Standard Normal Variate Transformation Research Articles

Near infrared spectroscopy as a fast and non-destructive technique for total acidity prediction of intact mango: Comparison among regression approaches

In the present study, the potential of near infrared spectroscopy (NIRS) as a rapid and non-destructive tools for quality attributes measurement of intact mango was investigated. Three different regression approaches namely partial least square regression (PLSR), support vector machine regression (SVMR), and artificial neural network (ANN) were used and compared in predicting total acidity (TA) of intact mangos. This quality parameter prediction models were established based on near infrared diffuse reflectance spectra acquired in wavelength range from 1000 to 2500 nm. Standard normal variate (SNV) transformation was applied as spectra enhancement prior to prediction models development. The results obtained show that ANN and SVMR are better than PLSR for TA prediction. The optimal prediction model for TA quality attribute were obtained by ANN with the first 4 principal components (PCs) scores as input. The coefficient determination of calibration (R2cal) and prediction (R2pred), the root-mean square error of calibration (RMSEC) and prediction (RMSEP), and the ratio of prediction to deviation (RPD) were 0.97, 0.89, 25.29 mg100g−1, 28.42 mg100g−1 and 4.02, respectively. The overall results satisfactorily demonstrate that NIRS technology combined with proper regression approaches has the promising results to determine TA of intact mango non-destructively.

Spectral Preprocessing Combined with Deep Transfer Learning to Evaluate Chlorophyll Content in Cotton Leaves.

Rapid determination of chlorophyll content is significant for evaluating cotton's nutritional and physiological status. Hyperspectral technology equipped with multivariate analysis methods has been widely used for chlorophyll content detection. However, the model developed on one batch or variety cannot produce the same effect for another due to variations, such as samples and measurement conditions. Considering that it is costly to establish models for each batch or variety, the feasibility of using spectral preprocessing combined with deep transfer learning for model transfer was explored. Seven different spectral preprocessing methods were discussed, and a self-designed convolutional neural network (CNN) was developed to build models and conduct transfer tasks by fine-tuning. The approach combined first-derivative (FD) and standard normal variate transformation (SNV) was chosen as the best pretreatment. For the dataset of the target domain, fine-tuned CNN based on spectra processed by FD + SNV outperformed conventional partial least squares (PLS) and squares-support vector machine regression (SVR). Although the performance of fine-tuned CNN with a smaller dataset was slightly lower, it was still better than conventional models and achieved satisfactory results. Ensemble preprocessing combined with deep transfer learning could be an effective approach to estimate the chlorophyll content between different cotton varieties, offering a new possibility for evaluating the nutritional status of cotton in the field.

Modeling and Prediction of Soil Organic Matter Content Based on Visible-Near-Infrared Spectroscopy

In order to explore the ever-changing law of soil organic matter (SOM) content in the forest of the Greater Khingan Mountains, a prediction model of the SOM content with a high accuracy and stability has been developed based on visible near-infrared (VIS-NIR) technology and multiple regression analysis. A total of 105 soil samples were collected from Cuifeng forest farm in Jagdaqi City, Greater Khingan Mountains region, Heilongjiang Province, China. Five classical preprocessing algorithms, including Savitzky−Golay convolution smoothing (S-G smoothing), standard normal variate transformation (SNV), multiplicative scatter correction (MSC), first derivative, second derivative, and the combinations of the above five methods were applied to the raw spectra. Wavelengths were optimized with five methods of competitive adaptive reweighted sampling (CARS), successive projections algorithm (SPA), uninformative variable elimination (UVE), synergy interval partial least square (SiPLS), and their combinations, and PLS models were developed accordingly. The results showed that when S-G smoothing is combined with SNV or MSC, both preprocessing strategies can improve the performance of the model. The prediction accuracy of SiPLS-PLS model and SiPLS-UVE-PLS model for the SOM content is higher than for other models, withan Rc2 of 0.9663 and 0.9221, RMSEC of 0.0645 and 0.0981, Rv2 of 0.9408 and 0.9270, and RMSEV of 0.0615 and 0.0683, respectively. The pretreatment strategies and characteristic variable selection methods used in this study could significantly improve the model performance and predicting efficiency.

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.

