d.b
d.b

Reputation: 32548

Identify overlapping points in scatter plot

Is there an easy way to identify the points of d1 as well as d2 that are contained inside the intersection of the red as well as the blue polygons (ie. the green polygon).

graphics.off()
set.seed(42)

#DATA
d1 = data.frame(x = rnorm(20,6,2), y = rnorm(20,5,1))
d2 = data.frame(x = rnorm(20,8.5,1), y = rnorm(20,6,1.5))
ind1 = chull(d1)
ind2 = chull(d2)
p1 = data.frame(PID = rep(1, length(ind1)), POS = seq_along(ind1), X = d1$x[ind1], Y = d1$y[ind1])
p2 = data.frame(PID = rep(1, length(ind2)), POS = seq_along(ind2), X = d2$x[ind2], Y = d2$y[ind2])
#library(PBSmapping)
#p3 = joinPolys(p1, p2)
p3 = structure(list(PID = c(1L, 1L, 1L, 1L, 1L), POS = 1:5, X = c(9.19051041605104, 
9.69947043508479, 6.94793595066407, 5.50690991684707, 8.73369696007428
), Y = c(4.69045214195459, 5.98112560649435, 6.85111759082941, 
4.67133555388548, 4.3553412183288)), .Names = c("PID", "POS", 
"X", "Y"), row.names = 0:4, class = c("PolySet", "data.frame"))

#PLOT
plot(x = c(d1$x,d2$x), y = c(d1$y, d2$y), type = "n", asp = 1)
points(d1$x, d1$y, pch = 19, col = "red")
points(d2$x, d2$y, pch = 19, col = "blue")
polygon(p1$X, p1$Y, border = "red")
polygon(p2$X, p2$Y, border = "blue")
polygon(p3$X, p3$Y, border = "green", lwd = 2)

enter image description here

Upvotes: 3

Views: 402

Answers (2)

d.b
d.b

Reputation: 32548

Based on 李哲源 Zheyuan Li's comment:

library(mgcv)
ind1_intersect = in.out(as.matrix(p3[,-(1:2)]), as.matrix(d1))
ind2_intersect = in.out(as.matrix(p3[,-(1:2)]), as.matrix(d2))

points(d1[ind1_intersect,], col = "black", pch = 19, cex = 2)
points(d2[ind2_intersect,], col = "black", pch = 19, cex = 2)

enter image description here

Upvotes: 3

G5W
G5W

Reputation: 37641

The sp package has a function point.in.polygon that does the trick

library(sp)
In1 = which(point.in.polygon(d1$x, d1$y, p3$X, p3$Y) != 0)
points(d1$x[In1], d1$y[In1], pch = 19, col = "orange")
In2 = which(point.in.polygon(d2$x, d2$y, p3$X, p3$Y) != 0)
points(d2$x[In2], d2$y[In2], pch = 19, col = "orange")

Upvotes: 4

Related Questions