Spectral Pretreatment Methods Research Articles

Research on Enhanced Detection of Benzoic Acid Additives in Liquid Food Based on Terahertz Metamaterial Devices.

It is very important for human health to supervise the use of food additives, because excessive use of food additives will cause harm to the human body, especially lead to organ failures and even cancers. Therefore, it is important to realize high-sensibility detection of benzoic acid, a widely used food additive. Based on the theory of electromagnetism, this research attempts to design a terahertz-enhanced metamaterial resonator, using a metamaterial resonator to achieve enhanced detection of benzoic acid additives by using terahertz technology. The absorption peak of the metamaterial resonator is designed to be 1.95 THz, and the effectiveness of the metamaterial resonator is verified. Firstly, the original THz spectra of benzoic acid aqueous solution samples based on metamaterial are collected. Secondly, smoothing, multivariate scattering correction (MSC), and smoothing combined with first derivative (SG + 1 D) methods are used to preprocess the spectra to study the better spectral pretreatment methods. Then, Uninformative Variable Elimination (UVE) and Competitive Adaptive Reweighted Sampling (CARS) are used to explore the optimal terahertz band selection method. Finally, Partial Least Squares (PLS) and Least square support vector machine (LS-SVM) models are established, respectively, to realize the enhanced detection of benzoic acid additives. The LS-SVM model combined with CARS has the best effect, with the correlation coefficient of prediction set (Rp) is 0.9953, the root mean square error of prediction set (RMSEP) is 7.3 × 10−6, and the limit of detection (LOD) is 2.3610 × 10−5 g/mL. The research results lay a foundation for THz spectral analysis of benzoic acid additives, so that THz technology-based detection of benzoic acid additives in food can reach requirements stipulated in the national standard. This research is of great significance for promoting the detection and analysis of trace additives in food, whose results can also serve as a reference to the detection of antibiotic residues, banned additives, and other trace substances.

Feasibility study on the use of near-infrared spectroscopy for rapid and nondestructive determination of gossypol content in intact cottonseeds

BackgroundGossypol found in cottonseeds is toxic to human beings and monogastric animals and is a primary parameter for the integrated utilization of cottonseed products. It is usually determined by the techniques relied on complex pretreatment procedures and the samples after determination cannot be used in the breeding program, so it is of great importance to predict the gossypol content in cottonseeds rapidly and nondestructively to substitute the traditional analytical method.ResultsGossypol content in cottonseeds was investigated by near-infrared spectroscopy (NIRS) and high-performance liquid chromatography (HPLC). Partial least squares regression, combined with spectral pretreatment methods including Savitzky-Golay smoothing, standard normal variate, multiplicative scatter correction, and first derivate were tested for optimizing the calibration models. NIRS technique was efficient in predicting gossypol content in intact cottonseeds, as revealed by the root-mean-square error of cross-validation (RMSECV), root-mean-square error of prediction (RMSEP), coefficient for determination of prediction (Rp2), and residual predictive deviation (RPD) values for all models, being 0.05∼0.07, 0.04∼0.06, 0.82∼0.92, and 2.3∼3.4, respectively. The optimized model pretreated by Savitzky-Golay smoothing + standard normal variate + first derivate resulted in a good determination of gossypol content in intact cottonseeds.ConclusionsNear-infrared spectroscopy coupled with different spectral pretreatments and partial least squares (PLS) regression has exhibited the feasibility in predicting gossypol content in intact cottonseeds, rapidly and nondestructively. It could be used as an alternative method to substitute for traditional one to determine the gossypol content in intact cottonseeds.

