distmap {EBImage}R Documentation

Distance map transform of binary images

Description

Distance map transform of binary images.

Usage

    distmap(x, tolerance=0.05, minBG=0.05, strict=FALSE, ...)

Arguments

x
An object of Image-class in the Grayscale mode.
tolerance
A numeric value in the range [0,1). Image pixels that differ from background (zero) by less than this value will be considered as background. This enables image-wide single value threshloding if necessary.

minBG
An estimated minimum fraction of background. If the estimated number of background pixels is smaller than this value the algorithm will terminate to avoid unnecessarily lengthy computations. This is a protection mechanism in case a very noise image is supplied without first thresholding it.
sctrict
If TRUE, single-pixel diagonal distances are calculated exactly (i.e. sqrt(2)), otherwise these are set to be 1, same as neighbouring pixels on the vertical and horizontal axes.
...
Reserved for future use.

Value

An object of Image-class in the Grayscale mode with pixels containing floor of distances to the nearest background. If strict is set to TRUE, single-pixel diagonal distances are calculated exactly (i.e. sqrt(2)), otherwise these are set to be 1, same as neighbouring pixels on the vertical and horizontal axes.To preview the results visually, use display( normalize(result) ).

Details

Given a binary or grayscale image with zero background, a distance map transformed image contains in each pixel a distance to the closest background pixel (zero). This function is used extensively in object detection. Distance map is a usual input for watershed and other segmentation algorithms that are used to identify and separate closely positioned objects.

Author(s)

Copyright (c) 2005-2006 Oleg Sklyar : osklyar@ebi.ac.uk

References

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

See Also

Image-class, watershed, thresh

Examples

  ## Not run: see ?watershed for examples

[Package EBImage version 2.0.1 Index]