noise.filter: Median, Mean and Gaussian Filter

In wvtool: Image Tools for Automated Wood Identification

Description

A funtion provides three kinds of noise reduction on an image, "median", "mean", and "gaussian". A typical pre-processing step to improve the results of later processing for example, glcm-haralick analysis.

Usage

noise.filter(x, n=3, method="median")

Arguments

x	A raster image or a matrix
n	filter size is given by n x n. Default is 3 x 3. Number has to be an odd number. For gaussian filter, only 3 or 5 is available.
method	"median", "mean", and "gaussian" can be selected. Default is "median".

Value

A raster or a matrix

References

T.S. Huang, G.J. Yang, G.Y. Tang (1979) A fast two-dimensional median filtering algorithm, IEEE transactions, Acoustics, Speech and Signal Processing, 27, 13-18.

See Also

glcm

Examples

data(camphora)
camphora <- crop(camphora,200,200)
par(mfrow=c(2,2))
image(rot90c(noise.filter(camphora,3,"median")),col=gray(c(0:255)/255), 
main="median", useRaster=TRUE, axes=FALSE, asp=1)
image(rot90c(noise.filter(camphora,3,"mean")),col=gray(c(0:255)/255), 
main="mean", useRaster=TRUE, axes=FALSE, asp=1)
image(rot90c(noise.filter(camphora,3,"gaussian")),col=gray(c(0:255)/255), 
main="gaussian", useRaster=TRUE, axes=FALSE, asp=1)

wvtool documentation built on May 1, 2019, 10:27 p.m.