error associated with the temperature for the resampled data was less than or equal to the error associated with the original data.

Extracting and Modifying the Vibration Characteristic Parameters of Watermelon Based on Experimental Modal Measurement and Finite Element Analysis for Hollow Heart Defect Detection

HighlightsAn impulse vibration method is proposed to excite watermelon for hollow heart defect detection.Experimental models of watermelon were acquired with 3D scanning laser vibrometry.The relationship between hollow heart defect and vibration characteristic parameters was investigated with finite element analysis.Better prediction of hollow heart defect in watermelon was achieved with the wavelet transform method.Abstract. Hollow heart defect seriously influences the taste and storability of watermelon. In this study, a non-destructive detection system based on an impulse vibration method was developed to detect hollow watermelon. First, acceptable agreement between the theoretical and experimental models of watermelon proved the suitability of investigating the relationship between hollow heart defect and vibration characteristic parameters by finite element analysis (FEA). Through modal analysis, the optimum location for the detection sensor was determined at the opposite location or 90° from the excitation point. The normalized second to fourth resonance frequencies (f2n, f3n, and f4n) and the peak value at the second frequency (A2) were extracted as latent variables for prediction of hollow watermelon. The technical parameters of the pressurized-air excitation device were then modified in orthogonal tests, and the best combination of technical parameters was as follows: air pressure of 275 kPa, excitation distance of 9 cm, and pulse width of 200 ms. In the qualitative discrimination of hollow watermelon, the results showed that a back-propagation neural network (BPNN) using 13 vibration characteristic parameters had the best classification performance, with accuracies of 91.7% and 88.9% for the calibration and prediction sets. In the quantitative analysis of hollow rate, the best prediction result was achieved with the BPNN (rp = 0.829, RMSEP = 0.016), which selected ten vibration characteristic parameters as input variables. Therefore, it is feasible to detect hollow watermelon by impulse vibration, and this method has potential to be applied in on-line defect detection. Keywords: Doppler vibrometry, Finite element analysis, Hollow heart defect, Laser modal analysis, Watermelon.

Qualitative Discrimination of Intact Tobacco Leaves Based on Near-Infrared Technology

To explore the application of near-infrared (NIR) technology to the quality analysis of raw intact tobacco leaves, a nondestructive discrimination method based on NIR spectroscopy is proposed. A “multiregion + multipoint” NIR spectrum acquisition method is developed, allowing 18 NIR diffuse reflectance spectra to be collected from an intact tobacco leaf. The spectral characteristics and spectral preprocessing methods of intact tobacco leaves are analyzed, and then different spectra (independent or average spectra) and different algorithms (discriminant partial least-squares (DPLS) and Fisher’s discriminant algorithms) are used to construct discriminant models for verifying the feasibility of intact leaf modeling and determining the optimal model conditions. Qualitative discrimination models based on the position, green-variegated (GV), and the grade of intact tobacco leaves are then constructed using the NIR spectra. In the application and verification stage, a multiclassification voting mechanism is used to fuse the results of multiple spectra from a single tobacco leaf to obtain the final discrimination result for that leaf. The results show that the position-GV discrimination model constructed using independent spectra and the DPLS algorithm and the grade discrimination model constructed using independent spectra and Fisher’s algorithm achieve optimal results with intact leaf NIR wavenumbers from 5006–8988 cm−1 and the first-derivative and standard normal variate transformation preprocessing method. Finally, when applied to new tobacco leaves, the position-GV model and the grade model achieve discrimination accuracies of 95.18% and 92.77%, respectively. This demonstrates that the two models have satisfactory qualitative discrimination ability for intact tobacco leaves. This study has established a feasible method for the nondestructive qualitative discrimination of the position, GV, and grade of intact tobacco leaves based on NIR technology.

