fields (version 5.02)

plot.Wimage: Plots 2-d wavelet coefficents by level and type

Description

Produces a panel of images using the split.screen tools that organize the wavelet coefficients from a multiresolution by location, resolution level and type.

Usage

## S3 method for class 'Wimage':
plot(x, cut.min, graphics.reset = TRUE, common.range = FALSE, 
    color.table = tim.colors(128), Nlevel = NULL, with.lines = FALSE, 
    omd.width = 0.2, ...)

Details

As with most complicated graphical figures you basically get what it draws although it should be easy to modify this function for customization. The split.screen set of graphical functions are used to divide up the plotting real estate into subplots of different sizes. The user can experiment with different outer margin space (omd.width) Use cex.axis argument in the call to change the size of the numerals in the color strip.

By setting grahics.reset to FALSE the function returns a matrix giving the split screen ids to reference each of the individual plots. Use the screen function to move to given plot and use a high level plotting function to overlay information. (See example below.) Use close.screen( all=TRUE) to turn off split.screen mode for subsequent and normal plotting.

Examples

old.par<- par(no.readonly=TRUE) # these functions may leave the device in 
# with some funny defaults

#
#multiresolution of John Lennon 
data(lennon) 
Wtransform.image( lennon, cut.min=16)->look 
plot.Wimage( look, cut.min=16)

# adding information 
plot.Wimage( look, cut.min=16, Nlevel=3, graphics.reset=FALSE)-> plot.layout
# plot.layout here is a 4X3 matrix with the screen numbers

# move to the smooth coefficients plot

 screen( plot.layout[1,1])
 plot( c(.5,16.5), c( .5,16.5), type="n",axes=FALSE) 
 points( 8,8, cex=2, pch="+")

# NOTE: just points( 8,8) will not work here. This has to do with split.screen not 
# reseting the plotting pars correctly. 

 box(col=6, lwd=2)# just for fun 

close.screen( all=TRUE)
par( old.par) # reset  to old settings