SPUTNIK (version 1.4.2)

gini.index: Gini index.

Description

gini.index returns the Gini index of the ion intensity vector as a measure of its sparseness. The intensity vector is first quantized in N levels (default = 256). A value close to 1 represents a high level of sparseness, a value close to 0 represents a low level of sparseness.

Usage

gini.index(x, levels = 256)

Value

A value between 0 and 1. High levels of signal sparsity are associated with values close to 1, whereas low levels of signal sparsity are associated with values close to 0.

Arguments

x

numeric. Peak intensity array.

levels

numeric (default = 256). Number of levels for the peak intensity quantization.

Author

Paolo Inglese

References

Hurley, N., & Rickard, S. (2009). Comparing measures of sparsity. IEEE Transactions on Information Theory, 55(10), 4723-4741.

See Also

scatter.ratio spatial.chaos

Examples

## Load package
library("SPUTNIK")

## Image
im <- matrix(rnorm(100), 10, 10)
im[im < 0] <- 0

## Spatial chaos
sc <- spatial.chaos(im, levels = 30, morph = TRUE)
stopifnot(sc <= 1)

## Gini index
gi <- gini.index(im, levels = 16)
stopifnot(gi >= -1 && gi <= 1)

## Scatter ratio
sr <- scatter.ratio(im)
stopifnot(sr <= 1)