Rapid, simultaneous and non-destructive determination of maize flour and soybean flour adulterated in quinoa flour by front-face synchronous fluorescence spectroscopy

Quinoa is a naturally gluten-free pseudocereal receiving growing attentions worldwide. Herein, front-face synchronous fluorescence spectroscopy (FFSFS) was applied for the fast and non-invasive authentication of quinoa flour adulteration with maize and soybean flours. The cases of both single and dual adulterants (6–50 wt%) were investigated. For single adulteration with either maize or soybean flour, principal component analysis coupled with linear discriminant analysis (PCA–LDA) was used for preliminary qualitative discrimination. Then first-order prediction models were constructed based on the combination of unfolded total synchronous fluorescence spectra and partial least square (PLS) regression, and were validated by five-fold cross-validation and external validation. For dual adulterants, a general PLS2 model for the simultaneous determination of maize and soybean flours was developed, which produced suitable results, with the determination coefficient of prediction (Rp2) > 0.9, root mean square error of prediction (RMSEP) < 5% and residual predictive deviation (RPD) > 3. The limits of detection (LODs) are 9.0% and 6.9% for maize flour and soybean flour, respectively. Furthermore, most spike recoveries are in the range of 80–120%. This study not only has important applications in the authentication of quinoa flour adulteration but may also shed light on the quantitative component analysis of other cereal flour mixtures from different botanical origins.

A novel analytical strategy for discriminating antibiotic mycelial residue adulteration in feed based on ATR-IR and microscopic infrared imaging

The Antibiotic mycelial residue (AMR) contains antibiotic residue, there are safety risks if it is used illegally in feed. This study investigated the feasibility of qualitative identification of AMR in protein feed and self-prepared feed based on attenuated total reflection mid-infrared spectrum (ATR-IR) and microscopic infrared imaging. Cottonseed meal (CM), soybean meal (SM), distillers dried grains with solubles (DDGS), nucleotide residue (NR), oxytetracycline residue (OR) and streptomycin sulfate residue (SR) and two self-prepared feed (broiler and pig) were used as research objects. The results showed that there were characteristic peaks at 1614 cm−1, 1315 cm−1, 779 cm−1, 514 cm−1 in the ATR-IR spectra of AMR, which were related to calcium oxalate hydrate. After detection, the content of total calcium and calcium oxalate in AMR were higher than those in protein feed. ATR-IR can quickly realize the qualitative discrimination of pure material samples. The combination of ATR-IR and partial least squares discriminant analysis (PLSDA) was effective in discriminating AMR from CM and SM with a single component (the classification errors were 0), but it cannot meet the discrimination of AMR from the fermented protein feed (such as DDGS and NR, the classification errors were 0.10 and 0.12) and self-prepared feed with complex components. Compared with ATR-IR, microscopic infrared imaging was less affected by the sample complexity. Multi-component samples belong to physical mixing and will not affect the infrared spectra of each component. Therefore, microscopic infrared imaging combined with effective information extraction algorithms such as cosine similarity can distinguish OR in the fermented protein feed and self-prepared feed. The above results showed that the advantages of ATR-IR and microscopic infrared imaging were complementary, which provided a new idea for the discrimination analysis of illegal feed additives.

Identification of aflatoxin B1 in peanut using near-infrared spectroscopy combined with naive Bayes classifier

Peanuts are easily contaminated by a variety of mycotoxins during growth, transportation, and storage, of which aflatoxin B1 is the most common. Aflatoxin B1 is one of the most toxic carcinogens known, and it can cause liver damage to varying degrees after ingestion. To explore the feasibility of detecting aflatoxin B1 contamination in peanuts by near-infrared spectroscopy, 115 peanut samples with aflatoxin B1 content in the range of 2.44 to 223.76 μg/kg were prepared. The near-infrared spectroscopy data of the peanut sample in the 940–1660 nm band was obtained, and the naive Bayes qualitative discrimination model based on the whole band was established. To improve accuracy and reduce dimensions, simplified models were built using characteristic wavelengths screened by Successive Projection Algorithm and Elimination of Uninformative Variables. The built model is verified internally and externally by omitting cross-validation. In comparison, the second derivative Savitzky–Golay Elimination of Uninformative Variables normal kernel density estimation naive Bayes model reached the optimum discriminant accuracy, with the comprehensive overall accuracy of validation set and prediction set were over 91.00%, areas under receiver operating characteristic curves was over 0.90. The results showed that the accuracy of the quantitative determination of aflatoxin B1 in peanuts by near-infrared spectroscopy was high, with a boundary of 20 μg/kg. This method is suitable for the qualitative determination of peanut aflatoxin B1.

