作者: Kaiyi Zheng , Xuan Zhang , Jibran Iqbal , Wei Fan , Ting Wu
DOI: 10.1002/CEM.2637
关键词: Mathematics 、 Spectral line 、 Canonical correlation 、 Piecewise 、 Extraction (chemistry) 、 Partial least squares regression 、 Biological system 、 Statistics 、 Interference (wave propagation) 、 Calibration (statistics) 、 Noise (signal processing)
摘要: A new calibration transfer method that applies canonical correlation analysis (CCA) to the informative components extracted from a spectral dataset is proposed reduce interference of noise, background and non-predicted properties. This employs partial least squares extract related predicted properties raw spectra then corrects based on CCA. The performance this algorithm was tested using three pairs batches: two corn one pair tri-component solvent spectra. results showed can significantly prediction errors compared with CCA piecewise direct standardization. Copyright © 2014 John Wiley & Sons, Ltd.