Critical quality attribute assessment of big brand traditional Chinese medicine: study on NIR field detection method of chemical properties of moisture in Tongren Niuhuang Qingxin Pills

For the field detection problems of critical quality attribute(CQA) of moisture content in traditional Chinese medicine(TCM) manufacturing process, big brand TCM Tongren Niuhuang Qingxin Pills were used as the carrier, to establish a moisture content NIR field detection model with or without cellophane in real world production with use of near infrared(NIR) spectroscopy combined with stoichiometry. With the moisture content determined by drying method as reference value, the partial least square method(PLS) was used to analyze the correlation between the spectrum and the moisture reference value. Then the spectral pretreatment methods were screened and optimized to further improve the accuracy and stability of the model. The results showed that the best quantitative model was developed by the spectral data pretreatment of standard normal variate(SNV) with the latent variable factor number of 2 and 7 of Tongren Niuhuang Qingxin Pills with or without cellophane samples. The prediction coefficient of determination(R_(pre)~2) and standard deviation of prediction(RMSEP) of the model with cellophane samples were 0.765 7 and 0.157 2%; R_(pre)~2 and RMSEP of the model without cellophane samples were 0.772 2 and 0.207 8%. The NIR quantitative models of moisture content of Tongren Niuhuang Qingxin Pills with and without cellophane both showed good predictive performance to realize the rapid, accurate and non-destructive quantitative analysis of moisture content in such pills, and provide a method for the field quality control of the critical chemical attributes of moisture in the manufacturing of big brand TCM.

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.

Effective Quantification of Tannin Content in Sorghum Grains Using Near-infrared Spectroscopy

This study was conducted to investigate the feasibility of determining tannin content in sorghum grains with near-infrared reflectance spectroscopy (NIRS). A total of 110 sorghum grain samples were collected. The data matrix of the pretreated NIRS was randomly divided into a calibration set (Nc=77 samples) and a prediction set (Np=33 samples). The analysis of tannin content was based on the colorimetric method of GBT 15686-2008. Diffuse reflectance spectra of 110 sorghum samples were generated on a Fourier-transform NIRS with a scanning range of 12800-4000 cm^-1 and resolution of 16 cm^-1 and 64 scans. Several spectra pretreatment methods were compared to for an optimum spectral pretreatment method. The optimal model was determined according to coefficient of determination for calibration (R²<SUB>CAL</sub>), root mean standard error of calibration (RMSE<SUB>CAL</sub>), coefficient of determination for cross-validation (R²<SUB>CV</sub>), root mean standard error of cross-validation (RMSE<SUB>CV</sub>) and the residual predictive deviation (RPD). The results showed that the tannin content of the sorghum grains ranged from 0.01% to 2.12% DM with the average of 0.58%, and first derivative was the optimal spectral pretreatment with the lowest RMSE<SUB>CV</sub> of 0.14. The absorption peaks of the optimal model mainly located at 9402-7492 cm^-1 and 5452-4244 cm^-1. The RPD of calibration, cross-validation and external validation were 6.22, 4.22 and 3.0, respectively. The findings suggest that the established model using NIRS is effective to quantify tannin content in sorghum grains rapidly.

Intelligent spectral algorithm for pigments visualization, classification and identification based on Raman spectra

Raman spectroscopy is a molecular vibrational spectroscopic technique has developed rapidly in recent years, especially in rapid field detection. In this paper, we discuss the Raman spectral pretreatment method and classification algorithm by using nearly 300 pigments spectral data as an example. Here, more than 5 kinds of classification algorithms such as SVM, KNN, ANN and et al are used to sovle the problem of pigments visualization, classification and identification via Raman spectral, and the results show that most of the algorithms fit well, with an accuracy of 90%. Moreover, SNR (Signal to noise ratio) is introduced to evaluate the stability of our algorithm. When the SNR is low, the accuracy of the algorithm decreases sharply. When the SNR was 1, the accuracy rate reached the highest value of 39.46%. In order to slove this problem, the flattopwin, hanning, blackman algorithm was introduced to denoise the signal with low SNR, even when SNR = 1, the signal is 80% accurate. It is proved that in the extreme case of this application, the algorithm still maintains good accuracy, and our research pave the way to use interlligent algorithms to solve the problems in the fields of Raman spectral detection.

