Rapid discrimination of Panax ginseng powder adulterated with various root plants by FT-IR spectroscopy coupled with multivariate analysis.

Panax ginseng powder adulterated with other root plants (arrowroot, bellflower, and lance asiabell) was discriminated using Fourier transform infrared (FT-IR) spectroscopy, combined with multivariate analysis. Principal component analysis visually diagnosed the adulteration by showing two distinct clusters based on presence of adulteration. Wavenumber regions (1000 cm-1 and 3300 cm-1) selected from the loading plot associated with the vibration of OH and CH bond in ginsenoside and aromatic compounds. A quantitative model for the content of ginsenosides and specific aromatic compounds as indicators of pure ginseng powder, was developed based on partial least square regression analysis. The performance of the prediction model preprocessed with the Savizky-Golay 1st derivative was improved to R2 of 0.9650, 0.9635, and 0.9591 for Rb1, Rc, and ß-Panasinsene, respectively. Therefore, FT-IR technology makes it possible to rapidly authenticate pure ginseng product based on the ginsenoside contents and aroma compound.

Detection of Red Pepper Powder Adulteration with Allura Red and Red Pepper Seeds Using Hyperspectral Imaging.

This study used shortwave infrared (SWIR) technology to determine whether red pepper powder was artificially adulterated with Allura Red and red pepper seeds. First, the ratio of red pepper pericarp to seed was adjusted to 100:0 (P100), 75:25 (P75), 50:50 (P50), 25:75 (P25), or 0:100 (P0), and Allura Red was added to the red pepper pericarp/seed mixture at 0.05% (A), 0.1% (B), and 0.15% (C). The results of principal component analysis (PCA) using the L, a, and b values; hue angle; and chroma showed that the pure pericarp powder (P100) was not easily distinguished from some adulterated samples (P50A-C, P75A-C, and P100B,C). Adulterated red pepper powder was detected by applying machine learning techniques, including linear discriminant analysis (LDA), linear support vector machine (LSVM), and k-nearest neighbor (KNN), based on spectra obtained from SWIR (1,000-1,700 nm). Linear discriminant analysis determined adulteration with 100% accuracy when the samples were divided into four categories (acceptable, adulterated by Allura Red, adulterated by seeds, and adulterated by seeds and Allura Red). The application of SWIR technology and machine learning detects adulteration with Allura Red and seeds in red pepper powder.

Determination of Freshness of Mackerel (Scomber japonicus) Using Shortwave Infrared Hyperspectral Imaging.

Shortwave infrared (SWIR) hyperspectral imaging was applied to classify the freshness of mackerels. Total volatile basic nitrogen (TVB-N) and acid values, as chemical compounds related to the freshness of mackerels, were also analyzed to develop a prediction model of freshness by combining them with hyperspectral data. Fresh mackerels were divided into three groups according to storage periods (0, 24, and 48 h), and hyperspectral data were collected from the eyes and whole body, separately. The optimized classification accuracies were 81.68% using raw data from eyes and 90.14% using body data by multiple scatter correction (MSC) pretreatment. The prediction accuracy of TVB-N was 90.76%, and the acid value was 83.76%. These results indicate that hyperspectral imaging, as a nondestructive method, can be used to verify the freshness of mackerels and predict the chemical compounds related to the freshness.

Effect of Moisture Content Difference on the Analysis of Quality Attributes of Red Pepper (Capsicum annuum L.) Powder Using a Hyperspectral System.

The variety of characteristics of red pepper makes it difficult to analyze at the production field through hyperspectral imaging. The importance of pretreatment to adjust the moisture content (MC) in the process of predicting the quality attributes of red pepper powder through hyperspectral imaging was investigated. Hyperspectral images of four types of red pepper powder with different pungency levels and MC were acquired in the visible near-infrared (VIS-NIR) and short-wave infrared (SWIR) regions. Principal component analysis revealed that the powders were grouped according to their pungency level, color value, and MC (VIS-NIR, Principal Component 1 = 95%; SWIR, Principal Component 1 = 91%). The loading plot indicated that 580-610, 675-760, 870-975, 1020-1130, and 1430-1520 nm are the key wavelengths affected by the presence of O-H and C-H bonds present in red pigments, capsaicinoids, and water molecules. The R2 of the partial least squares model for predicting capsaicinoid and free sugar in samples with a data MC difference of 0-2% was 0.9 or higher, and a difference of more than 2% in MC had a negative effect on prediction accuracy. The color value prediction accuracy was barely affected by the difference in MC. It was demonstrated that adjusting the MC is essential for capsaicinoid and free sugar analysis of red pepper.

