Multiplicative Scatter Correction Method Research Articles

Study on detecting main ingredients of silicone rubber based on terahertz spectrum

AbstractThe authors investigated the ingredient detection technique of silicone rubber based on the Terahertz spectrum. For this purpose, 18 diverse high‐temperature vulcanised silicone rubber (HTVSR) formulations were customised, 8 of which are used as calibration set while the rest 10 as prediction set. Based on the Beer‐Lambert Law, the partial‐least‐square (PLS) regression model and the least‐squares support‐vector machines (LS‐SVM) regression model were used to yield the relationships between the absorption spectrums and the content percentages of polydimethylsiloxane (PDMS), alumina trihydrate (ATH), and silica in HTVSR. The results showed that for the formulations tested, the prediction accuracy of all three main ingredients by the PLS regression model could be improved by changing the spectrum range from 0.2–4 to 0.5–2 THz. If the data were pre‐processed by the Savitzky–Golay smoothing method or multiplicative scatter correction method, the prediction accuracy of PDMS could be further enhanced. However, this would lead to a slight decrease in the prediction accuracy of ATH. For the LS‐SVM regression model, the radial basis function (RBF) kernel and the linear kernel were studied. It was found that the prediction accuracy of both kernels was better than that of the PLS regression model. With the LS‐SVM regression model using the RBF kernel, the correlated coefficients of PDMS and ATH in the prediction set could be up to 0.9915 and 0.9742, respectively.

Near Infrared Technology Coupled with Different Spectra Correction Approaches for Fast and Non-Destructive Prediction of Chlorogenic Acid on Intact Coffee Beans

Abstract The primary objective of this research was to utilise near-infrared reflectance spectroscopy as a swift, non-destructive method for identifying chlorogenic acid in whole coffee beans. Additionally, this investigation explored the efficacy of different spectral improvement techniques alongside partial least square regression to construct predictive models. NIR spectral data was gleaned from whole coffee beans spanning a wavelength range of 1000–2500 nm, while the chlorogenic acid content was ascertained via high-performance liquid chromatography procedures. Our findings revealed that the highest coefficient of determination reached for chlorogenic acid was 0.97, and the root mean square error for calibration was 0.31% when using the multiplicative scatter correction method. Furthermore, upon testing the model using an external validation dataset, a determination coefficient of 0.91 and a ratio error to range index of 11.56 with a root mean square prediction error at 0.51% was attained. From these results, it can be inferred that the near-infrared technology, coupled with an effective spectral enhancement process, can facilitate quick, non-invasive determination of chlorogenic acid in whole coffee beans.

The synergic approach between machine learning, chemometrics, and NIR hyperspectral imagery for a real-time, reliable, and accurate prediction of mass loss in cement samples

Alternative and non-destructive analytical methods that predict analyte concentration accurately and immediately in a specific matrix are becoming vital in the analytical chemistry domain. Here, a new innovative and rapid method of predicting mass loss of cement samples based on a combination of Machine Learning (ML) and the emerging technique called Hyperspectral Imaging (HSI) is presented. The method has proved its reliability and accuracy by providing a predictive ML model, with satisfactory best validation scores recorded using partial least squared regression, with a reported ratio of performance to inter-quartile distance and root mean squared error of 12,89 and 0.337, respectively. Moreover, the possibility of optimizing and boosting the performance of the method by optimizing the predictive model performance has been suggested. Therefore, a features selection approach was conducted to disqualify non-relevant wavelengths and stress only relevant ones in order to make them the only contributors to a final optimized model. The best selected features subset was composed of 28 wavelengths out of 121, found by applying genetic algorithm combined to partial least squares regression as a feature selection method, on spectra preprocessed consecutively by the first-order savitzky-golay derivative calculated with 7-point quadratic SG filter, and multiplicative scatter correction method. The overall results show the possibility of combining HSI and ML for fast monitoring of water content in cement samples.

