WOLFRAM

Wolfram Language & System Documentation Center

GrowCutComponents[image,{marker1,marker2,…}]

creates a segmentation from image by growing each markeri.

GrowCutComponents[video,…]

returns segmentation for each frame in video.

Details and Options
Details and Options Details and Options
Examples  
Basic Examples  
Scope  
Options  
CornerNeighbors  
MaxIterations  
Applications  
Properties & Relations  
See Also
Related Guides
History
Cite this Page

GrowCutComponents

GrowCutComponents[image,{marker1,marker2,…}]

creates a segmentation from image by growing each markeri.

GrowCutComponents[video,…]

returns segmentation for each frame in video.

Details and Options

  • GrowCutComponents models the image using cellular automata where the automata evolution leads to an image segmentation.
  • GrowCutComponents works with arbitrary 2D and 3D images.
  • Each marker markeri can be given either as an image or a list of points in the standard image coordinate system.
  • The following options can be specified:
  • CornerNeighbors Truewhether to include corner neighbors
    MaxIterations Automaticmaximum number of iterations to use
  • With MaxIterations->Automatic the algorithm runs until convergence.

Examples

open all close all

Basic Examples  (1)

Grow-cut segmentation of an image:

Wolfram Language code: GrowCutComponents[[image], {[image], [image], [image]}]//Colorize

Scope  (2)

Grow-cut binarization of a color image:

Wolfram Language code: c = GrowCutComponents[i = [image], {[image], [image]}]; Colorize[c]

Mask the image with the detected foreground area:

Wolfram Language code: SetAlphaChannel[i, Image[c - 1]]

Segment a 3D image:

Wolfram Language code: img = [image]; m1 = Image3D[DiskMatrix[10, ImageDimensions[img]]];m2 = ImagePad[ConstantImage[0, ImageDimensions[img] - {2, 2, 1}], {{1, 1}, {1, 1}, {0, 1}}, 1];
Wolfram Language code: ArrayComponents[GrowCutComponents[img, {m2, m1}], 3, {1 -> 0, 2 -> 1}]//Image3D

Options  (2)

CornerNeighbors  (1)

By default, CornerNeighbors->True is used:

Wolfram Language code: i = [image]; GrowCutComponents[i, {[image], [image]}]//Colorize

Use CornerNeighbors->False:

Wolfram Language code: GrowCutComponents[i, {[image], [image]}, CornerNeighbors -> False]//Colorize

MaxIterations  (1)

By default, the algorithm is iterated until convergence:

Wolfram Language code: img = [image];m1 = [image];m2 = [image]; GrowCutComponents[img, {m1, m2}]//Colorize

Specify the number of iterations:

Wolfram Language code: img = [image];m1 = [image];m2 = [image]; GrowCutComponents[img, {m1, m2}, MaxIterations -> 10]//Colorize

Applications  (3)

Use grow-cut segmentation to separate foreground and background in a complex image:

Wolfram Language code: i = [image];
Wolfram Language code: markers = {[image], [image]};
Wolfram Language code: c = GrowCutComponents[i, markers]; SetAlphaChannel[i, Image[c - 1]]

Separate foreground and background in an image:

Wolfram Language code: i = [image];
Wolfram Language code: bg = [image];fg = [image];
Wolfram Language code: mask = Image[GrowCutComponents[i, {bg, fg}] - 1]

Use the binary mask as an alpha channel:

Wolfram Language code: SetAlphaChannel[i, mask]

Blur the binary mask for a smoother separation of foreground and background:

Wolfram Language code: SetAlphaChannel[i, Blur[mask]]

Place the detected foreground on a different background:

Wolfram Language code: i = [image];
Wolfram Language code: m1 = [image];m2 = [image];

Detect the foreground:

Wolfram Language code: fg = SetAlphaChannel[i, Image[GrowCutComponents[i, {m1, m2}] - 1]]

Compose image foreground to a new background:

Wolfram Language code: background = [image];
Wolfram Language code: ImageCompose[background, fg]

Properties & Relations  (3)

Switching the foreground and background marker will result in the complement mask:

Wolfram Language code: i = [image];
Wolfram Language code: m1 = [image];m2 = [image];
Wolfram Language code: GrowCutComponents[i, {m1, m2}]//Colorize
Wolfram Language code: GrowCutComponents[i, {m2, m1}]//Colorize

Not all pixels are necessarily segmented:

Wolfram Language code: i = [image];
Wolfram Language code: marker1 = [image];marker2 = [image];
Wolfram Language code: GrowCutComponents[i, {marker1, marker2}]//Colorize

Use ArrayComponents to relabel the array and convert the label matrix into an Image object:

Wolfram Language code: c = GrowCutComponents[i = [image], {[image], [image]}];
Wolfram Language code: ArrayComponents[c, 2, {1 -> 0, 2 -> 1}]//Image
Wolfram Research (2014), GrowCutComponents, Wolfram Language function, https://reference.wolfram.com/language/ref/GrowCutComponents.html (updated 2025).

Text

Wolfram Research (2014), GrowCutComponents, Wolfram Language function, https://reference.wolfram.com/language/ref/GrowCutComponents.html (updated 2025).

CMS

Wolfram Language. 2014. "GrowCutComponents." Wolfram Language & System Documentation Center. Wolfram Research. Last Modified 2025. https://reference.wolfram.com/language/ref/GrowCutComponents.html.

APA

Wolfram Language. (2014). GrowCutComponents. Wolfram Language & System Documentation Center. Retrieved from https://reference.wolfram.com/language/ref/GrowCutComponents.html

BibTeX

@misc{reference.wolfram_2026_growcutcomponents, author="Wolfram Research", title="{GrowCutComponents}", year="2025", howpublished="\url{https://reference.wolfram.com/language/ref/GrowCutComponents.html}", note=[Accessed: 13-June-2026]}

BibLaTeX

@online{reference.wolfram_2026_growcutcomponents, organization={Wolfram Research}, title={GrowCutComponents}, year={2025}, url={https://reference.wolfram.com/language/ref/GrowCutComponents.html}, note=[Accessed: 13-June-2026]}