Modified linear model correction: A calibration transfer method without standard samples

Abstract

Calibration transfer without standard samples is essential for practical applications of near infrared spectroscopy because, sometimes, it is difficult or even impossible to obtain the standard samples for measuring their spectra on the secondary instrument. In this work, a modified linear model correction method is proposed for improving the transfer accuracy and computational efficiency. The constraint of linear model correction was replaced by a robust convex equation to restrict the model similarity in the optimization. The near infrared dataset of pharmaceutical tablet measured with different instruments are used to test the performance of the method. The result shows that a modified linear model correction achieves a high efficiency of the transfer while the computational complexity can be considerably reduced. The method may provide a robust way in practical application.