Application of NIRS adaptive technology for prediction of in vitro digestibility parameters of feed ingredients from Fermented Cocoa Pods

Near Infrared Reflectance Spectroscopy (NIRS) is an alternative method that can be applied in the evaluation of feed ingredients. The purpose of this study was to evaluate the in vitro digestibility value of fermented cocoa pods using the NIRS method of feed analysis. This study used 18 samples of fermented cocoa shell waste, where the material used was fermented cocoa pod skin (KBK) then in vitro analysis was carried out. Samples were acquired spectrum by Thermo Integrative. Chemical analysis is intended to compare the data from chemical analysis and the NIRS method. The parameters analyzed were pH, dry matter digestibility (KCBK), and organic matter digestibility (KCBO). Spectrum data were processed using PLS with the pre-treatment methods of multiplicative scatter correction (MSC) and DeTrending (DT). Based on the analysis that has been carried out using the NIRS method with a prediction model that has been built on pH parameters, KCBK has a good predictive model with the DT pre-treatment method and KCBO has a good predictive model with the MSC method where (pH = LV: 8, r: 0 .79, R2: 0.62, RMSEC: 0.04 and RPD: 2.00; KCBK= LV: 8, r: 0.86, R2: 0.74, RMSEC: 1.30 and RPD: 2.02 ; KCBO= LV: 8, r: 0.88, R2: 0.78, RMSEC: 1.39 and RPD: 2.21). Rough predictions for pH were obtained with Non pre-treatment and MSC pre-treatment (pH and MSC = LV: 8, r: 0.74, R2: 0.55, RMSEC: 0.05, RPD 1.58) and rough predictions on KCBK were obtained with Non Preatment (KCBK=LV: 8, r: 0.83 R2:0.70, RMSEC: 1.30, and RPD 1.87).

Hyperspectral imaging technigue for monitoring moisture content of blueberry during the drying process

Changes in the moisture content (MC) of blueberries during drying was monitored by hyperspectral image analysis, and the degree of drying was determined using the partial least squares (PLS) model. Blueberries (n=820) were dried at 35°C for 0 (control), 3, 6, 9 and 12 days. The PLS discriminant analysis prediction accuracy of smoothing the pre-processed data was the highest. Regression coefficients were high at 706, 790, 827, 868, and 894 nm, corresponding to water molecules and carbohydrates (830-840 nm). To develop a prediction model for blueberry MC, 150 hyperspectral images were obtained from 30 samples per group. The MC of each group was also analyzed. The accuracy of the MC prediction model pretreated by the multiplicative scatter correction method was the highest at 0.9302. As indicated by Pearson’s correlation analysis, the blueberry MC showed a high correlation of 0.95 with the total soluble solid contents, brightness, and total flavonoid contents. These results suggest that hyperspectral imaging techniques can be used to predict and monitor various quality characteristics as well as the MC of blueberries during drying.

Detection of coconut (Cocos nucivera) sugar adulteration in palm (Arenga pinnata Merrill) sugar by Fourier Transform Infrared (FT-IR) Spectroscopy

Palm sugar which is also named brown sugar is powdered sugar produced from palm extract. Due to the high price of palm sugar, its contamination of materials that are cheap or low quality is inevitable. Usually, adulteration detection is done by conventional methods such as HPLC, TLC, or NMR which are time-consuming and require high-priced equipment, thus impractical for routine and large sample analysis. The aim of this research was to detect adulteration in palm sugar using Fourier Transform Infrared (FT-IR) spectroscopy. The samples used in this study were palm sugar as the main ingredient and coconut sugar as the adulterant. Two chemometric methods namely principal component analysis (PCA) and partial least squares regression (PLSR) were used for analysis. The absorbance data were taken at wavenumber 4000-650 cm-1 . Several concentrations of coconut sugar as an adulterant ranging from 0 to 100% were added to palm sugar. A total of 110 spectra of both pure and adulterated palm sugar samples were divided into two groups, i.e. 73 samples for developing calibration model and 37 samples for developing prediction model. The spectral obtained were pre-processed and analyzed using The Unscrambler X version 10.4. a total of six pre-processing methods were used, i.e., Normalization, Standard Normal Variate (SNV), Multiplicative Scatter Correction (MSC), and Baseline. Results showed that PCA was able to classify palm sugar based on adulterant concentrations. PLSR calibration model with a coefficient of determination (Rc2 ) of 0.94 and root mean square error of calibration (RMSEC) of 8% was obtained by applying the MSC method. The model was able to predict coconut sugar adulteration in palm sugar with Rp2 of 0.89 and root mean square error of prediction (RMSEP) of 10.68%. The results confirmed the potential of FT-IR spectroscopy for detecting adulteration in palm sugar.