A portable detector on main compositions of raw and homogenized milk

Fat, protein, lactose and total solids are important indicators in evaluating the nutritional and commercial values of milk. To make the main compositions of milk be detected rapidly and conveniently, and to realize the detection in field or on-site, a portable detector on main compositions of milk was developed. The detector mainly consisted of a visible/near-infrared micro-spectrometer (650–1100 nm), a Raspberry Pi board, a micro-halogen lamp, a self-made “Y” shape optical fiber probe, a lithium battery and an organic light emitting diode display. The software of the detector was developed by using Python language to realize calibration, diffuse reflectance spectral collection, spectral pretreatment and calculation on main compositions of milk, i.e, fat, protein, lactose and total solids. One hundred and twenty raw milk and their homogenized milk were used as samples to establish the partial least squares regression (PLSR) models to predict the main compositions. The effect of several spectral preprocessing methods on prediction performance of established models were evaluated, and the results indicated that Savitzky-Golay smoothing was the best spectral pretreatment method for raw milk and Savitzky-Golay (S-G) smoothing combined with standard normal variate was the best method for homogenized milk. The precision of the detector were validated by another batch of 40 raw milk and their homogenized milk samples. The validation results showed that the root-mean-squares error of the explored detector for fat, protein, lactose and total solids contents, respectively, were 0.14%, 0.14%, 0.08% and 0.27% for raw milk, and were 0.10%, 0.12%, 0.08% and 0.21% for homogenized milk. The detector was just about 1 kg in weight and the measurement time for one sample was less than 3 s.

Systematic vs. stepwise parameter optimization for discriminant model development: A case study of differentiating Pinellia ternata from Pinellia pedatisecta with near infrared spectroscopy

Near infrared (NIR) spectroscopy is an effective technique for adulteration detection in traditional Chinese medicine. The aim is to develop a discriminant model with the aid of chemometrics tools. The discriminant model is conventionally established by the means of stepwise optimization. This approach is often limited to trial-and-error and considered as a burden. In this study, a systematic optimization approach was proposed to develop the discriminant model with the aid of the design of experiment tools and applied to a case study of differentiating Pinellia ternata from Pinellia pedatisecta and adulterated Pinellia ternata using NIR spectroscopy. Spectral pretreatment, variable selection, and discriminant methods were identified as critical factors. The classification accuracy and no-error rate of the calibration set, cross-validation, and the prediction set were calculated to evaluate the performance of discriminant models. A full factorial design was applied to analyze the effect of critical factors at different levels on the model performance and optimize these factors. Three discriminant models including discriminant analysis coupled with principal component analysis (PCA-DA), partial least squares – discriminant analysis (PLS-DA), and k-nearest neighbors (KNN) were obtained by systematic optimization. The performance of PCA-DA and PLS-DA models obtained by systematic optimization was very good, and no samples were misclassified, which were better than those obtained by stepwise optimization. The performance of the KNN model obtained by systematic optimization was not desired and it was equal to that obtained by stepwise optimization. The results showed that Pinellia ternata could be successfully discriminated from Pinellia pedatisecta and adulterated Pinellia ternata by the PCA-DA and PLS-DA models. Compared to the stepwise optimization approach, the systematic optimization approach can improve the PCA-DA and PLS-DA model performance for differentiating Pinellia ternata from Pinellia pedatisecta and adulterated Pinellia ternata.

A rapid analytical method for the quantitative determination of the sugar in acarbose fermentation by infrared spectroscopy and chemometrics.

