feemgrid: Interpolate FEEMs on a given wavelength grid

Description

Use interpolation to change the wavelength grid of a single FEEM or unify the grid of a collection of them.

Usage

feemgrid(x, ...)
  # S3 method for feem
feemgrid(
    x, emission, excitation,
    method = c("whittaker", "loess", "kriging", "pchip"), ...
  )
  # S3 method for feemcube
feemgrid(
    x, emission, excitation, ..., progress = TRUE
  )
  # S3 method for list
feemgrid(
    x, emission, excitation, ..., progress = TRUE
  )

Value

An object of the same kind (FEEM object / FEEM cube / list of them) with emission and excitation wavelengths as requested.

Arguments

x: A feem object, a feemcube, or a list of feem objects.
emission, excitation: Desired wavelength grid, as numeric vectors. Must be specified for a single FEEM. If not specified for a collection of FEEMs, all wavelengths falling in the range of the intersection all wavelengths intervals are chosen.
method: Interpolation method, see feemscatter for details.
...: Passed from generics to feemgrid.feem, then to interpolation methods. See feemscatter for details.
progress: Set to FALSE to disable the progress bar.

Details

The algorithm doesn't know how to distinguish between NAs that haven't been measured and NAs that resulted from combining different wavelength grids, so it tries to interpolate all of them. As a result, leaving large areas of the spectrum undefined (e.g. anti-Stokes area) is not recommended, since it would result in extrapolation and introduce strong artefacts.

Examples

Run this code

  data(feems)
  x <- feemscatter(feems$a, rep(25, 4))
  y <- feemgrid(x, seq(240, 600, 5), seq(230, 550, 10))
  plot(plot(x, main = 'Original'    ), split = c(1, 1, 2, 1), more = TRUE)
  plot(plot(y, main = 'Interpolated'), split = c(2, 1, 2, 1))