denoise {EBImage} | R Documentation |
Functions to enhance image colours.
denoise(x, r=0, ...) mediansmooth(x, r=2, ...) despeckle(x, ...) sharpen(x, r=0, s=0.5, ...) umask(x, r=0, s=0.5, amount=5, t=2, ...) blur(x, r=0, s=0.5, ...) gblur(x, r=0, s=0.5, ...) noise(x, type="gaussian", ...)
x
Image-class
. r
s
sharpen
) or
Gaussian (umask, blur, gblur
), in pixels. For reasonable results of
most functions, r
must be larger than s
. amount
t
type
...
A transformed image in an object of Image-class
.
despeckle
reduces the speckle noise.
mediansmooth
smoothes a noisy image by replacing each pixel value by
the median in a set of neighbouring pixels as defined by radius.
blur, gblur
blur an image. blur
differs from Gaussian blur,
gblur
, in that it uses a separable kernel which is faster but
mathematically equivalent to the non-separable kernel (implemented in
ImageMagick
).
sharpen, umask
sharpen an image. umask
uses the unsharp mask
algorithm. The the image is convolved with a Gaussian operator of the given
radius and standard deviation, s
.
noise
adds noise to an image. Supported noise types are:
'uniform', 'gaussian'
(default) , 'multi', 'impulse',
'laplace'
and 'poisson'
.
Copyright (c) 2005-2006 Oleg Sklyar : osklyar@ebi.ac.uk
ImageMagick: http://www.imagemagick.org.