R Script - 3D Surface via Interpolation of several X,Y,Z points

Question

I would like to use R and have it consume the few X,Y,Z points of data that I have (the black dots in the image) and then generate 300x300 XYZ points that would represent the 3D Surface Plot seen colored in the image. I would like the output to simple be three columns, the X,Y, and Z points. I've tried the following, but it doesn't seem to generate the data I need. I need the X,Y,Z points to generate an image in another program (spotfire) and do not need to do any graphics in R. Any help would be appreciated! Thanks.

3D Surface Plot with Points

> require(akima)
> points = read.csv("points.csv", header = TRUE)
> print(points)
         x       y    z
1   400.01  650.01 2.70
2   150.00  780.00 3.40
3   250.00  780.00 3.10
4   800.00  700.00 1.00
5  1000.00  680.00 1.00
6  1500.00  680.00 0.75
7  1700.00  700.00 1.00
8  1700.00  790.00 1.25
9  1600.00 1000.00 1.20
10 1750.00 1000.00 1.00
11 1800.00 1650.00 0.60
12 1400.00 1650.00 0.65
13 1000.00 1650.00 0.80
14  500.00 1650.00 0.90
15  150.00 1650.00 1.30
16  150.00 1050.00 3.90
17  500.00 1000.00 3.00
> s=interp(points$x,points$y,points$z,xo=seq(min(points$x),max(points$x),length=300),yo=seq(min(points$y),max(points$y),length=300))

Answer 1

It works for me. I think all you need to do is transform the final list into a matrix.

s1 <- interp2xyz(s)
dim(s1)
# [1] 90000     3

A lot of points generate a NA values for z value

sum(is.na(s1[,3]))
# [1] 4897

You might choose to exclude those points or use extrapolation.