Interpolating data in R

Question

Let's suppose I have a 3 by 5 matrix in R:

4  5  5  6  8
3  4  4  5  6
2  3  3  3  4

I would like to interpolate in between these values to create a matrix of size 15 by 25. I would also like to specify if the interpolation is linear, gaussian, etc. How can I do this?

For example, if I have a small matrix like this

2 3
1 3

and I want it to become 3 by 3, then it might look like

  2    2.5   3
  1.5  2.2   3
  1    2     3

this post may be of some help? http://stackoverflow.com/questions/3093455/3d-geometry-how-to-interpolate-a-matrix — Arun, Apr 15 '13 at 22:16
@Arun that seems useful although there must be a built in function in R to do this... — CodeGuy, Apr 15 '13 at 22:21

score 6 · Accepted Answer · answered Apr 15 '13 at 22:40

6

app <- function(x, n) approx(x, n=n)$y # Or whatever interpolation that you want

apply(t(apply(x, 1, function(x) app(x, nc))), 2, function(x) app(x, nr))
     [,1] [,2] [,3]
[1,]  2.0 2.50    3
[2,]  1.5 2.25    3
[3,]  1.0 2.00    3

answered Apr 15 '13 at 22:40

Matthew Lundberg

42,009
6
90
112

To followup on my comments at my answer: if I'm reading this code correctly, setting `nc` and `nr` to the number of cols/rows you want for the output matrix will allow arbitrary asymmetric expansion. – Carl Witthoft Apr 16 '13 at 20:28

score 0 · Answer 2 · answered Apr 16 '13 at 12:07

Long time ago I wrote a similar toy, except I never got around to defining the interpolation function. There's also raster::disaggregate .

zexpand<-function(inarray, fact=2, interp=FALSE,  ...)  {
# do same analysis of fact to allow one or two values, fact >=1 required, etc.
fact<-as.integer(round(fact))
switch(as.character(length(fact)),
            '1' = xfact<-yfact<-fact,
            '2'= {xfact<-fact[1]; yfact<-fact[2]},
            {xfact<-fact[1]; yfact<-fact[2];warning(' fact is too long. First two values used.')})
if (xfact < 1) { stop('fact[1] must be > 0') } 
if (yfact < 1) { stop('fact[2] must be > 0') }
bigtmp <- matrix(rep(t(inarray), each=xfact), nrow(inarray), ncol(inarray)*xfact, byr=T)  #does column expansion
bigx <- t(matrix(rep((bigtmp),each=yfact),ncol(bigtmp),nrow(bigtmp)*yfact,byr=T))
# the interpolation would go here. Or use interp.loess on output (won't
# handle complex data). Also, look at fields::Tps which probably does
# a much better job anyway.  Just do separately on Re and Im data
return(invisible(bigx))
}

Thanks for the response. How is this different or better than Math Lundberg's answer above? — CodeGuy, Apr 16 '13 at 18:06
@CodeGuy If I actually put an interpolation function in there (basically replacing the `rep` function which just duplicates rows or columns), it would be doing almost exactly the same thing, except that I allow for different expansions in the column- and row- dimensions. As Matthew wrote, you can swap in whatever interpolator you want in is code as well. — Carl Witthoft, Apr 16 '13 at 20:20

Interpolating data in R

2 Answers2