Hyperspectral imaging technology for monitoring of moisture contents of dried persimmons during drying process.

The moisture content of persimmons during drying was monitored by hyperspectral imaging technology. All persimmons were dried using a hot-air dryer at 40 °C and divided into seven groups according to drying time: semi-dried persimmons (Cont), 1 day (DP-1), 2 days (DP-2), 3 days (DP-3), 4 days (DP-4), 5 days (DP-5), and 6 days (DP-6). Shortwave infrared hyperspectral spectra and moisture content of all persimmons were analyzed to develop a prediction model using partial least squares regression. There were obvious absorption bands: two at approximately 971 nm and 1452 nm were due to water absorption related to O-H stretching of the second and first overtones, respectively. The R-squared value of the optimal calibration model was 0.9673, and the accuracy of the moisture content measurement was 95%. These results indicate that hyperspectral imaging technology can be used to predict and monitor the moisture content of dried persimmons during drying.

Estimating the Composition of Food Nutrients from Hyperspectral Signals Based on Deep Neural Networks.

There is an increasing demand for acquiring details of food nutrients especially among those who are sensitive to food intakes and weight changes. To meet this need, we propose a new approach based on deep learning that precisely estimates the composition of carbohydrates, proteins, and fats from hyperspectral signals of foods obtained by using low-cost spectrometers. Specifically, we develop a system consisting of multiple deep neural networks for estimating food nutrients followed by detecting and discarding estimation anomalies. Our comprehensive performance evaluation demonstrates that the proposed system can maximize estimation accuracy by automatically identifying wrong estimations. As such, if consolidated with the capability of reinforcement learning, it will likely be positioned as a promising means for personalized healthcare in terms of food safety.

Prediction of shelf-life and changes in the quality characteristics of semidried persimmons stored at different temperatures.

This study was performed to investigate changes in the quality characteristics and shelf-life of semidried persimmons stored at different temperatures using acceleration experiments. In order to estimate quality changes in the samples, we evaluated the physicochemical properties, microbiological changes, and sensory features of the samples periodically after storage at -20, -10, 0, and 10 °C. At all storage temperatures, CIE L * a * b * values decreased significantly. Based on the results of this study, regression equations are set up. L * had the highest correlation and were therefore used to determine quality factor. The activation energy, which was calculated using the Arrhenius equation, was found to be 12.98 kcal/mol, and the Q10-values were 3.81, 2.07, and 2.06 at -20 to -10 °C, -10 to 0 °C, and 0 to 10 °C, respectively. Therefore, the expected expiration dates of the semidried persimmons were estimated to be 203.83, 53.46, 22.00, and 8.71 days at -20, -10, 0, and 10 °C.

Qualitative properties of roasting defect beans and development of its classification methods by hyperspectral imaging technology.

Qualitative properties of roasting defect coffee beans and their classification methods were studied using hyperspectral imaging (HSI). The roasting defect beans were divided into 5 groups: medium roasting (Cont), under developed (RD-1), over roasting (RD-2), interior under developed (RD-3), and interior scorching (RD-4). The following qualitative properties were assayed: browning index (BI), moisture content (MC), chlorogenic acid (CA), trigonelline (TG), and caffeine (CF) content. Their HSI spectra (1000-1700nm) were also analysed to develop the classification methods of roasting defect beans. RD-2 showed the highest BI and the lowest MC, CA, and TG content. The accuracy of classification model of partial least-squares discriminant was 86.2%. The most powerful wavelength to classify the defective beans was approximately 1420nm (related to OH bond). The HSI reflectance values at 1420nm showed similar tendency with MC, enabling the use of this technology to classify the roasting defect beans.

Image analysis to evaluate the browning degree of banana (Musa spp.) peel.

Image analysis was applied to examine banana peel browning. The banana samples were divided into 3 treatment groups: no treatment and normal packaging (Cont); CO2 gas exchange packaging (CO); normal packaging with an ethylene generator (ET). We confirmed that the browning of banana peels developed more quickly in the CO group than the other groups based on sensory test and enzyme assay. The G (green) and CIE L(∗), a(∗), and b(∗) values obtained from the image analysis sharply increased or decreased in the CO group. And these colour values showed high correlation coefficients (>0.9) with the sensory test results. CIE L(∗)a(∗)b(∗) values using a colorimeter also showed high correlation coefficients but comparatively lower than those of image analysis. Based on this analysis, browning of the banana occurred more quickly for CO2 gas exchange packaging, and image analysis can be used to evaluate the browning of banana peels.