reclassify values of a raster

function to reclassify values of a raster object or raster file based on a reclassification matrix which specifies which value in the output should be taken by different intervals of values of the input raster

recategorize_rast(in_rast, class_matrix, out_file = NULL,
  out_type = "rastobject", overwrite = FALSE, verbose = TRUE)

Arguments

in_rast	Input raster file or "R" raster layer to be reclassified
class_matrix	`data.frame` 3 column data frame with column names equal to `start`, `end`, `new` and `label`. Each line specifies an interval of values in the original raster and the value they will assume in the output. IMPORTANT ! Intervals in `class_matrix` will be considered as CLOSED on THE LEFT and OPEN ON THE RIGHT ! - see examples !!!! -
out_file	`character` Name of the output raster file. If NULL and to_file == TRUE , the raster is saved to a file in the `R` temporary folder in "R" tempdir(), Default: NULL
out_type	`character` if == "rastobj", return a `Raster` object; if == "filename", return the filename of the masked layer (GTiff or gdal vrt format depending on other arguments - see below) Default: "rastobj" (If an invalid string is provided, defaults to `rastobj`)
overwrite	`logical` If TRUE and `out_file`` already exists, the file is overwritten, Default: FALSE
verbose	`logical` if TRUE, extended processing information is sent to the console in the form of messages

Value

if out_type == rastobject: *raster corresponding to the reclassified raster
- if out_type == "rastfile": name of the file where it was saved (either corresponding to out_file or to the name of the temporary file created (if out_file == NULL))

Details

The function is a wrapper around the raster::reclassify function, providing a (somehow) easier and extended I/O functionality and possibility to easily set the class names and eventually plotting colors for the new raster using raster::ratify

class_matrix must be a 3-columns data.frame (or similar), including the start, end, newand label columns. It can be easily created by, for example:

class_matrix <- tribble(~start, ~end, ~new, ~label
                             0,    1,   NA, NA         # values >=0 and < 1 will be set to NA
                             1,    5,   1,  "Class_1", # values >=1 and < 5 will be set to 1
                             5,    7,   2,  "Class_2", # values >=5 and < 7 will be set to 2
                           ...,  ..., ...,  ...,
                           ...,  ..., ...,  ...,
                            11,  100,  NA,  NA)   # values >=11 and < 100 will be set to NA

, or:

class_matrix <- data.frame(start = c(0,  1, 5, ..., ..., 11),
                           end   = c(1,  5, 7, ..., ..., 100),
                           new   = c(NA, 1, 2, ..., ..., NA),
                           label = c(NA, "Class_1", "Class_2", ..., ..., NA)
                           )

Note that it is FUNDAMENTAL for proper functioning that ALL values of the original raster are "covered" in class_matrix. Values not falling in any of the intervals will be kept at their original values ! (checks for this will be implemented in the future!)

Examples


# create a raster with values equal to rows number and assign a RAT
# to it
  library(magrittr)
#> 
#> Attaching package: ‘magrittr’
#> The following object is masked from ‘package:raster’:
#> 
#>     extract
#> The following objects are masked from ‘package:testthat’:
#> 
#>     equals, is_less_than, not
  in_rast <- raster::raster(ncol = 20, nrow = 20) %>%
     raster::init("row")
  plot_rast_gg(in_rast, rast_type = "continuous", scalebar = FALSE,
               direction = -1)

# build a reclassification matrix

  class_matrix <- tibble::tribble(
                     ~start, ~end, ~new, ~label, ~color,
                     -Inf,   5,    1, "pippo",   "red", # Values  < 5 --> 1
                        5,   8,    2, "pluto",   "green",# >=5 and < 8 --> 2
                        8,   12,   2, "pluto",   "green",# >=8 and < 12 --> 2
                       12,  15,   NA, NA,         NA,#>=12 and < 15 --> NA
                       15,  Inf,   3, "paperino", "red")# Values >=15  --> 3
# reclassify and assign class names
out_rast <- recategorize_rast(in_rast,
                        class_matrix)
#> recategorize_rast--> Reclassifying: in_rast on values of: class_matrix
out_rast
#> class       : RasterLayer 
#> dimensions  : 20, 20, 400  (nrow, ncol, ncell)
#> resolution  : 18, 9  (x, y)
#> extent      : -180, 180, -90, 90  (xmin, xmax, ymin, ymax)
#> coord. ref. : +proj=longlat +datum=WGS84 +no_defs +ellps=WGS84 +towgs84=0,0,0 
#> data source : /tmp/Rtmpzfhi9J/file78523f0b1718.tif 
#> names       : file78523f0b1718 
#> values      : 1, 3  (min, max)
#> attributes  :
#>  ID    Class Color
#>   1    pippo   red
#>   2    pluto green
#>   3 paperino   red
#> 
plot_rast_gg(out_rast, scalebar = FALSE)


# reclassify a land cover map with N classes to a 0-1 mask, retaining classes 5
# and 9, putting the rest to 0 and values >= 11 to NA
# Open  the masking file

# in_mask <- raster(in_maskfile)
# setup the reclassification matrix

#  class_matrix <- tibble::tribble(
#                ~start, ~end, ~new, ~label, ~color,
#                     0,   0,   NA, NA, NA,
#                     1,   5,   0,  NA, "black",
#                     5,   6,   1,  NA, "white",
#                     6,   9,   1,  NA, "white",
#                     9,  11,   1,  NA, "white",
#                    11, 100,  NA,  NA, NA)

# reclass_file = "/home/lb/Temp/buttami/pippo_reclass.tif"
# outmask = categorize_rast(in_rast,
#                        class_matrix,
#                        r_out = TRUE)
# plot_rast_gg(outmask)

Arguments

Value

Details

See also

Examples

Contents