Estimating Soil Properties and Nutrients by Visible and Infrared Diffuse Reflectance Spectroscopy to Characterize Vineyards

Visible, near, and shortwave infrared (VIS-NIR-SWIR) reflectance spectroscopy, a cost-effective and rapid means of characterizing soils, was used to predict soil sample properties for four vineyards (central and north-western Spain). Sieved and air-dried samples were measured using a portable spectroradiometer (350–2500 nm) and compared for pistol grip (PG) versus contact probe (CP) setups. Raw data processed using standard normal variate (SVN) and detrending transformation (DT) were grouped into four subsets (VIS: 350–700 nm; NIR: 701–1000 nm; SWIR: 1001–2500 nm; and full range: 350–2500 nm) in order to identify the most suitable range for determining soil characteristics. The performance of partial least squares regression (PLSR) models in predicting soil properties from reflectance spectra was evaluated by cross-validation. The four spectral subsets and transformed reflectances for each setup were used as PLSR predictor variables. The best performing PLSR models were obtained for pH, electrical conductivity, and phosphorous (R2 values above 0.92), while models for sand, nitrogen, and potassium showed moderately good performances (R2 values between 0.69 and 0.77). The SWIR subset and SVN + DT processing yielded the best PLSR models for both the PG and CP setups. VIS-NIR-SWIR reflectance spectroscopy shows promise as a technique for characterizing vineyard soils for precision viticulture purposes. Further studies will be carried out to corroborate our findings.

Identification and quantification of additives in bituminous binders based on FTIR spectroscopy and multivariate analysis methods

Bitumen is a crucial building material in road construction, which is exposed to continuously higher stresses due to higher traffic loads and changing climatic conditions. Therefore, various additives are increasingly being added to the bitumen complicating the characterisation of the bituminous binder, especially concerning the reuse of reclaimed asphalt. Therefore, this work aimed to demonstrate that the combination of Fourier transform infrared (FTIR) spectroscopy with attenuated total reflexion (ATR) technique and multivariate evaluation is a very well-suited method to reliable identify and quantify additives in bituminous binders. For this purpose, various unmodified and modified binders, directly and extracted from laboratory and reclaimed asphalts, were investigated with FTIR-ATR spectroscopy. The determined spectra, pre-processed by standard normal variate (SNV) transformation and the determination of the 1st derivation, were evaluated using factor analysis (FA), linear discriminant analysis (LDA) and partial least square regression (PLSR). With this multivariate evaluation, first, a significant model with a very high hit rate of over 90% was developed allowing for the identification of styrene-butadiene copolymers (SBC), ethylene-copolymer bitumen (ECB) and different waxes (e.g. amide and Fischer-Tropsch wax) even if the additives do not show any additional peaks or the samples are multi-modified. Second, a quantification of the content is possible for SBC, ECB, and amide wax with a mean error of RMSE ≤ 0.4 wt% and a coefficient of determination of R2 > 90%. Based on these results, FTIR identification and quantification of additives in bituminous binders is a very promising method with a great potential.

Nondestructive Detection of Moisture Content in Walnut Kernel by Near-Infrared Diffuse Reflectance Spectroscopy

The rapid and accurate detection of the moisture content is of great significance to the quality evaluation and oil extraction process of walnut kernel. Near-infrared (NIR) spectroscopy is an ideal method for measuring the moisture content in walnut kernel. In this study, a regression model for moisture content in walnut kernel was developed based on NIR diffuse reflectance spectroscopy using chemometric methods. The different spectral pretreatment methods were adopted to preprocess the original spectral data. The whole spectra band was divided into 5 subbands, 10 subbands, 15 subbands, and 20 subbands to screen specific wavelengths relevant to the walnut kernel moisture content. PLS (partial least square regression), MLR (multivariate linear regression), PCR (principle component regression), and SVR (support vector regression) were used to establish the relationship model between the spectral data and measurement values of the moisture content. In comparison, the optimized modeling conditions were determined as follows: detection wavelength 1349–1490 nm, SNV-FD (standard normal variate transformation and first derivative) preprocessing method, and PLS algorithm. Under these conditions, the square correlation coefficient (R2) and root mean square error of prediction (RMSEP) of the prediction model were 0.9865 and 0.0017, respectively. The results of this study provided a feasible method for the rapid detection of moisture content in walnut kernel. To improve the performance and applicability of the model, it is necessary to continuously expand the size of the sample set.