Alternative particle compensation techniques for online water quality monitoring using UV–Vis spectrophotometer

Particles in the water can significantly affect UV–Vis absorption measurements. There is a need for the water industry to develop a reliable technique to eliminate particle impact on on-line water quality monitoring using UV–Vis spectroscopy. This study aims to develop and use digital techniques for particle compensation: single wavelength compensation, linear regression compensation and multiplicative scatter correction method for on-line UV–Vis spectrophotometers. Water quality data were collected from three selected water sources in water treatment plants which represent different water qualities in terms of particles and organic matters. UV254 measurements were determined with these three software compensation techniques in comparison with the proprietary instrument built-in compensation algorithm using Bland-Altman analysis. Linear correction methods were found to be able to adjust the three compensation techniques to achieve acceptable compensated UV254 results, particularly for raw waters. UV254 measurements using single wavelength compensation, linear regression compensation and multiplicative scatter correction techniques with the assistant of linear correction methods were confirmed to be comparable to the instrument built-in compensation method. Our results reveal that these particle compensation techniques can make the UV254 technology reliable for online water quality monitoring in water treatment network. This paper demonstrated the advantage of using software compensation method to establish local compensation and calibration models instead of relying on the predetermined global calibrations for online water quality monitoring.

Rapid Nondestructive Detection of Water Content and Granulation in Postharvest "Shatian" Pomelo Using Visible/Near-Infrared Spectroscopy.

Visible/near-infrared (VIS/NIR) spectroscopy is a powerful tool for rapid, nondestructive fruit quality detection. This technology has been widely applied for quality detection of small, thin-peeled fruit, though less so for large, thick-peeled fruit due to a weak spectral signal resulting in a reduction of accuracy. More modeling work should be focused on solving this problem. “Shatian” pomelo is a traditional Chinese large, thick-peeled fruit, and granulation and water loss are two major internal quality factors that influence its storage quality. However, there is no efficient, nondestructive detection method for measuring these factors. Thus, the VIS/NIR spectral signal detection of 120 pomelo samples during storage was performed. Information mining (singular sample elimination, data processing, feature extraction) and modeling were performed in different ways to construct the optimal method for achieving an accurate detection. Our results showed that the water content of postharvest pomelo was optimally detected using the Savitzky–Golay method (SG) plus the multiplicative scatter correction method (MSC) for data processing, genetic algorithm (GA) for feature extraction, and partial least squares regression (PLSR) for modeling (the coefficient of determination and root mean squared error of the validation set were 0.712 and 0.0488, respectively). Granulation degree was best detected using SG for data processing and PLSR for modeling (the detection accuracy of the validation set was 100%). Additionally, our research showed a weak relationship between the pomelo water content and granulation degree, which provided a reference for the existing debates. Therefore, our results demonstrated that VIS/NIR combined with optimal information mining and modeling methodswas feasible for determining the water content and granulation degree of postharvest pomelo, and for providing references for the nondestructive internal quality detection of other large, thick-peeled fruits.

Investigating aquaphotomics for temperature-independent prediction of soluble solids content of pure apple juice

