Identification of tea varieties by mid‐infrared diffuse reflectance spectroscopy coupled with a possibilistic fuzzy c‐means clustering with a fuzzy covariance matrix

• DOI: 10.1111/jfpe.13298
• Publication Date: 2019-10-31T13:15:49Z
• Authors: Xiaohong Wu; Jing Zhou; Bin Wu; Jun Sun; Chunxia Dai
• Keywords: Matrix (chemical analysis); Diffuse reflection

Abstract Mid‐infrared diffuse reflectance spectroscopy was used to rapidly and nondestructively identify tea varieties together with the proposed possibilistic fuzzy c‐means (PFCM) clustering a covariance matrix. The mid‐infrared spectra of 96 samples three different (Emeishan Maofeng, Level 1, 6 Leshan trimeresurus) were acquired using FTIR‐7600 infrared spectrometer. First, multiplicative scatter correction implemented pretreat spectral data. Second, principal component analysis employed compress data after preprocessing. Third, linear discriminant utilized for extracting identification information required by algorithms. Ultimately, (FCM) clustering, allied (AFCM) PFCM matrix cluster processed data, respectively. highest accuracy reached at 100% compared those FCM (96.7%), AFCM (94.9%), (96.3%), partial least squares discrimination (PLS‐DA) algorithm (33.3%). It is sufficiently demonstrated that coupled valid method identifying varieties. Practical applications variety vitally important evaluate quality in market. deemed be convenient, rapid, accurate, nondestructive detection technology comparison traditional methods. In this article, can determine quickly correctly. experimental results indicate application potential examination fake products discrimination.