segment {EBImage}R Documentation

Segmentation and edge detection

Description

Fucntions to segment images and detect edges.

Usage

  ## S4 method for signature 'Image':
  edge(x, r=0, ...)
  ## S4 method for signature 'Image':
  segment(x, cl=10, s=1.5, ...)
  ## FIXME: 'fill' implemented incorrectly because of unclear API
  ## S4 method for signature 'Image':
  fill(x, col, xoff, yoff, fuzz=10, method="floodfill", ...) 

Arguments

x An object of Image.
r The radius of the pixel neighbourhood to take into account. The 0 value enables automatic radius selection.
cl Minimum cluster size in pixels .
s The smoothing threshold.
col The fill colour in any format (character, gray value or integer byte-coded value).
xoff, yoff The point at which to start the fill.
fuzz Colour tolerance, within this interval the colours are considered the same.
method Fill method.
... Reserved.

Details

edge returns an image of edges between different colours in the original image. Most effective on binary images.

segment segment an image by analyzing the histograms of the color components and identifying units that are homogeneous with the fuzzy C-means technique (source and implementation ImageMagick). The smoothing threshold eliminates noise in the second derivative of the histogram. As the value is increased, you can expect a smoother second derivative.

fill should allow colour fill, but ImageMagick API is not clear and the current implementation does not seem to produce any reasonable result. Therefore the function will currently produce an error message, should be corrected in future releases.

Value

A transformed image in an object of Image.

Author(s)

Oleg Sklyar, osklyar@ebi.ac.uk, 2005-2006

References

ImageMagick: http://www.imagemagick.org.

See Also

Image


[Package EBImage version 2.2.0 Index]