Soil Nutrient Estimation and Mapping in Farmland Based on UAV Imaging Spectrometry.

Soil nutrient is one of the most important properties for improving farmland quality and product. Imaging spectrometry has the potential for rapid acquisition and real-time monitoring of soil characteristics. This study aims to explore the preprocessing and modeling methods of hyperspectral images obtained from an unmanned aerial vehicle (UAV) platform for estimating the soil organic matter (SOM) and soil total nitrogen (STN) in farmland. The results showed that: (1) Multiplicative Scattering Correction (MSC) performed better in reducing image scattering noise than Standard Normal Variate (SNV) transformation or spectral derivatives, and it yielded a result with higher correlation and lower signal-to-noise ratio; (2) The proposed feature selection method combining Successive Projections Algorithm (SPA) and Competitive Adaptive Reweighted Sampling algorithm (CARS), could provide selective preference for hyperspectral bands. Exploiting this method, 24 and 22 feature bands were selected for SOM and STN estimation, respectively; (3) The particle swarm optimization (PSO) algorithm was employed to obtain optimized input weights and bias values of the extreme learning machine (ELM) model for more accurate prediction of SOM and STN. The improved PSO-ELM model based on the selected preference bands achieved higher prediction accuracy (R2 of 0.73 and RPD of 1.91 for SOM, R2 of 0.63, and RPD of 1.53 for STN) than support vector machine (SVM), partial least squares regression (PLSR), and the ELM model. This study provides an important guideline for monitoring soil nutrient for precision agriculture with imaging spectrometry.

The application of Raman spectroscopy for monitoring product quality attributes in perfusion cell culture

In-situ Raman spectroscopy provides enhanced capabilities to monitor and control the mammalian cell culture process in real-time, which conforms to the concepts Process Analytical Technology (PAT) and Quality by Design (QbD) raised by U.S. Food and Drug Administration (FDA), and help us to overcome the challenges encountered in the Pharmaceutical R&D. Product quality control was addressed in FDA’s guidance to encourage continuous manufacturing, few work was done to maintain a steady output of drug product in the continuous mammalian cell culture process with the aid of Raman spectroscopy. Here, the state-of-the-art intensified perfusion cell culture - Raman integrated system was set up where product quality attributes on-line monitoring was archived, which could enhance our product quality attributes (PQA) control capabilities as well as ensuring product quality stability in the cell culture process.In this study, an intensified perfusion cell culture system integrated with real-time Raman analyzer was built-up at 3 L bench-top scale for product quality attributes monitoring. Real-time Raman spectrums were correlated with the PQA data of one-step Protein A purified sample after the alternating tangential flow (ATF) column via partial least square (PLS) modeling, results showed that the Raman spectrum could reflect the PQA changes as a function of elapsed time. Thus the real-time monitoring of 3 representative analytical items for cell culture, i.e., Non-reduced Microchip CE-SDS (SDS_Caliper-NR), size-exclusion chromatography (SEC) and N-linked glycan liquid chromatography (N-glycan LC) were evaluated through Raman analyzer. For different analytical items, around 6∼8 kinds of PLS models with different pre-processing methods were evaluated, which considered the effects of Raman shift range, spectrum pre-processing (1st order derivate or 2nd order derivate), utilization of variable importance in projection (V.I.P) on Raman shift range on the model performance. Then the trained models were tested in another intensified perfusion batch with changed process.As for SDS_Caliper-NR main peak, SEC monomer, high molecule weight (HMW) species, Mannose 5 (Man5) percentage, our optimized models could predict these items accurately, with relatively low root mean square error of cross validation (RMSECV) of 0.37 %, 0.44 %, 0.24 % and 0.51 %, respectively. Adapting these models in another 3 L intensified perfusion culture bioreactor with different processes, these models still have predictability, yielded root mean square error of prediction (RMSEP) of 1.88 %, 1.74 %, 0.90 %, 2.79 %, suggesting that Raman spectrums could capture the PQA profiling trends and these models were quite robust against some process changes like pH strategy, perfusion rate, additional feeding. We also found that the 1st order derivate with standard normal variate transformation (SNV), Raman wavenumber from 800 to 1800 cm−1 and V.I.P > 0.8 usually obtained the best performance in the training set, suggesting a golden point for Raman PQA monitoring of cell culture process.This work demonstrated that real-time Raman spectroscopy is an effective PAT tool for on-line product quality attributes monitoring in the cell culture process, especially for continuous perfusion cell culture, allowing us to explore the possibility of PQA on-line tuning in the continuous manufacturing.

