Up: Component summary Component

SegmentGraphCut

Segmentation based on graph cutting. Suitable especially for nucleus segmentation.

The component uses a tool found in FARSight: http://www.farsight-toolkit.org/wiki/Nuclear_Segmentation

To produce the result images, ImageMagick is required!

Version 1.1
Bundle anima
Categories Image Processing
Authors Ville Rantanen (ville.rantanen@helsinki.fi)
Issue tracker View/Report issues
Requires bash ; imagemagick (DEB) ; libqtgui4 (DEB) ; FARSight segment_nuclei ; installer (bash)
Source files component.xml segment_run.sh
Usage Example with default values

Inputs

Name Type Mandatory Description
in ImageList Mandatory A directory with grayscale image files. Images will be converted to 8bit grayscale, you may have to auto-level brightness of 16bit images.

Outputs

Name Type Description
mask ImageList BW PNG images.
perimeter ImageList BW PNG images of the perimeter of the mask.

Parameters

Name Type Default Description
UseDistanceMap boolean true a boolean parameter used to enable the use of the distance constraint to select the LoG scales
binary string "segment_nuclei" Binary to use. You can also provide the full path to the binary.
continueNext boolean true Continue to next image after failure, or timeout. Produces an empty mask
finalizeSegmentation boolean true a boolean parameter used to enable the segmentation refinement step
highSensitivity boolean false a boolean parameter used to enable high sensitivity binarization. Default value is false.
maxScale float 6 the maximum scale of the Laplacian of Gaussian filter.
minScale float 4 the minimum scale of the Laplacian of Gaussian filter.
refinementRange int 5 If segmentation finalization (refinement) is enabled, this parameter sets the maximum distance that an initial contour can be shifted
samplingRatio int 2 ratio of image sampling along Z to the sampling in X-Y
timeout string "30m" Timeout individual image segmentation process. Express as seconds, or with suffixes: s,m,h or d; e.g. "5m" for 5 minutes.
xyClusteringRes int 2 resolution of the local maximum clustering (in the x-y plane) used in the initial binarization step
zClusteringRes int 1 resolution of the local maximum clustering (along z) used in the initial binarization step

Test cases

Test case Parameters IN
in
OUT
mask
OUT
perimeter
case1 properties in mask perimeter

minScale=7,
maxScale=10,
highSensitivity=true


Generated 2019-02-08 07:42:07 by Anduril 2.0.0