The package corr2D implements two-dimensional (2D) correlation
analysis in R. It allows the user to calculate and plot homo as well as
hetero 2D correlation spectra from an input matrix containing the
(preprocessed) spectra. The user can freely choose the reference
spectrum used to create the dynamic spectra, how the perturbation
variables vector should be handled, if and how the resulting 2D spectra
should be scaled and how many cores should be used to calculate the
correlation matrix.
The plot functions offer a wide variety of options to customize the look
of 2D spectra. In this context the appearance of the main spectrum can
be altered and the 1D reference spectra as well as the color legend can
be controlled.
The package also features an example dataset containing preprocessed
FT-Raman spectra of a self-healing polymer sample.
First we need to make the package available and get an example dataset. For illustration purposes we use the FuranMale dataset from the present corr2D package.
library("corr2D")
data(FuranMale, package = "corr2D")As the next step we calculate the homo 2D correlation spectra of the FuranMale dataset using the function corr2d() and the first spectrum of the dataset as reference spectrum.
twod <- corr2d(FuranMale, Ref1 = FuranMale[1, ], corenumber = 1)
#> HOMO-Correlation: 1 cores used for calculation
#> 12:27:42 - Fast Fourier Transformation and multiplication
#> to obtain a 145 x 145 correlation matrix
#> 12:27:42 - DoneThe last step is to plot the resulting 2D correlation spectra. As an example we plot the synchronous correlation spectrum of the FuranMale dataset. The appearance of the 2D plot can be altered by i.e. using the xlab and ylab options to control the labels at the x- and y-axes.
plot_corr2d(twod, xlab = expression(paste("relative Wavenumber" / cm^-1)),
ylab = expression(paste("relative Wavenumber" / cm^-1)))
#> [1] 3
For an in-depth tutorial as well as a detailed description of the code see R. Geitner et al. (2019) https://doi.org/10.18637/jss.v090.i03.