The qualitative and quantitative determination of the components of textile fibers takes an important position in quality control. A fast and nondestructive method of simultaneously analyzing four fiber components in blended fabrics was studied by nearinfrared (NIR) spectroscopy combined with multivariate calibration. Two sample sets including 39 and 25 samples were designed by simplex mixture lattice design methods and used for experiment. Four components include wool, polyester, polyacrylonitrile, and nylon and their mixture is one of the most popular formulas of textiles. Uninformative variable eliminationpartial least squares (UVEPLS) and the fullspectrum partial least squares (PLS) were used as the tool. On the test set, the mean standard error of prediction (SEP) and the mean ratio of the standard deviation of the response variable and SEP (RPD) of the fullspectrum PLS model and UVEPLS model were 0.38, 0.32 and 7.6, 8.3, respectively. This result reveals that the UVEPLS can construct local models with acceptable and better performance than the fullspectrum PLS. It indicates that this method is valuable for nondestructive analysis in the field of wool content detection since it can avoid timeconsuming, costly, and laborious wet chemical analysis.
To blend fibers of different types is a common practice to obtain expected characteristics in the textile industry. According to the national standard of China, textile products have to be marked with fabric type and composition on the label. Also, this quantitative composition is mandatory information [
Especially, nearinfrared (NIR) spectroscopy has shown great potential and gained wide acceptance in food industry [
In NIRbased quantitative applications, a reliable calibration model is of great importance and its predictive performance even directly determines its availability [
In the present work, a fast and nondestructive method of simultaneously analyzing four components in blended fabrics was studied by nearinfrared (NIR) spectroscopy combined with multivariate calibration. Two sample sets including 39 and 25 samples were designed by simplex mixture lattice design methods and used as the training set and the independent test set, respectively. Four components include wool, polyester, polyacrylonitrile, and nylon and represent one of the most popular formulas of textiles. Uninformative variable eliminationpartial least squares (UVEPLS) and the fullspectrum partial least squares (PLS) were used as the tool of variable selection and multivariate calibration. This result reveals that the UVEPLS can construct local models with acceptable and better performance than the fullspectrum PLS. It indicates that this method can serve as a tool of fast and nondestructive analysis of fiber contents since it can avoid timeconsuming, costly, and laborious wet chemical analysis.
Partial least squares (PLS) [
Uninformative variable elimination (UVE) is a classic method of variable selection by analyzing the stability of the regression coefficient [
First PLS regression is made on instrumental signal matrix (
Then a noise matrix with an approximate size is generated and its elements are random numbers in the interval of 01. And the elements are multiplied by a small constant so as to make their influence on the model negligible. Such a noise matrix is appended to the original signal matrix to form an extended matrix.
PLS models are constructed on the extended matrix (
The reliability of each variable is quantitatively measured by the stability value, which is defined as the mean of the corresponding column divided by the standard deviation of that column in the matrix of regression coefficients.
Based on the fact that any variable with less stability than a random variable is uninformative and should be eliminated, a cutoff value is calculated as the maximum of the stability values among the random variables. Every original variable with lower stability values than the cutoff value is assumed to contain nothing but noise and is therefore eliminated.
Based on the remaining variables, a final PLS model can be constructed and optimized.
This work used the simplex lattice design for preparing the fourcomponent mixture samples. For an fourcomponent system, the regular simplex is a tetrahedron where each vertex represents a straight component, an edge represents a binary system, and a face represents a ternary one. Points inside the tetrahedron correspond to quaternary systems. The basis of designing experiments of this kind is a uniform scatter of experimental points on the socalled simplex lattice. Points, or design points, form a [
Mixture simplex lattice design of four components for training and test sets.
No  Training set lattice design  Test set lattice design  

