spc.NA.approx: Impute missing data points

In GegznaV/hyperSpec2: Work with Hyperspectral Data, i.e. Spectra + Meta Information (Spatial, Time, Concentration, ...)

Description

Replace NAs in the spectra matrix by interpolation. With less than 4 points available linear interpolation of the 2 neighbour points is used. For larger numbers of neighbour points, smoothing interpolation is performed by smooth.spline.

Usage

spc.NA.approx(spc, neighbours = 1, w = rep(1, 2 * neighbours), df = 1 +
  .Machine$double.eps, spar = NULL, debuglevel = hy.getOption("debuglevel"))

spc.NA.linapprox(...)

Arguments

spc	hyperSpec object with spectra matrix containing NAs
neighbours	how many neighbour data points should be used to fit the line
w, df, spar	see smooth.spline
debuglevel	see options
...	ignored

Value

hyperSpec object

Note

The function has been renamed from spc.NA.linapprox to spc.NA.approx

Author(s)

Claudia Beleites

Examples

fluNA <- hyperSpec:::fluNA
spc.NA.approx (fluNA [,, min ~ 410], debuglevel = 1)
spc.NA.approx (fluNA [1,, min ~ 410], debuglevel = 2)
spc.NA.approx (fluNA [4,, min ~ 410], neighbours = 3, df = 4, debuglevel = 2)

GegznaV/hyperSpec2 documentation built on Dec. 27, 2019, 9:43 a.m.