rgdal (version 0.5-5)

SGDF2PCT: Convert RGB three band to single band colour table

Description

This function converts a three-band SpatialGridDataFrame into a single band of colour indices and a colour look-up table using RGB2PCT. vec2RGB uses given breaks and colours (like image) to make a three column matrix of red, green, and blue values for a numeric vector.

Usage

SGDF2PCT(x, ncolors = 256, adjust.bands=TRUE)
vec2RGB(vec, breaks, col)

Arguments

x

a three-band SpatialGridDataFrame object

ncolors

a number of colours between 2 and 256

adjust.bands

default TRUE; if FALSE the three bands must lie each between 0 and 255, but will not be streched within those bounds

vec

a numeric vector

breaks

a set of breakpoints for the colours: must give one more breakpoint than colour

col

a list of colors

Value

• The value returned is a list:
• idx
  a vector of colour indices in the same spatial order as the input object
• ct
  a vector of RGB colours

Details

The function makes a system() call to the GDAL utility rgb2pct.py, which is only available if GDAL has been built with Python support. It is hoped to replace this extra dependency with compiled code if the function proves useful.

References

http://www.remotesensing.org/gdal/

Examples

logo <- system.file("pictures/Rlogo.jpg", package="rgdal")[1]
SGlogo <- readGDAL(logo)
cols <- SGDF2PCT(SGlogo)
SGlogo$idx <- cols$idx
image(SGlogo, "idx", col=cols$ct)
SGlogo <- readGDAL(logo)
cols <- SGDF2PCT(SGlogo, ncolors=64)
SGlogo$idx <- cols$idx
image(SGlogo, "idx", col=cols$ct)
SGlogo <- readGDAL(logo)
cols <- SGDF2PCT(SGlogo, ncolors=8)
SGlogo$idx <- cols$idx
image(SGlogo, "idx", col=cols$ct)
data(meuse.grid)
coordinates(meuse.grid) <- c("x", "y")
gridded(meuse.grid) <- TRUE
fullgrid(meuse.grid) <- TRUE
summary(meuse.grid$dist)
opar <- par(no.readonly=TRUE)
par(mfrow=c(1,2), mar=c(1,1,1,1)+0.1)
image(meuse.grid, "dist", breaks=seq(0,1,1/10), col=bpy.colors(10))
RGB <- vec2RGB(meuse.grid$dist, breaks=seq(0,1,1/10), col=bpy.colors(10))
summary(RGB)
meuse.grid$red <- RGB[,1]
meuse.grid$green <- RGB[,2]
meuse.grid$blue <- RGB[,3]
cols <- SGDF2PCT(meuse.grid[c("red", "green", "blue")], ncolors=10, adjust.bands=FALSE)
is.na(cols$idx) <- is.na(meuse.grid$dist)
meuse.grid$idx <- cols$idx
image(meuse.grid, "idx", col=cols$ct)
par(opar)
# Note: only one wrongly classified pixel after NA handling/dropping
# The functions are not written to be reversible
sort(table(findInterval(meuse.grid$dist, seq(0,1,1/10), all.inside=TRUE)))
sort(table(cols$idx))