The methods of aquaphotomics were explored as an aid to improve near infrared spectroscopic predictive modelling of the soluble solids content of pure apple juice at different temperatures. The study focussed on the first overtone region of the O–H stretching vibration of water (1300–1600 nm). A transmission-based FT-NIR (Fourier transform near infrared) spectrometer was used to acquire 103 spectra of freshly expressed juice samples from individual ‘Braeburn’ apples over the wavelength range of 870–1800 nm with a 1 mm cuvette at three temperatures, 20, 25 and 30°C. The aquagram of the first overtone water region showed a trend of increasing bound water absorption with rising soluble solids content, from 7.3 to 13.7°Brix, and increasing free water absorption with rising temperature from 20 to 30°C. Predictive models for apple juice soluble solids content at 25°C were developed using partial least squares regression with spectral pre-processing by standard normal variate (SNV) followed by second derivative transformation (SNV + 2D) or no pre-processing on absorbance spectra at all. The best result, with lowest standard error of prediction of 0.38°Brix, was obtained using the first overtone water region with partial least squares regression on the SNV + 2D spectra. The method of extended multiplicative scatter correction was used, as an additional pre-processing step, to improve apple juice soluble solids content prediction at different temperatures. The interference component selected for the extended multiplicative scatter correction method was the first principal component loading measured using pure water samples taken at the same three temperatures (20, 25 and 30°C). Such extended multiplicative scatter correction pre-processing greatly reduced the soluble solids content prediction bias, when applying the partial least squares regression model developed at 20°C to samples measured at 25 and 30°C, from 0.23 to 0.08 and 0.36 to 0.13°Brix, respectively. Model precision (in terms of standard error of prediction) was also slightly improved by 0.02°Brix in each case, from 0.40 to 0.38 and 0.46 to 0.44°Brix at 25 and 30°C respectively.

Non-Destructive Classification of Fruits Based on Vis-nir Spectroscopy and Principal Component Analysis

Fruits are one of the sources of nutrition needed for health. Fruit quality is generally assessed by physical and chemical properties. Measurement of fruit internal quality is usually done by destructive techniques. Ultraviolet, visible and near-infrared (UV-Vis-NIR) spec-troscopy is a non-destructive technique to measure fruit quality. This technique can rapidly measure the fruit quality, the measured fruit still remains intact, and can be marketed. Besides, UV-Vis-NIR spectrosco-py can also be used to classify fruits. The study aimed to classify var-ious types of fruits using UV-Vis-NIR spectroscopy with wavelengths of 300-1041 nm and Principal Component Analysis (PCA). First de-rivative savitzky-golay with 9 smoothing points (dg1) and multiplica-tive scatter correction (MSC) were applied to correct the spectra. The results showed that the use of uv-vis-nir spectroscopy and PCA com-bined with spectra pre-treatment of the MSC method were able to clas-sify various types of fruits with 100% success rate in all fruit samples including sapodilla, ridge gourd, mango, guava, apple and zucchini. 

Perbandingan Metode Koreksi Pencaran pada Data Hasil Alat Pemantau Kadar Glukosa Darah Non-Invasif

Scatter correction is one of the methods in data preprocessing that aim at eliminating the physical properties of the spectrum and reducing the variance between samples. The most commonly methods of scatter correction used are the Multiplicative Scatter Correction (MSC) and Standard Normal Variate (SNV) methods. The MSC method corrects the spectrum by utilizing the results of simple linear regression parameter estimation. The SNV method performs spectral correction with the median and standard deviation. Another alternative method of scatter correction is the Orthogonal Scatter Correction (OSC) applying the principle of orthogonality. The methods  used in this research were MSC, SNV, and OSC methods in order to correct the result data of non-invasive blood glucose measuring instrument. The result of this research showed that the time domain spectrum data and intensity had different amount so that the summarized data was needed. Furthermore, this research found that the OSC method with the five series of statistics gained a good correction result compared to the other methods. The OSC method produced a smaller average value of the variance than the other methods.

基于近红外波长组合快速检测油页岩含油率

To detect rapidly the oil yield of an oil shale on a single point measurement,a detection method was proposed based on combination of wavelengths by aportable near infrared spectroscopy.By taking the kaolin mixed with oil simulating samples as the research object,how to select the combination of wavelengths by near infrared spectroscopy was performed.With proposed method,the full spectral reflectivity data were firstly acquired by the developed portable spectrometer.Then the preliminary selection for the combination of wavelengths was implemented by differential multiplicative scatter correction method and correlation coefficient method,and the optimum combination of wavelengths was determined by combination generation method and leave one cross validation Multiple Linear Regression(MLR).Finally,the data for the optimum combination of wavelengths were verified by MLR modeling.The results show that the optimum combination of wavelengths is 1 644,1 720,2 210 and 2 260 nm,and the determination coefficients for calibration set(R2c),prediction set(R2p)of30modeling samples and validation(R2p)of 11 validation samples are 0.995 4,0.997 7and 0.990 1,respectively.The method provides a basis for design of a rapid detecting spectrometer in oil yields of oil shapes.

