NorthernOntario_1.R

library(raster)
library(sp)
library(rgdal)
library(dplyr)

#######################################################################

#' set working directory and identify name of image file

wd <- setwd("c:/personal/r")

par(mai=c(1.02,0.82,0.82,0.42))

image.lst <- c(1000, 1001, 1002, 1003, 1004, 1005)

#' Specify out raster with "filename" argument also test the pixel value (pix).
#' Pixel values can be adjusted for each TIF tile specifically. 
#' identify variable (val) to encode the pixels greater than a certain value
val <- 1200
pix <- c(20, 20, 20, 20, 20, 20)

# start loop

for (i in 1: length(image.lst)){
  
  # call each image
  image.name <- paste0("Ontario", image.lst[i],".tif")
  
  #######################################################################
  
  #' load Tif as raster image
  no <- raster(image.name)
  plot(no)    # plot image
  
  #' use the locator function to identify the pixel value to use in the 
  #' calculator function below
  
  #pxy <- locator(1) # click on plot where you want to query pixel value
  #extract(no, cbind(pxy$x, pxy$y))) # returns value associated with click query
  
  r05 <- no
  
    #' write the raster to points while subsetting for user defined variable: PIX
  
  r2.points <- rasterToPoints(r05, fun = function(x){x>pix[i]})
  
  # write points out to CSV
  write.csv(r2.points, file = paste0("Ontario", image.lst[i], "_", pix[i], ".csv"))
  
  
}