In this article, a rapid analytical method for the quantitative determination of the glucose and maltose in the industrial acarbose fermentation was established by the combination application of infrared spectroscopy and chemometrics. The spectra of the 398 acarbose samples were collected by a portable infrared fast analyzer and the concentration of the glucose and maltose in the acarbose fermentation solution were determinate by high performance liquid chromatography (HPLC) as the referent database. Four spectral pretreatment methods, first derivative (FD), second derivative (SD), Savitzky-Golay (SG) convolution smoothing and mean center (MC) were employed to eliminate the optical interference from background and other noise information. The best result was obtained with FD+SG(21, 3)+MC method. The effects of different principal component numbers (PCs) on the parameters were also optimized. Two models of PLS and MLR, were used to predict the concentration of the glucose and maltose. The FD+SG(21, 3)+MC method was chosen as best method, with 12 PCs for glucose and 11 for maltose as optimized parameters. The PLS model was significantly better than the MLR model. Furthermore, both the predicted values and the reference values of glucose and maltose models showed superior linear relationship within the calibration range. The absolute errors of the predicted values and their corresponding reference values of glucose and maltose in the PLS model were within ±0.14 and ±0.35 confidence intervals, respectively. The prediction correct rate was 98.3%, which indicated that the prediction results of model were excellent.

Hyperspectral Inversion of Petroleum Hydrocarbon Contents in Soil Based on Continuum Removal and Wavelet Packet Decomposition

Hyperspectral remote sensing is widely used to detect petroleum hydrocarbon pollution in soil monitoring. Different spectral pretreatment methods seriously affect the prediction and analysis of petroleum hydrocarbon contents (PHCs). This study adopted a combined spectral data preprocessing technique that improves the prediction accuracy of petroleum hydrocarbons in soil. We combined continuum removal and wavelet packet decomposition (CR–Daubechies 3 (db3)) to process the hyperspectral reflectance data of 26 soil samples in the oil production work area in China and judged the correlation between spectral reflectance and petroleum hydrocarbons in soil. Partial least squares regression was used to construct an optimal model for the inversion of PHCs in soil and the leave-one-out cross-validation was used to select the best factor number. The best model of soil petroleum hydrocarbon inversion was determined by comprehensively comparing the initial spectrum, db3 to high-frequency spectrum, db3 to low-frequency spectrum, after-continuum removal spectrum, CR-db3 to high-frequency spectrum, and CR-db3 to low-frequency spectrum comprehensively. The main contributions of this study are as follows: (1) three-layer decomposition with CR-db3 can improve the correlation between spectral reflectance and PHCs and effectively improve the sensitivity of the spectrum to PHCs; (2) the prediction accuracy of the high-frequency spectrum of wavelet packet decomposition for PHCs in soil is higher than that of low-frequency information; (3) the proposed petroleum hydrocarbon prediction model based on CR-db3 processed spectra to obtain high-frequency information is optimal (coefficient of determination = 0.977, root mean square error of calibration = 3.078, root mean square error of cross-validation = 4.727, root mean square error of prediction = 4.498, ratio of performance to deviation = 6.12).

Feasibility of NIR spectroscopy detection of moisture content in coco-peat substrate based on the optimization characteristic variables

Moisture content is an important index to evaluate the water content in substrate. Near-infrared (NIR) spectroscopy was used for rapid quantitative detection of moisture content of coco-peat substrate. The different spectral pretreatment methods were adopted to pre-process the spectral data. Successive projection algorithm (SPA), elimination of uninformative variables algorithm (UVE) and synergy interval partial least squares algorithm (Si-PLS) were used to screen characteristic variables of coco-peat substrate original spectral data and different pretreatment spectral data. The partial least squares (PLSR) and multiple linear regression (MLR) were used to establish the relationship model between the spectral data and reference measurement value of moisture content. In comparison, the best and simplest spectral prediction model was established when SPA was used to screen the characteristic variables of Savitzky-Golay (S-G) smoothing spectral data and MLR was used to establish the model. And the corresponding correlation coefficient and root mean square error of calibration set were 0.9976 and 1.0989%, respectively; the correlation coefficient and root mean square error of prediction set were 0.9963 and 1.4029%, respectively, and RPD was 11.28. The results of this study provided a feasible method for the rapid detection of moisture content of coco-peat substrate.