Spectroscopic analysis of chia seeds

Chia seeds are becoming more and more popular in modern diets. In this contribution NIR and 2D-fluorescence spectroscopy were used to determine their nutritional values, mainly fat and protein content. 25 samples of chia seeds were analysed, whereof 9 samples were obtained from different regions in Kenya, 16 samples were purchased in stores in Germany and originated mostly from South America. For the purchased samples the nutritional information of the package was taken in addition to the values obtained for fat and protein, which were determined at the Hohenheim Core Facility. For the first time the NIR and fluorescence spectroscopy were used for the analysis of chia. For the spectral evaluation two different pre-processing methods were tested. Baseline correction with subsequent mean-centring lead to the best results for NIR spectra whereas SNV (standard normal variate transformation) was sufficient for the evaluation of fluorescence spectra. When combining NIR and fluorescence spectra, the fluorescence spectra were also multiplied with a factor to adjust the intensity levels. The best prediction results for the evaluation of the combined spectra were obtained for Kenyan samples with prediction errors below 0.2 g/100 g. For all other samples the absolute prediction error was 0.51 g/100 g for fat and 0.62 g/100 g for protein. It is possible to determine the amount of protein and fat of chia seeds by fluorescence and NIR spectroscopy. The combination of both methods is beneficial for the predictions. Chia seeds from Kenya had similar protein and lipid contents as South American seeds.

Determination of sugar content in Lingwu jujube by NIR-hyperspectral imaging.

Near infrared hyperspectral imaging (NIR-HSI) with a spectral range of 900 to 1700nm was for the first time used to predict the changes of sugar content in Lingwu jujube during storage. Monte Carlo method was adopted to detect outliers, and multiple scattering correction (MSC), standard normal variate transformation (SNV), and Baseline were used to optimize modeling. Competitive adaptive reweighted sampling (CARS), interval variable iterative space shrinkage approach (iVISSA), and interval random frog (IRF) were used to select optimal wavelengths. In addition, partial least square regression (PLSR) and support vector machine (SVM) modeling based on optimal wavelengths were compared. The results showed that 30, 30, and 24 wavelengths were selected by CARS; 106, 87, and 112 feature wavelengths were selected by iVISSA; and 96, 71, and 83 optimal wavelengths were selected by IRF for sucrose, fructose, and glucose, respectively. The CARS-PLSR models provided the best results for fructose and glucose, and iVISSA-SVM model was better for sucrose. The results indicated that NIR-HSI model may be used as a rapid and nondestructive method for the determination of sugar content in jujubes.

A novel spectral index for estimation of relative chlorophyll content of sugar beet