Spatial distribution of total polyphenols in multi-type of tea using near-infrared hyperspectral imaging

Polyphenols are the key taste and health components of tea. Typically, chemical analysis is used to determine the total polyphenol (TP) content of tea. However, this process is time consuming. In this study, the TP content of six tea types, namely green, white, yellow, oolong, black, and dark, was assessed using near-infrared hyperspectral imaging. Here, 100% accuracy was achieved for both the calibration and prediction sets for qualitative discrimination of tea by using the principal component analysis-K-nearest neighbor model. Important wavelengths of TP were selected using the regression coefficients (RCs) of the partial least squares regression (PLSR) model. The proposed RC-PLSR model yielded satisfactory prediction results with a residual predictive deviation of 3.34. The differences in the spatial distribution of TP in various tea samples were visualized using distribution maps. This study provided a rapid and nondestructive method for the identification of tea types and the visualization of the TP content of tea.

Rapid quantitative detection of glucose using biological sensor system as combined with electrochemical data treatment

ABSTRACT In this paper, glucose quantitative detection by biological sensor and electrochemical capacitance data treatment was studied. The taste cells were fixed on carbon screen-printed electrode. Electrochemical capacitance responses of cell-based sensor to sucrose in different concentrations and related mixtures were measured. Glucose quantitative analysis results demonstrated that the proposed method had quantitative determination abilities toward glucose and related mixtures. The signal-to-noise ratio index presented qualitative discrimination abilities to different glucose mixtures. Negative control results indicated that the receptor-free cells showed no perception ability for glucose. The proposed method is promising in sugar rapid detection.

Heavy metal Hg stress detection in tobacco plant using hyperspectral sensing and data-driven machine learning methods.

Accurate detection of heavy metal stress on the growth status of plants is of great concern for agricultural production and management, food security, and ecological environment. A proximal hyperspectral imaging (HSI) system covered the visible/near-infrared (Vis/NIR) region of 400-1000nm coupled with machine learning methods were employed to discriminate the tobacco plants stressed by different concentration of heavy metal Hg. After acquiring hyperspectral images of tobacco plants stressed by heavy metal Hg with concentration solutions of 0mg·L-1 (non-stressed groups), 1, 3, and 5mg·L-1 (3 stressed groups), regions of interest (ROIs) of canopy in tobacco plants were identified for spectra processing. Meanwhile, tobacco plant's appearance and microstructure of mesophyll tissue in tobacco leaves were analyzed. After that, clustering effects of the non-stressed and stressed groups were revealed by score plots and score images calculated by principal component analysis (PCA). Then, loadings of PCA and competitive adaptive reweighted sampling (CARS) algorithm were employed to pick effective wavelengths (EWs) for discriminating non-stressed and stressed samples. Partial least squares discriminant analysis (PLS-DA) and least-squares support vector machine (LS-SVM) were utilized to estimate the stressed tobacco plants status with different concentrations Hg solutions. The performances of those models were evaluated using confusion matrixes (CMes) and receiver operating characteristics (ROC) curves. Results demonstrated that PLS-DA models failed to offer relatively good result, and this algorithm was abandoned to classify the stressed and non-stressed groups of tobacco plants. Compared to LS-SVM model based on full spectra (FS-LS-SVM), the LS-SVM model established EWs selected by CARS (CARS-LS-SVM) carried 13 variables provided an accuracy of 100%, which was promising to achieve the qualitative discrimination of the non-stressed and stressed tobacco plants. Meanwhile, for revealing the discrepancy between 3 stressed groups of tobacco plants, the other FS-LS-SVM, PCA-LS-SVM, and CARS-LS-SVM models were setup and offered relatively low accuracies of 55.56%, 51.11% and 66.67%, respectively. Performance of those 3 LS-SVM discriminative models was also poorly performing to differentiate 3 stressed groups of tobacco plants, which might be caused by low concentration of heavy metal and similar canopy (especially in fresh leaves) of plant. The achievements of the research indicated that HSI coupled with machine learning methods had a powerful potential to discriminate tobacco plant stressed by heavy metal Hg.

