Non-destructive identification of wool and cashmere fibers based on improved LDA using NIR spectroscopy
Author:
Chen Xin12, Lan Qingle1, Zhu Yaolin1
Affiliation:
1. School of Electronics Information, Xi’an Polytechnic University , No.19 Jinhua South Road , Xi’an , 710048, Shaanxi , China 2. School of Automation, Northwest University of Technology , No.1 Dongxiang Road , Xi’an , 710129, Shaanxi , China
Abstract
Abstract
With the advancement of near-infrared (NIR) spectroscopy and chemometrics technology, non-destructive qualitative testing has been widely applied in many fields. Both wool and cashmere are keratin protein fibers with many similarities in tissue structure, making it very difficult to distinguish between them. In order to achieve rapid and non-destructive identification of wool and cashmere, an improved linear discriminant analysis (ILDA) algorithm combined with NIR spectroscopy technology is proposed. The proposed method can also be used for the classification of extremely similar fibers and substances, with better classification performance. First, the spectral data of wool and cashmere are collected using an NIR spectrometer so as to reduce the influence of noise in the spectra; data preprocessing methods are used to correct the collected fiber spectra. Then, principal component analysis (PCA), linear discriminant analysis (LDA), and ILDA are used to extract the characteristic variables from the spectral data. Finally, the extracted characteristic variables are input into the machine learning algorithm K-nearest neighbor (K-NN) classifier. In the experimental stage, three dimensionality reduction methods (PCA, LDA, and ILDA) are evaluated using the K-NN classification model. The fiber classification accuracy can reach 97% when using the ILDA method for dimensionality reduction. The results show that the proposed method is effective for the qualitative detection of different types of wool and cashmere fibers.
Publisher
Walter de Gruyter GmbH
Reference28 articles.
1. Jin, M., Lu, J., Fei, X., Lu, Z., Quan, K., Liu, Y., et al. (2020). Genetic signatures of selection for cashmere traits in Chinese goats. Animals, 10(10), 1905. 2. Wang, J., Sui, J., Mao, C., Li, X., Chen, X., Liang, C., et al. (2021). Identification of key pathways and genes related to the development of hair follicle cycle in cashmere goats. Genes, 12(2), 180. 3. Zhang, X., Wu, X., Yang, H., Zheng, H., Zhou, Y. (2023). Identification of cashmere and wool by DNA barcode. Journal of Natural Fibers, 20(1), 2175100. 4. Atav, R., Ergünay, U., Gürkan Unal, P. (2023). Determining the effect of pigmentation on some physical and mechanical properties of yak and cashmere down fibers. Journal of Natural Fibers, 20(1), 2149939. 5. Sun, M. R., Fei, J., Cai, J. S., Zhao, J. (2016). Application of dna analysis in quantifying cashmere and wool binary blend. Key Engineering Materials, 671, 378–384.
|
|