factor.RdThese functions allow for defining a RasterLayer as a categorical variable. Such a RasterLayer is linked to other values via a "Raster Attribute Table" (RAT). Thus the cell values are an index, whereas the actual values of interest are in the RAT. The RAT is a data.frame. The first column in the RAT ("ID") has the unique cell values of the layer; this column should normally not be changed. The other columns can be of any basic type (factor, character, integer, numeric or logical). The functions documented here are mainly available such that files with a RAT can be read and processed; currently there is not too much further support. Whether a layer is defined as a factor or not is currently ignored by almost all functions. An exception is the 'extract' function (when used with option df=TRUE).
Function 'levels' returns the RAT for inspection. It can be modified and set using levels <- value (but use caution as it is easy to mess things up).
as.factor and ratify create a layer with a RAT table. Function 'deratify' creates a single layer for a (or each) variable in the RAT table.
is.factor(x)
as.factor(x)
levels(x)
# S4 method for class 'Raster'
ratify(x, filename="", count=FALSE, ...)
factorValues(x, v, layer=1, att=NULL, append.names=FALSE)
deratify(x, att=NULL, layer=1, complete=FALSE, drop=TRUE, fun='mean', filename='', ...)
asFactor(x, ...)Raster* object
integer cell values
integer > 0 indicating which layer to use (in a RasterStack or RasterBrick)
numeric or character. Which variable(s) in the RAT table should be used. If NULL, all variables are extracted. If using a numeric, skip the first two default columns
logical. Should names of data.frame returned by a combination of the name of the layer and the RAT variables? (can be useful for multilayer objects
character. Optional
logical. If TRUE, a columns with frequencies is added
additional arguments as for writeRaster
logical. If TRUE, the layer returned is no longer a factor
logical. If TRUE a factor is converted to a numerical value if possible
character. Used to get a single value for each class for a weighted RAT table. 'mean', 'min', 'max', 'smallest', or 'largest'
Raster* object; list (levels); boolean (is.factor); matrix (factorValues)
asFactor is deprecated and should not be used
set.seed(0)
r <- raster(nrow=10, ncol=10)
values(r) <- runif(ncell(r)) * 10
is.factor(r)
#> [1] FALSE
r <- round(r)
f <- as.factor(r)
is.factor(f)
#> [1] TRUE
x <- levels(f)[[1]]
x
#> ID
#> 1 0
#> 2 1
#> 3 2
#> 4 3
#> 5 4
#> 6 5
#> 7 6
#> 8 7
#> 9 8
#> 10 9
#> 11 10
x$code <- letters[10:20]
levels(f) <- x
levels(f)
#> [[1]]
#> ID code
#> 1 0 j
#> 2 1 k
#> 3 2 l
#> 4 3 m
#> 5 4 n
#> 6 5 o
#> 7 6 p
#> 8 7 q
#> 9 8 r
#> 10 9 s
#> 11 10 t
#>
f
#> class : RasterLayer
#> dimensions : 10, 10, 100 (nrow, ncol, ncell)
#> resolution : 36, 18 (x, y)
#> extent : -180, 180, -90, 90 (xmin, xmax, ymin, ymax)
#> crs : +proj=longlat +datum=WGS84 +no_defs
#> source : memory
#> names : layer
#> values : 0, 10 (min, max)
#> attributes :
#> ID code
#> from: 0 j
#> to : 10 t
#>
r <- raster(nrow=10, ncol=10)
values(r) = 1
r[51:100] = 2
r[3:6, 1:5] = 3
r <- ratify(r)
rat <- levels(r)[[1]]
rat$landcover <- c("Pine", "Oak", "Meadow")
rat$code <- c(12,25,30)
levels(r) <- rat
r
#> class : RasterLayer
#> dimensions : 10, 10, 100 (nrow, ncol, ncell)
#> resolution : 36, 18 (x, y)
#> extent : -180, 180, -90, 90 (xmin, xmax, ymin, ymax)
#> crs : +proj=longlat +datum=WGS84 +no_defs
#> source : memory
#> names : layer
#> values : 1, 3 (min, max)
#> attributes :
#> ID landcover code
#> 1 Pine 12
#> 2 Oak 25
#> 3 Meadow 30
#>
# extract values for some cells
i <- extract(r, c(1,2, 25,100))
i
#> [1] 1 1 3 2
# get the attribute values for these cells
factorValues(r, i)
#> landcover code
#> 1 Pine 12
#> 2 Pine 12
#> 3 Meadow 30
#> 4 Oak 25
# write to file:
# rr <- writeRaster(r, rasterTmpFile(), overwrite=TRUE)
# rr
# create a single-layer factor
x <- deratify(r, "landcover")
x
#> class : RasterLayer
#> dimensions : 10, 10, 100 (nrow, ncol, ncell)
#> resolution : 36, 18 (x, y)
#> extent : -180, 180, -90, 90 (xmin, xmax, ymin, ymax)
#> crs : +proj=longlat +datum=WGS84 +no_defs
#> source : memory
#> names : landcover
#> values : 1, 3 (min, max)
#> attributes :
#> ID landcover
#> 1 Pine
#> 2 Oak
#> 3 Meadow
#>
is.factor(x)
#> [1] TRUE
levels(x)
#> [[1]]
#> ID landcover
#> 1 1 Pine
#> 2 2 Oak
#> 3 3 Meadow
#>