Citizen Science for Students: Fundamentals of Gas Sensing and Realization of Environmental Studies

Introduction Citizen Science describes the collaboration between citizens and scientists on scientific issues and has gained in importance in recent years [1]. The development of the topic Citizen Science shows that citizens should no longer act as mere data collectors, but are more and more integrated into the scientific work. This includes the imparting of background knowledge, the introduction to scientific methods as well as joint data evaluation and interpretation. Through this approach, science is made accessible, and citizens often draw more conclusions regarding their own actions, behaviors and habits [2]. Especially young people are developing ever greater environmental awareness due to the already noticeable effects of climate change. Current topics in the media about Fridays for Future, but also exhaust emissions of cars (“Diesel scandal”) and increasing concentrations of particulate matter in inner cities show that air quality is becoming more and more important for them [3].In the project “SUSmobil” students aged between 12 and 18 receive fundamental information on gas sensing, consisting of functionality, calibration process and data evaluation, and then have the opportunity to develop their own environmental studies, such as an “air pollution map” or “the investigation of the air composition in beehives”. The project is divided into four modules, each of which focuses on one aspect of gas sensing and air quality. The concept is presented below. Fundamentals of gas sensing Beginning with the theoretical fundamentals on how semiconductor gas sensors work, the first module describes the functionality of metal oxide semiconductor (MOS) gas sensors. By examining the sensor reaction to the headspace of different substances, water, apple juice and non-alcoholic beer, the students learn about the relevant processes on the sensor surface. The second module offers an insight into modern methods of calibration of semiconductor gas sensors. In contrast to the first module, which was dealing with qualitative discrimination of substances, the focus here is laid on the quantification of one substance. The sensor model previously developed in module 1 is extended by the temperature cycled operation (TCO), which can be used to improve sensitivity, selectivity and stability of the sensor [4]. By addressing the topics of data analysis, feature extraction and mathematical modeling using neural networks, a comprehensive picture of the scientific work with air quality sensors is provided.The third module addresses environmental measurements and relevant parameters, e.g. particulate matter (PM), volatile organic compounds (VOCs) and CO2. Students learn about different sensor principles such as non-dispersive infrared absorption (NDIR) and laser scattering. Threshold limit values and possible sources of harmful VOCs are described and tested with model substances. Example of an environmental study: Air pollution map Motivated students are given the opportunity to create their own environmental studies using a citizen science approach supported by scientists. One example for an environmental study is the creation of an “air pollution map”, which two students (age 17 and 18) developed as part of the German youth research competition "Jugend forscht” (figure 1). The aim of this project was to perform air quality measurements inside and outside of the city and visualize the data using a freely accessible web page. The students designed a portable 3D-printed sensor system and integrated the air quality sensor “BME680” from Bosch (figure 2). The integrated gas sensor measures temperature, air pressure, humidity and a so called “air quality index” (AQI) derived from the raw signals of a MOS sensor and humidity. This index with a range between 0 (“Good”) and >300 (“Hazardous”) is not an official index, since the sensor was not calibrated based on an excepted test standard. Yet this value gives a quick indication of air quality. The system is controlled by the internet-enabled microcontroller “ESP32”. Via an internet hotspot, provided by a smartphone, the collected data, including GPS information of the smartphone, is sent to the online database of the free to use IoT platform “Blynk”. The corresponding smartphone app allows visualization of the data in real time (figure 3). Furthermore the students created a website, which collects the data from the online server and displays the values on a “Google Maps” based heatmap (figure 4). The color of the marked areas corresponds to the air quality index. Outlook The students were able to create a basic setup to measure air quality with a mobile sensor system. This can be adapted and further extended in subsequent studies by other students, for example using other sensors and including rigorous calibration and testing. Acknowledgement The project is funded by the environmental foundation DBU (“Deutsche Bundesstiftung Umwelt”).

