Abstract

The need may arise to use spectral data with a sample interval and slit width that differ from those used when the data were recorded. Results are presented of an examination of the quantitative accuracy of conversion of these parameters using Raman spectra. Assuming adequate sampling in the original set, the interval can be changed by forward and inverse Fourier transform with zero-filling. This procedure only permits expansion by powers of two, but by overexpansion of the original data set, points can be selected to achieve or very closely approximate other factors. Expansion by a factor of ten is demonstrated. The effect of varying slit function is compensated for by deconvoluting the data against the slit function originally used and then convoluting with that which corresponds to the desired condition. Conditions for carrying out this conversion with less than 0.1% error are described.

Multiframe deconvolution with space-variant point-spread functions by use of inverse filtering and fast Fourier transform

Mats G. Löfdahl
Appl. Opt. 46(21) 4686-4693 (2007)

Instrumental Corrections in Measurements of Gas Dispersion. A Three-Slit Problem*

J. H. Jaffe, J. Meiron, and N. Jacobi
J. Opt. Soc. Am. 52(1) 8-14 (1962)

Determination of the instrument function of a grating spectrometer by using white-light interferometry

V. Nirmal Kumar and D. Narayana Rao
Appl. Opt. 36(19) 4535-4539 (1997)

Correction of instrument line shape in Fourier transform spectrometry using matrix inversion

Raphaël Desbiens, Jérôme Genest, Pierre Tremblay, and Jean-Pierre Bouchard
Appl. Opt. 45(21) 5270-5280 (2006)

Estimation of spectral slit width and blind deconvolution of spectroscopic data by homomorphic filtering

Yasuhiro Senga, Keiichiroh Minami, Satoshi Kawata, and Shigeo Minami
Appl. Opt. 23(10) 1601-1608 (1984)