Identification of varieties of natural textile fiber based on Vis/NIR spectroscopy technology

A new method for discriminating the varieties of natural textile fiber based on visible/near infrared spectroscopy (Vis/NIRS) was developed. In order to achieve the rapid identification of the varieties of natural fiber, four kinds of fiber of cotton, flax, silk and cashmere were selected for analysis. Firstly, the spectra with wavelength 350-1800nm of each variety fiber were scanned by spectrometer, principal component analysis (PCA) method were used to analyze the characteristics of the pattern of Vis/NIR spectra. Principal component scores scatter plot (PCI PC2 PC3) of fiber indicated that the classification effect of four varieties of fibers. The former 6 principal components (PCs) were selected according with the quantity and size of PCs. The PCA classification model was optimized by using the least-squares support vector machines (LS-SVM) method. We use the 6 PCs extracted by PCA as the inputs of LS-SVM, PCA-LS-SVM model was built to achieve varieties validation as well as mathematical model building and optimization analysis. 180 samples (45 samples for each variety of fibers) of four varieties of fibers were used for calibration of PCA-LS-SVM model, and the other 60 samples (15 samples for each variety of fibers) were used for validation, the result of validation show that Vis/NIR spectroscopy technique combined with PCA-LS-SVM had a powerful classification capability. It provides a new method for identification of varieties of fiber rapidly and real-timely, so it has important significance for protecting the rights of consumers, ensuring the quality of textiles, and implementing rationalization production and transaction of textile materials and its production.