Qualitative discrimination of Chinese dianhong black tea grades based on a handheld spectroscopy system coupled with chemometrics.

The evaluation of Chinese dianhong black tea (CDBT) grades was an important indicator to ensure its quality. A handheld spectroscopy system combined with chemometrics was utilized to assess CDBT from eight grades. Both variables selection methods, namely genetic algorithm (GA) and successive projections algorithm (SPA), were employed to acquire the feature variables of each sample spectrum. A partial least‐squares discriminant analysis (PLS‐DA) and support vector machine (SVM) algorithms were applied for the establishment of the grading discrimination models based on near‐infrared spectroscopy (NIRS). Comparisons of the portable and benchtop NIRS systems were implemented to obtain the optimal discriminant models. Experimental results showed that GA‐SVM models by the handheld sensors yielded the best predictive performance with the correct discriminant rate (CDR) of 98.75% and 100% in the training set and prediction set, respectively. This study demonstrated that the handheld system combined with a suitable chemometric and feature information selection method could successfully be used for the rapid and efficient discrimination of CDBT rankings. It was promising to establish a specific economical portable NIRS sensor for in situ quality assurance of CDBT grades.

Authentication of paprika using HPLC-UV fingerprints

In this work we combine simple extraction and HPLC-UV methodologies with chemometric pattern-recognition strategies in order to obtain characteristic fingerprints of phenolic compounds that allow the authentication of paprika samples. To illustrate the potential of the proposed approach, two different adulteration scenarios were considered, namely adulteration of paprika based on its type (sweet, bittersweet and spicy) as well as on its region (Murcia, la Vera and Czech Republic). Upon preparation of a proper set of samples, those were analysed using a C18 reversed-phase column and registered chromatograms were then compressed employing fast Fourier transform (FFT) to reduce the large dimensionality of the data set, while preserving all relevant features. Next, data were analysed using linear discriminant analysis (LDA) for the qualitative discrimination of adulterated samples, or by partial least-squares regression (PLS) modelling to quantitatively assess the adulteration degree.

Determination of the bruise degree for cherry using Vis-NIR reflection spectroscopy coupled with multivariate analysis.

Determination and classification of the bruise degree for cherry can improve consumer satisfaction with cherry quality and enhance the industry’s competiveness and profitability. In this study, visible and near infrared (Vis-NIR) reflection spectroscopy was used for identifying bruise degree of cherry in 350–2500 nm. Sampling spectral data were extracted from normal, slight and severe bruise samples. Principal component analysis (PCA) was implemented to determine the first few principal components (PCs) for cluster analysis among samples. Optimal wavelengths were selected by loadings of PCs from PCA and successive projection algorithm (SPA) method, respectively. Afterwards, these optimal wavelengths were empolyed to establish the classification models as inputs of least square-support vector machine (LS-SVM). Better performance for qualitative discrimination of the bruise degree for cherry was emerged in LS-SVM model based on five optimal wavelengths (603, 633, 679, 1083, and 1803 nm) selected directly by SPA, which showed acceptable results with the classification accuracy of 93.3%. Confusion matrix illustrated misclassification generally occurred in normal and slight bruise samples. Furthermore, the latent relation between spectral property of cherries in varying bruise degree and its firmness and soluble solids content (SSC) was analyzed. The result showed both colour, firmness and SSC were consistent with the Vis-NIR reflectance of cherries. Overall, this study revealed that Vis-NIR reflection spectroscopy integrated with multivariate analysis can be used as a rapid, intact method to determine the bruise degree of cherry, laying a foundation for cherry sorting and postharvest quality control.