如何从 R 中的多边形值中删除光栅图像的像素

Question

I have a question to process an image. I clipped some polygons from my raster, and then extracted the band values:

library(raster)
library(sf)

#Read raster
EX1<-raster::stack("ueg.tif")

#Plot raster
plotRGB(EX1, r = 1, g = 2, b = 3)

#Read shapefile
cxs <- sf::st_read("polygons_cx.shp")

#Clip polygon in raster
rr <- mask(EX1, cxs)

#Extract values RGB
ex <- raster::extract(rr, cxs, sp=T)

I would like to remove the pixel values from polygons 26 and 19 from the image (as they represent a color that I am using as a control in the analysis). But I'm not sure how I would do that... I thought of something like transforming these values in NA and then applying a new mask, but I don't know how, could anyone give any tips?

ex[c(26,19)] <- NA

Answer 1

Without a reproducible example I'm not sure if ex is a single layer raster or a stack of several. But the basic principle is to replace values that match the values you want to replace with NA . Its an extension of the same thing in base R vectors:

> Z = c(1,2,3,4,5)
> Z[Z %in% c(2,5)] = -999
> Z
[1]    1 -999    3    4 -999

Test with a tiny 3x3 raster, made from a matrix:

> m = matrix(c(1,2,3,26,26,5,NA,19,9), 3,3)
> m
     [,1] [,2] [,3]
[1,]    1   26   NA
[2,]    2   26   19
[3,]    3    5    9
> ex = raster(m)
> as.matrix(ex)
     [,1] [,2] [,3]
[1,]    1   26   NA
[2,]    2   26   19
[3,]    3    5    9

and replace all 26s and 19s with NA :

> ex[ex %in% c(26, 19)]  = NA

see what we got:

> as.matrix(ex)
     [,1] [,2] [,3]
[1,]    1   NA   NA
[2,]    2   NA   NA
[3,]    3    5    9
> 

This also seems to work okay if ex is a stack.