A feasibility quantitative analysis of NIR spectroscopy coupled Si-PLS to predict coco-peat available nitrogen from rapid measurements

Available nitrogen was an important index to evaluate the supply capacity of nitrogen fertilizer in planting environment. The study of coco-peat available nitrogen content rapid detection technology was of great significance for instructing scientific fertilization. In this study, near infrared (NIR) spectroscopy was used to realize rapid quantitative detection of available nitrogen in coco-peat. Seven different spectral pretreatment methods were adopted to pre-process spectral data with detection band range of 1000–2500 nm (full-band), and synergy interval partial least squares (Si-PLS) was used to screen the optimum combination sub-intervals reflecting coco-peat available nitrogen content from the original full-band spectral data and various pre-treated spectral data. The spectral prediction models of coco-peat available nitrogen based on full-band spectral data and optimal combined sub-intervals spectral data were respectively established, the improvement effects of different pretreatment methods on the accuracy of coco-peat available nitrogen spectral prediction models were analyzed, and the performances of the optimal full-band spectral prediction model and combined sub-intervals prediction model were compared. By analysis and comparison, the first derivative combined with Savitzky-Golay (S-G) smoothing was used to pre-process spectral data, Si-PLS was used to screen the spectral data of 1724–1784 nm, 1852–1922 nm, 1923–1999 nm and 2175–2272 nm, and then the optimal spectral prediction model of available nitrogen content in coco-peat could be established by using the four bands spectral data. For the optimal model, the correlation coefficient and root mean square error of calibration set were 0.994 and 6.998 mg/100 g respectively, the correlation coefficient and root mean square error of prediction set were 0.993 and 7.390 mg/100 g respectively, and the RPD was 8.062. It showed that the combination of NIR spectroscopy and Si-PLS could realize coco-peat available nitrogen quickly and accurately quantitative detection, Si-PLS could effectively reduce the input variables of the established model and simplify the complexity of the model. It also provided a reference for development of a coco-peat available nitrogen content rapid detection device based on characteristic band sub-intervals spectral data.

A deep learning based regression method on hyperspectral data for rapid prediction of cadmium residue in lettuce leaves

In order to effectively realize the spectral detection of heavy metal content, a deep learning method which consisted of stacked auto-encoders (SAE) and partial least squares support vector machine regression (LSSVR) is proposed to obtain depth features and establish cadmium (Cd) detection model. The Vis-NIR hyperspectral images of 1120 lettuce leaf samples were obtained and the whole region of lettuce leaf sample spectral data was collected and preprocessed with different spectral pre-treatment methods. Successive projections algorithm (SPA), partial least squares regression (PLSR) and SAE were used to acquire the optimum wavelength, respectively. Besides, the characteristic wavelengths were used to build partial least squares support vector machine regression (LSSVR) models. Furthermore, the best prediction performance for detecting Cd content in lettuce leaves was obtained by Savitzky-Golay combined with first derivative (SG-1st) pre-processing method, with Rp2 of 0.9487, RMSEP of 0.01049 ​mg/kg and RPD of 3.330 using SAE-LSSVR method. The results of this study indicated that deep learning method coupled with hyperspectral imaging technique has great potential for detecting heavy metal content in lettuce leaves.

A Rapid and Nondestructive Approach for the Classification of Different-Age Citri Reticulatae Pericarpium Using Portable Near Infrared Spectroscopy.

