Up: Component summary Component

Watershed

Methods to divide objects in a mask, or grayscale images. The method choices are:

The input mask may also be grayscale image, but the minsize/maxsize will not work with such images.

Version 2.0
Bundle anima
Categories Image Processing
Authors Ville Rantanen (ville.rantanen@helsinki.fi)
Issue tracker View/Report issues
Requires Matlab
Source files component.xml imagews.m voronoicut.m delaunaycut.m execute.m
Usage Example with default values

Inputs

Name Type Mandatory Description
images ImageList Mandatory A directory with grayscale or BW image files.
seed ImageList Optional A directory of seed masks, used in Voronoi method.

Outputs

Name Type Description
mask ImageList BW PNG cut mask images.
lines ImageList BW PNG images of the lines used for watershed.
perimeter ImageList BW PNG images of the perimeter of the mask.

Parameters

Name Type Default Description
blur int 0 Sigma of gaussian filter for blurring the images. 0 to disable. Applicable to ds and ws methods.
invert boolean false Invert image before watershed.
maxArea int 0 Maximum accepted pixel area in all methods. Use 0 for no limit.
method string "ds" Method for image watershed. Choose from 'ds', 'ws', 'vo', 'de'.
minArea int 0 Minimum accepted pixel area in all methods. Works only with BW mask images.
sharpen int 0 Sigma of gaussian filter for sharpening the images via unsharp mask. 0 to disable.

Test cases

Test case Parameters IN
images
IN
seed
OUT
mask
OUT
lines
OUT
perimeter
case1 properties images (missing) mask lines perimeter

method=ds,
minArea=0,
maxArea=0,
blur=2

case2_voronoi properties images seed mask lines perimeter

method=vo,
minArea=80,
maxArea=0


Generated 2019-02-07 07:42:15 by Anduril 2.0.0