The parameters optimization selection of Savitzky-Golay filter and its application in smoothing pretreatment for FTIR spectra

Gas infrared spectrum should be reconstructed according to the collected spectral data by the spectrum detecting instrument which was used in the online gas Fourier transform infrared spectroscopy measurement. However, due to the influence of the factors such as ambient noise and the spectrum detecting instrument itself noise, spectrum reconstruction was mixed by noise and the results of subsequent analysis may be deviated from the true value. Therefore, the Savitzky-Golay filter was selected to smooth and denoise the spectrum in the procedure of the Fourier transform infrared spectra reconstruction. The main performance index of the Savitzky-Golay smoothing filter was decided by the polynomial order and frame size when the Savitzky-Golay filter was used for smoothing. Through the traversal search method, the relation between the main filter design index and the polynomial order, frame size were explored for the problems of the polynomial order and frame size selected arbitrary. And then, when the polynomial order and frame size should be used in the actual smoothing and filter, the above results can be choosed. According to the characteristics of the spectrum which was measured in this paper, the optimal polynomial order and frame size were selected as 8 and 11 respectively for smoothing and denoising the spectra data. The smoothing spectra for the actual spectra of 1% concentration of CH4 were very close coincidence to the standard spectra; At the same time, the converted absorbance of standard spectra, raw spectra and smoothed spectra for 1% concentration CH4 were 26.2720, 25.9017 and 26.2489 respectively, and the relative errors of the raw spectra and smoothed spectra were 1.4095% and 0.0880% respectively, in the second absorption peak area. The treatment effect was more apparent according to the method of the selection the polynomial order and frame size for the Savitzky-Golay smoothing filter, thereby a new method was provided for the Savitzky-Golay filter in other areas application.

[1]  Steven D. Brown,et al.  Near-optimal smoothing using a maximum entropy criterion , 1992 .

[2]  Achim Kohler,et al.  Optimizing Savitzky–Golay Parameters for Improving Spectral Resolution and Quantification in Infrared Spectroscopy , 2013, Applied spectroscopy.

[3]  J. Staggs Savitzky-Golay smoothing and numerical differentiation of cone calorimeter mass data , 2005 .

[4]  Tailoring noise frequency spectrum to improve NIR determinations. , 2009, Talanta.

[5]  I. Obrusník,et al.  General Least-Squares Smoothing and Differentiation by the Convolution (Savitzky-Golay) Method , 1990 .

[6]  Jianwen Luo,et al.  Savitzky-Golay smoothing and differentiation filter for even number data , 2005, Signal Process..

[7]  Jianwen Luo,et al.  Properties of Savitzky-Golay digital differentiators , 2005, Digit. Signal Process..

[8]  Quan Quan,et al.  Time-domain analysis of the Savitzky-Golay filters , 2012, Digit. Signal Process..

[9]  Andre Ivanov,et al.  Fully Automated High-Performance Signal-to-Noise Ratio Enhancement Based on an Iterative Three-Point Zero-Order Savitzky—Golay Filter , 2008, Applied spectroscopy.

[10]  P. Schoenmakers,et al.  Automatic selection of optimal Savitzky-Golay smoothing. , 2006, Analytical chemistry.

[11]  Lei Zhang,et al.  [Spectral baseline correction by piecewise dividing in Fourier transform infrared gas analysis]. , 2013, Guang pu xue yu guang pu fen xi = Guang pu.

[12]  Małgorzata Jakubowska,et al.  Adaptive-degree polynomial filter for voltammetric signals , 2004 .

[13]  P. Barak Smoothing and Differentiation by an Adaptive-Degree Polynomial Filter , 1995 .

[14]  Norbert Michael Mayer,et al.  A multiscale polynomial filter for adaptive smoothing , 2007, Digit. Signal Process..

[15]  Qian Chen,et al.  Transport-of-intensity phase imaging using Savitzky-Golay differentiation filter--theory and applications. , 2013, Optics express.

[16]  P. A. Gorry General least-squares smoothing and differentiation by the convolution (Savitzky-Golay) method , 1990 .

[17]  G. Schroeder,et al.  Thin film thickness determination using X-ray reflectivity and Savitzky-Golay algorithm , 2011 .

[18]  T. S. West Analytical Chemistry , 1969, Nature.

[19]  Lei Zhang,et al.  [An improved characteristic spectral selection method for multicomponent gas quantitative analysis based on Tikhonov regularization]. , 2012, Guang pu xue yu guang pu fen xi = Guang pu.

[20]  H. Aboul‐Enein,et al.  Recent Applications of Fourier Transform Infrared Spectrophotometry in Herbal Medicine Analysis , 2011 .