Citri Reticulatae Pericarpium (CRP), has been used in China for hundreds of years as a functional food and medicine. However, some short-age CRPs are disguised as long-age CRPs by unscrupulous businessmen in order to obtain higher profits. In this paper, a rapid and nondestructive method for the classification of different-age CRPs was established using portable near infrared spectroscopy (NIRS) in diffuse reflectance mode combination with appropriate chemometric methods. The spectra of outer skin and inner capsule of CRPs at different storage ages were obtained directly without destroying the samples. Principal component analysis (PCA) with single and combined spectral pretreatment methods was used for the classification of different-age CRPs. Furthermore, the data were pretreated with the PCA method, and Fisher linear discriminant analysis (FLD) with optimized pretreatment methods was discussed for improving the accuracy of classification. Data pretreatment methods can be used to eliminate the noise and background interference. The classification accuracy of inner capsule is better than that of outer skin data. Furthermore, the best results with 100% prediction accuracy can be obtained with FLD method, even without pretreatment.

Estimating Soil Arsenic Content with Visible and Near-Infrared Hyperspectral Reflectance

Soil arsenic (AS) contamination has attracted a great deal of attention because of its detrimental effects on environments and humans. AS and inorganic AS compounds have been classified as a class of carcinogens by the World Health Organization. In order to select a high-precision method for predicting the soil AS content using hyperspectral techniques, we collected 90 soil samples from six different land use types to obtain the soil AS content by chemical analysis and hyperspectral data based on an indoor hyperspectral experiment. A partial least squares regression (PLSR), a support vector regression (SVR), and a back propagation neural network (BPNN) were used to establish a relationship between the hyperspectral and the soil AS content to predict the soil AS content. In addition, the feasibility and modeling accuracy of different interval spectral resampling, different spectral pretreatment methods, feature bands, and full-band were compared and discussed to explore the best inversion method for estimating soil AS content by hyperspectral. The results show that 10 nm + second derivative (SD) + BPNN is the optimum method to predict soil AS content estimation; R v 2 is 0.846 and residual predictive deviation (RPD) is 2.536. These results can expand the representativeness and practicability of the model to a certain extent and provide a scientific basis and technical reference for soil pollution monitoring.

Identification of Rice Varieties and Transgenic Characteristics Based on Near-Infrared Diffuse Reflectance Spectroscopy and Chemometrics.

Background: In recent years, genetically modified technology has developed rapidly, and the potential impact of genetically modified foods on human health and the ecological environment has received increasing attention. The currently used methods for testing genetically modified foods are cumbersome, time-consuming, and expensive. This paper proposed a more efficient and convenient detection method. Methods: Near-infrared diffuse reflectance spectroscopy (NIRDRS) combined with multivariate calibration methods, including principal component analysis (PCA), partial least squares discriminant analysis (PLS-DA), and support vector machines (SVM), were used for identification of different rice varieties and transgenic (Bt63)/non-transgenic rice. Spectral pretreatment methods, including Norris–Williams smooth (NWS), standard normal variate (SNV), multiplicative scatter correction (MSC), and Savitzky–Golay 1st derivative (SG 1st-Der), were used for spectral noise reduction and effective information enhancement. Accuracy was used to evaluate the qualitative discriminant models. Results: The results showed that the SG 1st-Der pretreatment method, combined with the SVM, provided the optimal model to distinguish different rice varieties. The accuracy of the optimal model was 98.33%. For the discrimination model of transgenic/non-transgenic rice, the SNV-SVM model, MSC-SVM model, and SG 1st-Der-PLS-DA model all achieved good analysis results with the accuracy of 100%. Conclusion: The results showed that portable NIR spectroscopy combined with chemometrics methods could be used to identify rice varieties and transgenic characteristics (Bt63) due to its fast, non-destructive, and accurate advantages.

