View License

Download apps, toolboxes, and other File Exchange content using Add-On Explorer in MATLAB.

» Watch video

Highlights from
Graph Cut for image Segmentation

2.2 | 9 ratings Rate this file 76 Downloads (last 30 days) File Size: 1.77 KB File ID: #40669 Version: 1.1
image thumbnail

Graph Cut for image Segmentation


Amarjot (view profile)


07 Mar 2013 (Updated )

The code segments the grayscale image using graph cuts.

| Watch this File

File Information

Running the Code:
% Maximum Image size 128 by 128

[Ncut] = graphcuts(I,pad,MAXVAL)
% function [Ncut] = graphcuts(I)
% Input: I image
% pad: spatial connectivity; eg. 3
% MAXVAL: maximum image value
% Output: Ncut: Binary map 0 or 1 corresponding to image segmentation

Required Products Image Processing Toolbox
MATLAB release MATLAB 7.6 (R2008a)
MATLAB Search Path
Tags for This File   Please login to tag files.
Please login to add a comment or rating.
Comments and Ratings (19)
24 Nov 2016 Jinny

Jinny (view profile)

I wouldn't use this code if I were you. Poorly perform

Comment only
09 Nov 2016 mog sam

07 Nov 2016 PL Yang

12 Aug 2016 Xiangfei

This is a very poor and extremely inefficient implementation of the normalized graph cut algorithm.

17 Jun 2016 ebrahim nasr

16 Apr 2016 Prerna Surbhi

hello, i tried implement the code on matlab.i am getting error on line 7
I = double(I); [H,W] = size(I);
error :
Error using graph cuts (line 7)
Not enough input arguments.

12 Feb 2016 Amruthavakkula Shiva

Sir, I am not able to understand the construction of weight matrix. Can you provide reference or construct algorithm of steps of the provided program

15 Oct 2015 Bashir Isa Dodo

hello,I tried running the segmentation on a retinal image and this is the error I get. Could anyone drop some points on how to solve it or is it that the data is large for the algorithm to work on
Error using zeros
Requested 507904x507904 (1922.0GB) array exceeds maximum array size preference. Creation of arrays greater than this limit may
take a long time and cause MATLAB to become unresponsive. See array size limit or preference panel for more information.

Error in graphcuts (line 20)
WM = zeros(H*W,H*W); countWM = 0;

Comment only
03 Mar 2015 Rituraj Soni

I am trying to excute the code but it is givng error
Error in graphcuts (line 18)
WM = zeros(H*W,H*W); countWM = 0;
Max variable size exceeded

Comment only
08 Dec 2014 suji

suji (view profile)

Thank you for sharing!

Comment only
04 Dec 2014 Amarjot

Amarjot (view profile)

@chandra Please use the image size 128 by 128

Comment only
04 Dec 2014 Chandra Shaker Balure

I am getting an error at line:18, which has the instruction WM = zeros(H*W,H*W);... I am using an image of 256x320, so the zero matrix forming is giving the problem. Is it that I have to use some smaller sized image? I have tried generating zeros(256*256), it also give error. What is the workaround to solve this problem?

Comment only
17 Nov 2014 joy angel


Comment only
05 Jun 2014 Amarjot

Amarjot (view profile)

@matlabuser1 I assume you are passing the to the function.In that case you should not get any error. Can you send me the file in which you are calling the function at I can look into the error.

Comment only
05 Jun 2014 matlabuser1

Firstly, I am getting an error:

??? Input argument "I" is undefined.

Error in ==> graphcuts at 7
I = double(I); [H,W] = size(I);


Also, I am getting the same graph cut for ever image. Please help me out.


Comment only
05 Sep 2013 Wouter

Wouter (view profile)

You might want to add a input check / limitation to the max. input size of an image.
For a 400x400 image, this code requires approximately 200Gb of memory...

The code is quite compact though. Thanks for sharing.

13 May 2013 abbad

abbad (view profile)


Comment only
12 May 2013 Zhao

Zhao (view profile)

Why there is a problem when I run this program!!!
Out of memory. Type HELP MEMORY for your options.
Error in graphcuts (line 18)
WM = zeros(H*W,H*W); countWM = 0;

15 Apr 2013 Anton Semechko

Anton Semechko (view profile)

This is a very poor and extremely inefficient implementation of the normalized graph cut algorithm.

04 Dec 2014 1.1

I added the maximum image size for which the code runs without error.

Contact us