Set built-in GDAL pixel functions of a VRT stack object

Set built-in GDAL pixel functions of a VRT stack object

Usage

vrt_set_gdal_pixelfun(x, pixfun, ..., band_idx, nodata_as_nan)

# S3 method for class 'vrt_block'
vrt_set_gdal_pixelfun(x, pixfun, ..., band_idx = NULL, nodata_as_nan = FALSE)

# S3 method for class 'vrt_collection'
vrt_set_gdal_pixelfun(x, pixfun, ..., band_idx = NULL, nodata_as_nan = FALSE)

Arguments

x

A vrt_stack object

pixfun

A built-in GDAL pixel function. See details for a list of available functions.

...

named arguments used within the pixel function see details.

band_idx

The indices of the bands to set the pixel function for. If NULL, the pixel function is set for all bands.

nodata_as_nan

A logical indicating if the NoData value should be set to NaN. Defaults to FALSE. Sometimes useful if the pixel functions ignore the NoData value. see https://github.com/OSGeo/gdal/issues/4746. If you use this you will probably need to specify the '-srcnodata' option on compute.

Details

The documentation for the GDAL pixel functions can be found at https://gdal.org/en/stable/drivers/raster/vrt.html#built-in-pixel-functions

Where a pixel function requires arguments, these can be passed as named arguments, in line with the specified naming in PixelFunctionArguments of the table at the above link.

Examples

s2files <- fs::dir_ls(system.file("s2-data", package = "vrtility"))
ex_collect <- vrt_collect(s2files)

t_block <- ex_collect[[1]][[1]]
ex_stack <- ex_collect |>
  vrt_set_maskfun(
    mask_band = "SCL",
    mask_values = c(0, 1, 2, 3, 8, 9, 10, 11),
    drop_mask_band = TRUE
  ) |>
  vrt_warp(
    t_srs = t_block$srs,
    te = t_block$bbox,
    tr = t_block$res
  ) |>
  vrt_stack()

plot(
  vrt_set_gdal_pixelfun(
    ex_stack,
    pixfun = "min",
    propagateNoData = TRUE
  ),
  bands = c(3, 2, 1)
)
#> ℹ You a plotting a warped raster - this might be okay...
#>   But, If this is taking a long time you are probably better
#>   off saving the file first with `vrt_compute` and then plotting
#>   with `plot_raster_src()`.


plot(
  vrt_set_gdal_pixelfun(
    ex_stack,
    pixfun = "min"
  ),
  bands = c(3, 2, 1)
)
#> ℹ You a plotting a warped raster - this might be okay...
#>   But, If this is taking a long time you are probably better
#>   off saving the file first with `vrt_compute` and then plotting
#>   with `plot_raster_src()`.