Prediction Model of the Key Components for Lodging Resistance in Rapeseed Stalk Using Near-Infrared Reflectance Spectroscopy (NIRS)

The chemical composition of rape stalk is the physiological basis for its lodging resistance. By taking the advantage of NIRS, we developed a rapid method to determine the content of six key composition without crushing the stalk. Rapeseed stalks in the mature stage of growth were collected from three cultivation modes over the course of 2 years. First, we used the near-infrared spectroscope to scan seven positions on the stalk samples and took their average to form the spectral data. The stalks were then crushed and sieved; then the ratio of carbon and nitrogen, ratio of acid-insoluble lignin and lignin, and the content of soluble sugar and cellulose were determined using the combustion method, weighing method, and colorimetric method, respectively. The partial least squares regression (PLSR) method was used to establish a prediction model between the spectral data and the chemical measurements, and all models were evaluated by an internal interaction verification and an external independent test set sample. To improve the accuracy of the model and reduce the computing time, some optimization methods have been applied. Some outliers were removed, and then the data were preprocessed to determine the best spectral information band and the optimal principal component number. The results showed that elimination of outliers effectively improved the precision of the prediction model and that no spectral pretreatment method exhibited the highest prediction accuracy. In summary, the NIRS-based prediction model could facilitate the rapid nondestructive detection in the key components of rapeseed stalk.

Spectral imaging application to discriminate different diets of live rainbow trout (Oncorhynchus mykiss)

The main aim of this study was to evaluate the feasibility of hyperspectral imagery for determining the correlation between fish skin changes and different diets. Rainbow trout (Oncorhynchus mykiss) were fed either a commercial based diet (N = 80) or a 100% plant-based diet (N = 80). Hyperspectral images were made using a push-broom hyperspectral imaging system in the spectral region of 394–1009 nm. All images were calibrated using dark and white reference, and the average spectral data from the region of interest were extracted. Seven spectral pre-treatment methods were used, including Savitzky-Golay (SG), First Derivative (FD), Second Derivative (SD), Standard Normal Variate (SNV), Multiplicative Scatter Correction(MSC) and Continuum removal (CR) then a support vector machine (SVM) with linear kernel was applied to establish the classification models. Overall classification models developed from full wavelengths with different preprocessing methods showed good performance (Correct Classification Rate (CCR) = 0.83, Kappa = 0.66) when coupled with SG and SD or SG and MSC. The overall results indicate that the integration of Vis/NIR hyperspectral imaging system and machine learning algorithms have promise for discriminating different diets based on the live fish skin. These procedures can be used to not only identify the diet used for fish feeding in the case where we are not sure but also monitor different diets impacts on live fish skin for more precise monitoring of fish status during cultivation and ultimately for better implementation of precision fish farming.

Development of a methodology to analyze leaves from Prunus dulcis varieties using near infrared spectroscopy

Near-infrared spectroscopy (NIRS) can be a faster and more economical alternative to traditional methods for screening varietal mixtures of nursery plants during the propagation process to ensure varietal purity and to avoid errors in the dispatch batches. The global objective of this work was to develop and optimize a NIR spectral collection method for construction of robust multivariate discrimination models. Three different varieties of Prunus dulcis (Avijor, Guara, and Pentacebas) of agricultural interest were used for this study. Sources of variation were investigated, including the position of the leaves on the trees, differences among trees of the same variety, and differences at the varietal level. Three types of processed samples were investigated. Fresh leaves, dried leaves, and dried leaves in powder form were included in each analysis. A study of spectral pre-treatment methods was also performed, and multivariate methods were applied to analyze the influence of different factors on classification. These included principal component analysis (PCA), partial least squares discriminant analysis (PLS-DA), and ANOVA simultaneous component analysis (ASCA). The results indicated that variety was the most important factor for classification. The spectral pre-treatment that provided the best results was a combination of standard normal variate (SNV), Savitzky-Golay first derivative, and mean-centering methods. With regard to the type of processed sample, the highest percentages of correct classifications were obtained with fresh and dried powdered leaves at both the training set and test set validation levels. This study represents the first step towards the consolidation of NIRS as a method to identify Prunus dulcis varieties.

锌溶液中痕量Cu2+、Co2+的检测光谱预处理方法

锌溶液中痕量Cu2+、Co2+的检测光谱预处理方法