A  B  C  D  A  B  C  D  
1  0.75  0.25  0  0  0  1  0  0 
2  0.25  0.5  0.25  0  0  0  0  1 
3  0.75  0  0  0.25  0  0  1  0 
4  0.5  0.25  0  0.25  0.667  0  0.333  0 
5  0  0  1  0  0  0.667  0  0.333 
6  0  0  0  1  0  0.333  0.333  0.333 
7  0  0  0.75  0.25  0  0.667  0.333  0 
8  0.125  0.125  0.125  0.625  0.25  0.25  0.25  0.25 
9  0  0.75  0.25  0  0.667  0  0  0.333 
10  0.625  0.125  0.125  0.125  0.333  0  0.333  0.333 
11  0.25  0.25  0.25  0.25  0  0.333  0.667  0 
12  0.25  0  0.75  0  1  0  0  0 
13  0.25  0.25  0.5  0  0.333  0.333  0.333  0 
14  0  0.25  0.75  0  0  0  0.667  0.333 
15  0  0.75  0  0.25  0.333  0.333  0  0.333 
16  0  0.5  0.25  0.25  0.333  0  0  0.667 
17  0.5  0.5  0  0  0  0.333  0  0.667 
18  0  0  0.25  0.75  0.125  0.125  0.625  0.125 
19  0  0.25  0.5  0.25  0.125  0.125  0.125  0.625 
20  0.25  0.75  0  0  0.333  0  0.667  0 
21  0.5  0  0.25  0.25  0.333  0.667  0  0 
22  0.125  0.625  0.125  0.125  0  0  0.333  0.667 
23  0.25  0  0  0.75  0.625  0.125  0.125  0.125 
24  1  0  0  0  0.125  0.625  0.125  0.125 
25  0.5  0  0.5  0  0.667  0.333  0  0 
26  0.25  0.25  0  0.5  
27  0  0.5  0  0.5  
28  0.75  0  0.25  0  
29  0  0  0.5  0.5  
30  0.5  0.25  0.25  0  
31  0  0.25  0.25  0.5  
32  0  0.25  0  0.75  
33  0.125  0.125  0.625  0.125  
34  0.5  0  0  0.5  
35  0.25  0.5  0  0.25  
36  0  1  0  0  
37  0.25  0  0.25  0.5  
38  0.25  0  0.5  0.25  
39  0  0.5  0.5  0 
The instrument used in this study was an Antaris II Fourier transform nearinfrared (FTNIR) spectrometer, which is equipped with an integrating sphere, an InGaAs detector, and a tungsten lamp as the light source. When collecting a spectrum, the mixed sample was poured into a rotatable sample cup with a 50 mm diameter, and the stacking height was controlled above 10 mm for preventing light leak. An internal gold reference was used for background collection. The rotation cup allows multipoint diffuse reflection measurements for the same sample. So, a final spectrum is actually the mean of the spectra measured at different locations. All NIR spectra were recorded in the region of 4000–10000 cm^{−1} with 32 coadded scans. The resolution was set as 3.856 cm^{−1}, and each spectrum consisted of 1557 data points. Taking into account the uniformity problem of solid samples, two spectra were recorded for each sample. Thus, a total of 78 spectra and 50 spectra were obtained for the training set and the test set, respectively. The experimental temperature and the related humidity were controlled at 25°C and 60%, respectively. To remove the influence of light scattering and pathlength variations on the spectra, standard normal transformation (SNV) was first used to preprocess all original spectra. That is, each spectrum was corrected individually by centered and scaled by its standard deviation [
Given a dataset, in general, the partition of available samples into a representative training/calibration set and a test set is of great importance. The training set is used to construct a calibration model while the test set is used to evaluate its performance. Theoretically, the evaluation is valid only when the test set has the same distribution as the training set. In this work, the socalled simplex lattice design was used to generate two independent sample sets, one for training and the other for test. As Table
The concentration of four fibers in the designed experimental sample set.
Actually, each spectrum was composed of 1557 data points, and its profile depended on the sample components. That is, the weight percentage of each component in each sample was associated with the respective spectrum. The NIR spectrum reflects composition variation of a sample. Pure NIR spectra of these fabrics have different profiles despite the difference being very small. Figure
Nearinfrared spectra of the training and test sets.
Besides the chemical composition of the samples, the NIR spectra of a mixture can be affected by other factors such as light scattering, baseline shift, and pathlength variations from heterogeneity. These situations will seriously complicate the calibration task. To reduce these effects, as described above, all spectra were first preprocessed by SNV and followed by calculating the first derivative. The preprocessed spectra were ready for subsequent variable selection and calibration.
Based on the training set, both fullspectrum PLS and UVEPLS methods were used for multivariate calibration. Actually, the latter includes the variable selection. When using UVEPLS, the maximum number of components, the number of random variables, the fold number of crossvalidation, and the cutoff level were set as 15, 300, 5, and 0.99, respectively, by trial and error. Figure
The selected variables for quantifying four fibers by uninformative variable eliminationpartial least squares (UVEPLS).
The key parameter in the PLS model is the number of components. Based on 5fold crossvalidation, a series of candidate PLS models with different number of components were constructed. Figure
Root mean squared error of crossvalidation (RMSECV) versus the number of components for the fullspectrum model and the model with selected variables.
Based on the optimal number of PLS components, final calibration models were constructed. Figures
The predicted versus the actual values of fiber concentrations based on the fullspectrum model. (a) Wool, (b) polyester, (c) polyacrylonitrile, and (d) nylon.
The predicted versus the actual values of fiber concentrations based on the local model with selected variables. (a) Wool, (b) polyester, (c) polyacrylonitrile, and (d) nylon.
Comparison of two kinds of calibration models based on three indices.
Method  Index  A  B  C  D  Mean 

PLS (fullspectrum)  SEC  0.034  0.041  0.050  0.026  0.038 
SEP  0.056  0.036  0.035  0.026  0.038  
RPD  4.8  7.4  7.7  10.5  7.6  


UVEPLS (local)  SEC  0.026  0.031  0.037  0.027  0.030 
SEP  0.034  0.031  0.034  0.031  0.032  
RPD  8.0  8.7  7.9  8.7  8.3 
From Table
This paper demonstrated a NIR spectroscopic method for simultaneously determining the fiber contents of fourcomponent blends. Simplex lattice design was used to produce two independent sample sets for the training and test purposes, respectively. As a result, the UVEPLS algorithm construct local PLS models with satisfactory performance compared to the fullspectrum PLS algorithm. This procedure is simple, fast, and environmentfriendly and has potential for quality control of textile products. The main challenge is maybe heterogeneous nature of textiles since it is difficult to blend whole fiber samples uniformly. Even so, it is still valuable and can serve as an alterative to some wet chemical methods for similar tasks.
The data used to support the findings of this study are available from the corresponding author upon request.
The authors declare that there are no conflicts of interest regarding the publication of this paper.
This work was supported by the National Natural Science Foundation of China (21375118, J1310041), Scientific Research Foundation of Sichuan Provincial Education Department of China (17TD0048), Scientific Research Foundation of Yibin University (2017ZD05), Sichuan Science and Technology Program of China (2018JY0504), and Opening Fund of Key Lab of Process Analysis and Control of Sichuan Universities of China (2018005).