Identification of Liquor Brands Based on near Infrared Spectroscopy

In this paper, the liquor brands were identified based on the near infrared spectroscopy method and the principal component analysis. 60 samples of 6 different brands liquor were measured by the spectrometer of USB4000. Then, in order to eliminate the noise caused by the external factors, the smoothing method and the multiplicative scatter correction method were used. After the preprocessing, we got the revised spectra of the 60 samples. The difference of the spectrum shape of different brands is not much enough to classify them. So the principal component analysis was applied for further analysis. The results showed that the first two principal components variance contribution rate had reached 99.06%, which can effectively represent the information of the spectrums after preprocessing. From the scatter plot of the two principal components, the 6 different brands of liquor were identified more accurate and easier than the spectra curves.

Classification of Biological Spectrum Based on Principal Component Cluster Analysis

Spectrums of 17 biological tissue phantoms were measured using the fiber-optic spectrometer. Then, the spectrum was preprocessed by multiplicative scatter correction method to devoice the spectrum. Afterwards the features of the spectrum were extracted via principal component analysis. Ultimately, we applied cluster analysis for the spectral features. The results showed that the accumulated credibility of the first 12 spectral principal components was 99.86% for the spectrum after preprocessing; indicating that this spectrum feature extraction might be done in the case of losing no key information. And the results showed that the 17 biological tissue phantoms can be divided into four main categories according their optical features.

Research on Quantitative Prediction of Total Ash in Pollen Typhae by NIRS

Near infrared spectroscopy is used for establishing a rapid and non-destructive analytical method to determine the quantitative of Total Ash in Pollen Typhae, a traditional chinese crude drug. Partial least squares regression (PLSR) was used for building the quantitative prediction model of Pollen Typhae different calibration methods were compared with spectral pretreated methods, wave-number selection and factors of NIRS were discussed in detail. The first derivative and multiplicative scatter correction method was used to predict total ash to correlation coefficients (R), root mean square error of calibration (RMESECV) residual prediction deviation (RPD) and offset obtained by PLSR model were 0.9994, 0.293, 29.4 and 0.013 respectively. The standard deviation was less than 8.2% by the total ash prediction model in the 8 batches of Pollen Typhae from 6% to 27.3%. The predictive result shows the method of this research is rapid, non-destructive and credible, which can be applied to the rapid predict the total ash of Pollen Typhae.

Related versions of the multiplicative scatter correction method for preprocessing spectroscopic data

Various multiplicative transformations of spectral variables have been used with some success as preprocessing methods for diffuse near-infrared spectroscopy data. We discuss first which additive/multiplicative transformations conserve the area under the spectral curve. Next we look at the relationship between the Multiplicative Scatter and the Standard Normal Variate transformations, and use a minimizing norm formulation to isolate three specific transformations that seem to be singled out as particularly interesting. In an empirical investigation using 8 different data sets, each with several constituents, the different transformation methods are compared.

Related versions of the multiplicative scatter correction method for preprocessing spectroscopic data

Various multiplicative transformations of spectral variables have been used with some success as preprocessing methods for diffuse near-infrared spectroscopy data. We discuss first which additive/multiplicative transformations conserve the area under the spectral curve. Next we look at the relationship between the Multiplicative Scatter and the Standard Normal Variate transformations, and use a minimizing norm formulation to isolate three specific transformations that seem to be singled out as particularly interesting. In an empirical investigation using 8 different data sets, each with several constituents, the different transformation methods are compared.

A Pathlength Correction Method for Near-Infrared Spectroscopy

A method for correction of pathlength effects in near-infrared (NIR) spectra is presented. This method, called Pathlength Correction with Chemical Modeling (PLC-MC), requires the use of standards in which the chemistry varies but the pathlength remains constant. Principal Components Analysis (PCA) is used to model the chemical variations in the spectra of the standards. The PCA model can then be used to estimate the pathlength of a future sample from the spectrum of the sample. Theoretical and experimental comparisons of the PLC-MC correction method with the Multiplicative Scatter Correction (MSC) method indicate that the PLC-MC method is more accurate than the MSC method if spectral variability from chemical variations is large.