eptmap: Multiscale Visualization of Ensemble Patch Transform of a...

In EPT: Ensemble Patch Transform, Visualization and Decomposition

Description

This function displays time-scale representation of ensemble patch transform of a signal for a sequence of size parameters.

Usage

eptmap(eptransf, taus = eptransf$parameters$tau, maptype = c("C", "D", "DC", "DD"), 
    stat = c("pstat", "Epstat", "pM", "EpM", "psd", "Epsd"),
    der = c("time", "tau"), ncolor = 100, ...) 

Arguments

eptransf	R object of ensemble patch transform by eptransf() or meptransf().
taus	specifies size parameters for time-scale visualization.
maptype	specifies "C" for centrality map, "D" for dispersion map, "DC" for derivative of centrality map and "DD" for derivative of dispersion map.
stat	"pstat" for centrality of patch transform, "Epstat" for centrality of ensemble patch transform, "pM" for mean envelope of patch transform, "EpM" for mean envelope of ensemble patch transform, "psd" for standard deviation of patch transform and "Epsd" for standard deviation of ensemble patch transform.
der	specifies derivative with respect to "time" or "tau".
ncolor	the number of colors (≥ 1) to be in the palette.
...	graphical parameters for image.

Details

This function performs multiscale visualization of ensemble patch transform of a signal for a sequence of size parameters. This function creates images with heat.colors(ncolor) colors.

Value

image

See Also

eptransf, meptransf, eptplot.

Examples

# a doppler signal
n <- 1000
tindex <- seq(0, 1, length=n)
j <- 5
f <- 10 * sqrt(tindex*(1-tindex)) * sin((2*pi*(1+2^((9-4*j)/5))) / (tindex+2^((9-4*j)/5)))

set.seed(7)
fnoise <- f + 0.4 * rnorm(n)

op <- par(mar=c(2,2,2,1))
plot(f, type="l", , xlab="", ylab="", ylim=range(fnoise))
points(fnoise, cex=0.3)


taus <- seq(4, 64, by=4)

# try1 : Multiscale EPT by average patch transform and average ensemble transform 
try1 <- meptransf(tindex=tindex, signal=fnoise, taus=taus, process=c("average", "average"))

par(mfrow=c(2,2))
eptmap(try1, maptype="C", stat="pstat", main="centrality of patch transform")
eptmap(try1, maptype="D", stat="psd", main="standard deviation of patch transform")
eptmap(try1, maptype="C", stat="Epstat", main="centrality of ensemble patch transform")
eptmap(try1, maptype="D", stat="Epsd", main="standard deviation of ensemble patch transform")

eptmap(try1, maptype="DC", stat="Epstat", der="time", 
    main="derivative of centrality w.r.t time")
eptmap(try1, maptype="DC", stat="Epstat", der="tau", 
    main="derivative of centrality w.r.t tau")
eptmap(try1, maptype="DD", stat="Epsd", der="time", 
    main="derivative of standard deviation w.r.t time")
eptmap(try1, maptype="DD", stat="Epsd", der="tau", 
    main="derivative of standard deviation w.r.t tau", ncolor=70)

# try2 : Multiscale EPT by envelope patch transform and average ensemble transform 
try2 <- meptransf(tindex=tindex, signal=fnoise, taus=taus, process=c("envelope", "average"),
    pquantile=c(0, 1))

eptmap(try2, maptype="C", stat="pM", main="mean envelope of patch transform")
eptmap(try2, maptype="C", stat="EpM", main="mean envelope of ensemble patch transform")
eptmap(try2, maptype="DC", stat="EpM", der="time", 
    main="derivative of mean envelope w.r.t time")
eptmap(try2, maptype="DC", stat="EpM", der="tau", 
    main="derivative of mean envelope w.r.t time")

par(op)

EPT documentation built on Jan. 5, 2022, 9:06 a.m.