r - Grid of points from polygon input

2
votes

I am writing a script that takes an input KML file, created in google earth, and plots a grid of coordinate points inside of the polygon.

So far, I have the polygon input and a grid of points for the bounding box of the polygon, but I want to have only points INSIDE the polygon.

I tried to do this using the over() function, but it is not working. Suggestions?

You can download my test KML file HERE.

library(rgdal)
library(sp)
library(maptools)

# ogrInfo() to find layer name... not as labelled in Google Earth?!
my.poly = readOGR(ds = "PolyNYC.kml", layer = "PolyNYC") 
proj4string(my.poly) <- "+proj=longlat +datum=WGS84 +no_defs"

# Creating grid of points
grdpts <- makegrid(my.poly)

# Converting from df to spdf
coords = cbind(grdpts$x1, grdpts$x2)
sp = SpatialPoints(coords)
spdf = SpatialPointsDataFrame(coords, grdpts, proj4string = CRS(proj4string(my.poly)))

# Using over() to select only those points in the polygon
inPoly = over(spdf, my.poly)
# This is not working

# Plotting the polygon with the points overlaid.
plot(my.poly)
points(spdf, pch = 3, col = "red")

#kmlPoints(obj = spdf, kmlfile = "BBoxFromPoly.kml", kmlname = "Testing123")
1
rgeospatialpolygonspatial

1 Answers

6
votes

I'm going to show a solution using library(sf), which is the successor to library(sp)

Loading data

library(sf)

## read the kml
my.poly <- sf::st_read("~/Downloads/PolyNYC.kml")

## create a grid of points
grdpts <- sf::st_make_grid(my.poly, what = "centers")

## convert it to an `sf` object, as opposed to an `sfc`
my.points <- sf::st_sf(grdpts)

Viewing data

To view the objects on the map I'm using my googleway package that plots it on a Google Map (therefor you need an API key), but you can use leaflet or whatever map you want

library(googleway)

set_key("your_api_key_here")

google_map() %>%
  add_polygons(my.poly) %>%
  add_markers(my.points)

enter image description here

Points In Polygon

You can use the function sf::st_join() to join geometries

pointsInside <- sf::st_join(x = my.points, y = my.poly, left = FALSE)

# Simple feature collection with 59 features and 2 fields
# geometry type:  POINT
# dimension:      XY
# bbox:           xmin: -74.1754 ymin: 40.63513 xmax: -73.75675 ymax: 40.8514
# epsg (SRID):    4326
# proj4string:    +proj=longlat +datum=WGS84 +no_defs
# First 10 features:
#   Name Description                   geometry
# 1  TestLayerNYC             POINT (-74.08237 40.63513)
# 2  TestLayerNYC             POINT (-74.03585 40.63513)
# 3  TestLayerNYC              POINT (-74.1754 40.65916)
# 4  TestLayerNYC             POINT (-74.12889 40.65916)
# 5  TestLayerNYC             POINT (-74.08237 40.65916)
# 6  TestLayerNYC             POINT (-74.03585 40.65916)
# 7  TestLayerNYC             POINT (-73.80326 40.65916)
# 8  TestLayerNYC              POINT (-74.1754 40.68319)
# 9  TestLayerNYC             POINT (-74.12889 40.68319)
# 10 TestLayerNYC             POINT (-74.08237 40.68319)

Here, pointsInside is all the points that are within the polygon

View the result

google_map() %>%
  add_polygons(my.poly) %>%
  add_markers(pointsInside)

enter image description here