The Piecewise Two Points Autolinear Correlated Correction Method for Fourier Transform Infrared Baseline Wander

Baseline wander always blurring or even deteriorating analytical results, it is necessary to correct in order to perform further data analysis. Several correction algorithms have been proved to have some effect on this correction. However, these methods require user intervention or may be complicated and time-consuming. A novel algorithm named the autolinear correlated correction basing on piecewise was proposed. This method first divided the raw spectrum into different regions according to no absorption or nonsensitive bands which had very low or no sensitivity to target gases. And then spectrum lines were selected in every band and their means were calculated. Spectrum lines which were closest to the average were selected as the band center. According to the algorithm, the parameters of constant and autocorrelation coefficient were calculated within each band. Finally, baseline was corrected by the determined parameters. Experimental results on real spectrum demonstrated the effectiveness and high efficiency of the algorithm, making it suitable for the fully automated and online baseline correction.