Fast and non-destructive estimation of canopy chlorophyll content from hyperspectral sensing is essential to monitor the physiological status of vegetation or to estimate crop above ground nitrogen status. The objective of this study is to propose an optimal method for the relative chlorophyll content (SPAD) estimation of sugar beet canopy using ground-based hyperspectral imagery. Field experiments were conducted over three years at three different growth stages, across three different sites, using different cultivars and nitrogen (N) application rates. Quantitative correlations between SPAD value and canopy hyperspectral reflectance of sugar beet canopy after different pretreatment algorithms were established. Ten classical spectral indexes selected from the literature for estimating SPAD value in sugar beet canopy were evaluated and compared to a novel modified chlorophyll index (MCI) produced in this study by introducing a parameter to chlorophyll index (CI) to improve the estimation accuracy. Normalized difference vegetation index (NDVI) and chlorophyll index (CI) were optimized by using all possible combinations of spectral bands from the range of 390 nm to 990 nm. The prediction performance of partial least squares (PLS) regression models for optimized indexes (e.g., NDVI, CI and MCI), compared to the corresponding classical spectral indexes (e.g., ND550, ND705, CIgreen and CIred edge) was examined. Results showed that standard normal variate transformation (SNV) was the best pretreatment method for the hyperspectral data of this study. Models resulted after bands combinations optimization were found to be more accurate than models developed using the classical spectral indexes. The performance of proposed spectral index, MCI (R747, R839), MCI (R861, R884) and MCI (R931, R770), were best for the prediction accuracy of SPAD value in sugar beet for the validation set with the coefficient of determination (R2) of 0.83, 0.70 and 0.75, the root mean square error (RMSE) of 2.37, 3.11 and 2.78, and the relative root mean square error (RRMSE) of 4.95%, 6.05% and 5.75%, for the rapid growth stage of leaf cluster, sugar growth stage and sugar accumulation stage, respectively. It can be concluded that the index proposed can be implemented for the prediction of SPAD value of sugar beet using proximal hyperspectral sensors under a wide range of environmental conditions.

Dynamic monitoring oxidation process of nut oils through Raman technology combined with PLSR and RF-PLSR model

During preservation of nuts, nut oils are easily oxidized; hence, peroxide value (PV) is an important evaluation index. In this study, a novel method for the determination of the PV of nuts based on partial-least-square regression (PLSR) and Forest random PLSR (RF-PLSR) model was established. Meanwhile, the Raman spectrum was processed by 24 spectral pretreatment methods to transform the whole Raman band, and the best band was selected by RF method. Among the whole bands, 36 wavenumbers were selected to establish the PLSR model. The R-square of the correction set (R2c) and prediction set (R2p) of the optimal Standard normal variate transformation + first Derivative-PLSR model and RF-PLSR were 0.9552, 0.8672, 0.8048, and 0.7927, while the root-mean-square error of calibration (RMSEC) and prediction (RMSEP) were 0.067, 0.1100, 0.1514, and 0.1547, respectively. These results showed that Raman spectroscopy combined with chemometrics could be used to establish a rapid, nondestructive, and precise method for the determination of oil oxidation index.

Method and system for nondestructive detection of freshness in Penaeus vannamei based on hyperspectral technology

To test the freshness of Penaeus vannamei rapidly and non-destructively, we constructed a quantitative prediction model for total volatile basic nitrogen (TVB-N) content by collecting 860–1700 nm hyperspectral data and TVB-N content of samples of P. vannamei during eight consecutive days. In this study, we compared the influence of six kinds of spectral pretreatment on the partial least-squares regression modeling (PLSR), which included multiple scattering correction (MSC), standard normal variate transformation (SNV), normalized mean centering, 11-point moving-average smoothing, and Savitzky–Golay smoothing. The results demonstrate that the prediction effect of the MSC-PLSR model has the best relative model accuracy, under the pretreatment of the MSC algorithm for spectral data. In addition, principal component analysis (PCA) and the Mahalanobis distance algorithm were used to remove anomalous samples. The results reveal that the prediction effect of the MSC-PLSR model is better than other models that we considered, and the correlation coefficient (R) of the validation set reached 0.8147. Finally, this study developed a graphical user interface system to predict freshness, based on our research, which provides a method and an analysis tool to rapidly detect the freshness of P. vannamei.

Detection of breast cancer of various clinical stages based on serum FT-IR spectroscopy combined with multiple algorithms.

