R - filter coordinates

问题:

I am new to R and I have a simple problem (by my opinion) but I haven't found a solution so far. I have a (long) set of 2D (x,y) coordinates - just points in 2D space, like this:

ID  x   y
1   1758.56 1179.26
2    775.67 1197.14
3   296.99  1211.13
4   774.72  1223.66
5   805.41  1235.51
6   440.67  1247.59
7   1302.02 1247.93
8   1450.4  1259.13
9   664.99  1265.9
10  2781.05 1291.12
etc.....

How do I filter points (rows in the table) that are in certain area (of any shape!)? How to filter dots that are within a subset of specified coordinates. How do I specify the wanted/unwanted area subsets? And how to put it in R? :) Thx a lot in advance!

回答1:

To check if points are inside a shape of any kind use the inpip function of the splancs package.

library(splancs)

set.seed(123)
my.shape <- matrix(runif(10), 5)
my.points <- data.frame(x=runif(500), y=runif(500))
my.points$in.shape <- 1:500 %in% inpip(my.points, my.shape)

plot(my.points[1:2], col=1 + my.points$in.shape)
polygon(my.shape)

To test for multiple shapes, put them in a list and use lapply:

set.seed(127)
multi.shapes <- lapply(1:3, function(...) matrix(runif(6), 3))
my.points$in.multi.shapes <- 1:500 %in%
    unlist(lapply(multi.shapes, function(p) inpip(my.points, p)))
plot(my.points[1:2], col=1 + my.points$in.multi.shapes)
for(p in multi.shapes) polygon(p)