将shapefile从多边形转换为点?

时间:2013-02-08 16:32:38

标签: r gis polygon point shapefile

我有一个非重叠的基于多边形的shapefile (.shp),它具有很大的空间范围和许多相关的属性。 shapefile以UTM格式投影。我希望将多边形转换为在30-m分辨率网格中间隔开的点,其中每个点将保留其所在多边形的属性。

输出只是一个点表:

X, Y, attribute1, attribute2, attribute 3,etc...

我理想情况下喜欢在R中执行此操作,或者(不太理想)我可以在Mac上运行的其他免费程序。

1 个答案:

答案 0 :(得分:5)

注意:我正在考虑这一点,以了解是否有一种更优雅的方式可以做到这一点。所以,请注意空间类型,并提出任何改进建议。

(特别是,第2步,设置一个"SpatialPoints"网格,其中包含要提取值的点,对我来说似乎总是很低级。)


这使用over()从为"SpatialPolygonDataFrame"对象中包含的坐标中提取"SpatialPoints"的属性。

library(rgdal)

## (1) Read in an example shapefile
dsn <- system.file("vectors", package = "rgdal")[1]
scot_BNG <- readOGR(dsn=dsn, layer="scot_BNG")
scot_BNG <- scot_BNG[1:5,]  # Let's just use part of it

## (2) Set up a SpatialPoints object with the grid of points 
##     for which you want to extract values
res <- 10000            ## Distance between grid points (30 in OP's question) 
BB <- bbox(scot_BNG)
BB <- res*round(BB/res) ## Pretty up the bounding box
GT <- GridTopology(cellcentre.offset = BB[,1], 
                   cellsize = c(res, res),
                   cells.dim = (c(diff(BB[1,]), diff(BB[2,]))/res) + 1)
SP <- SpatialPoints(GT, proj4string = CRS(proj4string(scot_BNG)))

## (3) Extract the values
vals <- over(SP, scot_BNG)
res <- cbind(coordinates(SP), vals)

## Finally, have a look at a few of the points.
x <- res[!is.na(res$SP_ID),]
rbind(head(x,3), tail(x,3))[1:10]
#          x      y SP_ID       NAME ID_x COUNT   SMR  LONG  LAT    PY
# 4   230000 970000     0 Sutherland   12     5 279.3 58.06 4.64 37521
# 5   240000 970000     0 Sutherland   12     5 279.3 58.06 4.64 37521
# 25  220000 960000     0 Sutherland   12     5 279.3 58.06 4.64 37521
# 425 260000 780000     4   Bedenoch   17     2 186.9 57.06 4.09 27075
# 426 270000 780000     4   Bedenoch   17     2 186.9 57.06 4.09 27075
# 427 280000 780000     4   Bedenoch   17     2 186.9 57.06 4.09 27075