Breast cancer screening is time consuming, requires expensive equipment, and has demanding requirements for doctors. Hence, a large number of breast cancer patients may miss screening and early treatment, which greatly threatens their health around the world. Infrared spectroscopy may be able to be used as a screening tool for breast cancer detection. Fourier transform infrared (FT-IR) spectroscopy of serum was combined with traditional machine learning algorithms to achieve an auxiliary diagnosis that could quickly and accurately distinguish patients with different stages of breast cancer, including stage 1 disease, from control subjects without breast cancer. FT-IR spectroscopy were performed on the serum of 114 non-cancer control subjects, 35 patients with stage I, 43 patients with stage II, and 29 patients with stage III & IV breast cancer. Due to the experimental sample imbalance, we used the oversampling to process the four classes of sample. The oversampling selected Synthetic Minority Oversampling Technique (SMOTE). Subsequently, we used the random discarding method in undersampling to do experiments as well. The average FT-IR spectroscopy results for the four groups showed differences in phospholipids, nucleic acids, lipids, and proteins between non-cancer control subjects and breast cancer patients at different stages. Based on these differences, four classification models were used to classify stage I, II, III & IV breast cancer patients and non-cancer control subjects. First, standard normal variate transformation (SNV) was used to preprocess the original data, and then partial least squares (PLS) was used for feature extraction. Finally, the five models were established including extreme learning machine (ELM), k-nearest neighbor (KNN), genetic algorithms based on support vector machine (GA-SVM), particle swarm optimization-support vector machine (PSO-SVM) and grid search-support vector machine (GS-SVM). In oversampling experiment, the GS-SVM classifier obtained the highest average classification accuracy of 95.45 %; the diagnostic accuracy of non-cancer control subjects was 100 %; breast cancer stage I was 90 %; breast cancer stage II was 84.62 %; and breast cancer stage III & IV was 100 %. In undersampling experiment, the GA-SVM model obtained the highest average classification accuracy of 100 %; the diagnostic accuracy of non-cancer control subjects was 100 %; breast cancer stage I was 100 %; breast cancer stage II was 100 %; and breast cancer stage III & IV was 100 %. The results show that FT-IR spectroscopy combined with powerful classification algorithms has great potential in distinguishing patients with different stages of breast cancer from non-cancer control subjects. In addition, this research provides a reference for future multiclassification studies of cervical cancer, ovarian cancer and other female high-incidence cancers through serum FT-IR spectroscopy.

Assessing black tea quality based on visible–near infrared spectra and kernel-based methods

Current evaluation systems for tea quality assessment are difficult to use and unstable. A rapid and cheap assessment method is required to distinguish the quality levels of tea. In this study, a visible-near-infrared (Vis–NIR) spectrometer and support vector machine (SVM)-based kernels were used for the qualitative categorization of black tea. First, a Vis–NIR system was used to acquire the spectral data of seven levels of tea samples. The obtained spectra were preprocessed using the Savitzky–Golay smoothing combined with the first derivative and standard normal variate transformation. Then, four characteristic wavelength selection algorithms, namely synergy interval partial least square, competitive adaptive reweighted sampling (CARS), the variable iterative space shrinkage approach (VISSA), and the interval VISSA, were used to obtain preprocessed spectral features. Finally, the SVM was used with four kernel functions, namely the linear, Gaussian, quadratic, and cubic functions, to develop models based on the variables obtained from the selected features for tea quality classification. The results revealed that the CARS–linear kernel SVM model exhibited the best results, with a correct identification rate of 91.85 % in the validation process. Our findings demonstrate that Vis-NIR spectroscopy can be a rapid, inexpensive, efficient, and alternative method for predicting the quality of black tea.

Fast determination of three components in milk thistle extract with a hand-held NIR spectrometer and chemometrics tools

Milk thistle extract that is widely supplied in the market has three main active ingredients, including silybin, silychristin and silydianin (SS), and isosilybin, they have a lot of valuable functions for the human body. The determination of these active ingredients is of great significance for customers, so it is urgent to develop convenient, fast, and on-site analytical methods. In this study, we developed a fast method to detect the active ingredients using a hand-held NIR spectrometer MicroNIR 1700 to record spectra and applied multiple pretreatment methods to process them. The random frog (RG) method was used to optimize wavelength variables with high information, and the partial least square (PLS) regression was used to calibrate the spectra data of silybin, SS, and isosilybin. The results showed that the standard normal variate transformation (SNV), SNV+ first derivative, and first derivative were the best pretreatment methods for silybin, SS, and isosilybin. The calibration performances of selected wavelength variables were significantly improved compared to the whole wavelength variables. The the minimum limit of quantification (LOQmin) of the silybin, SS, and isosilybin were 9.53%, 10.86%, and 2.64%, respectively. For the validation of new unknown samples, the root mean squared error of prediction (RMSEP) of silybin, SS, and isosilybin were 0.666%, 0.510%, and 0.157%, and the corresponding R2 were 0.992, 0.898, and 0.936, respectively, it demonstrated that the proposed methods are accurate and robust. Benefiting from the portable hand-held NIR spectrometer and the reliable algorithm, the present method offers a promising approach for the determination of the active ingredients in milk thistle extract because it is rapid, convenient, non-polluting, and on-site, does not require professional skills and expensive apparatus.

