crop a raster file/rast_file on a given extent

Source: R/crop_rast.R

function to crop a raster file/rast_file on a given extent. The cropping extent can be:

  1. directly passed (as a sprawlext object or numeric array);

  2. derived from a different object from which a sprawlext object object can be derived (see details)

crop_rast(rast_object, ext_object, mask = FALSE, pad = 1,
  out_type = "rastobject", out_file = NULL, out_dtype = NULL,
  out_nodata = NULL, compress = "None", parallel = FALSE,
  cores = NULL, verbose = TRUE)

Arguments

rast_object

either an R object of class Raster, or a character string corresponding to a raster filename (with full path)
ext_object

either an object of class sprawlext, or an R object or filename from which an object of class sprawlext can be obtained (see sprawl::get_extent())
mask

logical if TRUE, and ext_object is a polygon shapefile or R object, the output is also masked on the boundaries of ext_object
pad

numeric extent of a "padding area" to be kept with respect to crop_extent (in number of pixels). Useful to be sure not to crop "too much" for example when cropping before extracting features suche as in extract_rast, Default: 1
out_type

character indicates the type of object to be returned:

  1. "rastobject" return a raster rast_file accessing the saved cropped file

  2. "rastfile" return the filename of the GTiff file corresponding to the cropped GTiff file. If out_file == NULL, this corresponds to a file saved in R temporary folder.

  3. "vrtfile" return the filename of the vrt file built for cropping the input raster (no saving to disk is performed) , Default: "rastobject"
out_file

character filename to be used to save the cropped raster.
out_dtype

character data type of the output masked files, according to gdal specifications for GTiff files ("Byte", "UInt16", "Int16", "UInt32", "Int32", "Float32", "Float64", "CInt16", "CInt32", "CFloat32" and "CFloat64"). If NULL, the data type is retrieved from the input, Default: NULL
out_nodata

numeric value to be used to fill (eventual) NoData areas outside the cropping extent (or the cropping polygon if mask = TRUE). If NULL, it is set automatically by raster::writeRaster depending on data type of the output raster.
compress

character compression option to be used to saved the cropped raster ("None", "PACKBITS", "LZW", "DEFLATE), Default: "None"
parallel

logical if TRUE, use ClusterR to implement multicore processing while masking. This speeds up execution for large rasters, Default: FALSE (Ignored if mask = FALSE)
cores

numeric Number of cores to use in case of parallel processing. If NULL, it defaults to parallel::detectCores()-2, Default = NULL (Ignored if mask = FALSE)
verbose

logical if FALSE, suppress processing messages, Default: TRUE

Value

either a Raster object containing the cropped raster, or a gdal vrt or GTiff filename corresponding to it, depending on out_type.

Examples


in_file <- system.file("extdata/OLI_test", "oli_multi_1000.tif",
                      package = "sprawl.data")
in_rast <- read_rast(in_file)

in_vect     <- create_fishnet(in_rast, pix_for_cell = 60,
                              verbose = FALSE)[55:60,]

plot_rast(in_rast[[1]], in_poly = in_vect)

out_cropped <- crop_rast(in_rast, in_vect)
#> crop_rast --> Cropping: in_rast on extent of: in_vect

get_extent(in_rast)
#> An object of class "sprawlext"
#> Slot "extent":
#>    xmin    ymin    xmax    ymax 
#>  407265 5156295  436665 5186085 
#> 
#> Slot "proj4string":
#> [1] "+proj=utm +zone=32 +datum=WGS84 +units=m +no_defs +ellps=WGS84 +towgs84=0,0,0"
#> 
get_extent(out_cropped)
#> An object of class "sprawlext"
#> Slot "extent":
#>    xmin    ymin    xmax    ymax 
#>  412635 5161665  423525 5163555 
#> 
#> Slot "proj4string":
#> [1] "+proj=utm +zone=32 +datum=WGS84 +units=m +no_defs +ellps=WGS84 +towgs84=0,0,0"
#> 

plot_rast(out_cropped[[1]])