On-site identification of Ophiocordyceps sinensis using multispectral imaging and chemometrics

For the reasonable and effective collection of Ophiocordyceps sinensis, a new method of on-site identification was attempted using a portable multispectral imaging (MSI) technique. Three dimensional (3D) data-cubes of representative Ophiocordyceps sinensis and weeds samples were acquired and pre-processed with standard normal variate transformation (SNV). Principal component analysis (PCA) and simulated annealing particle swarm optimisation (SAPSO) algorithms were used to extract characteristic images and develop the support vector classification (SVC) models. Results show that the fused feature model of SAPSO-SVC has the best performance, resulting in a recognition accuracy of the prediction set of 96.30%. Moreover, on-site distribution map of Ophiocordyceps sinensis and weeds was created using the spectral feature model of SAPSO-SVC, and the target could be easily identified from the distribution map. This work demonstrates the potential for on-site identification of Ophiocordyceps sinensis in the Qinghai–Tibet Plateau using a portable MSI technique combined with the SAPSO-SVC algorithm. Keywords: Ophiocordyceps sinensis, MSI, SAPSO-SVC, On-site distribution map DOI: 10.25165/j.ijabe.20201306.5425 Citation: Duan H W, Tong X, Cui R X, Han L J, Huang G Q. On-site identification of Ophiocordyceps sinensis using multispectral imaging and chemometrics. Int J Agric & Biol Eng, 2020; 13(6): 166–170.

Prediction of Soil Properties by Visible and Near-Infrared Reflectance Spectroscopy

The major goal of this research was to assess the ability of visible-near-infrared (VIS-NIR) spectroscopy combined with multivariate regressions based on four modeling algorithms, namely partial least squares regression (PLSR), principal component regression (PCR), support vector regression(SVR), and random forest (RF) to quantify some quality soil parameters. Saturated electrical conductivity (ECe), organic carbon (OC), and texture (including sand, silt, and clay) were measured. Overall, 114 soil samples were taken from 30 profiles (0–100 cm) in Doviraj plain, western Iran. For improving spectral data, various preprocessing techniques including Savitzky-Golay first deviation (SGD), Normalization (N), and Standard Normal Variate transformation (SNV) were employed. A leave-one-out cross-validation procedure was used. Results revealed that RF model produced much more precise predictions compared with PCR, SVR, and PLSR models and specifying the most important variables for the model. Also, the highest accuracy for RF model was achieved by SGD preprocessing for ECe, OC, sand, and clay with validation coefficient of determination (R2val) equal to 0.57, 0.75, 0.70, and 0.73, respectively; also, validation root mean square error (RMSEval) was equal to 0.74 dS/m, 0.13, 12.97, and 0.14%, respectively, However, SNV preprocessing with RF model caused to best result with R2val = 0.65 and RMSE val = 9.21% for silt. This methodology displays the capability of the VIS-NIR for estimation of ECe, OC, and texture.

Rapid discrimination of intact beef, venison and lamb meat using Raman spectroscopy

With increasing demand for fast and reliable techniques for intact meat discrimination, we explore the potential of Raman spectroscopy in combination with three chemometric techniques to discriminate beef, lamb and venison meat samples. Ninety (90) intact red meat samples were measured using Raman spectroscopy, with the acquired spectral data preprocessed using a combination of rubber-band baseline correction, Savitzky-Golay smoothing and standard normal variate transformation. PLSDA and SVM classification were utilized in building classification models for the meat discrimination, whereas PCA was used for exploratory studies. Results obtained using linear and non-linear kernel SVM models yielded sensitivities of over 87 and 90 % respectively, with the corresponding specificities above 88 % on validation against a test set. The PLSDA model yielded over 80 % accuracy in classifying each of the meat specie. PLSDA and SVM classification models in combination with Raman spectroscopy posit an effective technique for red meat discrimination.