How can I get the elements (r,c) in the GLCM?

I am developing my thesis for my master degree and I am working with images in gray-levels

So, I know that using graycoprops I can get the properties of the GLCM.

"graycoprops normalizes the gray-level co-occurrence matrix (GLCM) so that the sum of its elements is equal to 1. Each element (r,c) in the normalized GLCM is the joint probability occurrence of pixel pairs with a defined spatial relationship having gray level values r and c in the image. graycoprops uses the normalized GLCM to calculate properties". stats = graycoprops(glcm, properties)

I need to get for every property the elements (r,c) of the GLCM. As an output I have to get a list value for each property respect each gray-level.

I got a code from the web that in theory "do" the mentioned before, but this code was wrote in Matlab R14 , and when I try to run this code in matlab 7.8 R2009a , I get an error.

• ??? Undefined function or method 'ParseInputs' for input arguments of type 'cell'. *

Following a part of the code:

function stats = GraycoProps(varargin) %%A lot of comments

allStats = {'Contrast','Correlation','Energy','Homogeneity'};

[glcm, requestedStats] = ParseInputs(allStats, varargin{:}); %THE ERROR IS HERE**

% Initialize output stats structure.

numStats = length(requestedStats);

numGLCM = size(glcm,3);

empties = repmat({zeros(1,numGLCM)},[numStats 1]);

stats = cell2struct(empties,requestedStats,1);

for p = 1 : numGLCM

if numGLCM ~= 1 %N-D indexing not allowed for sparse. 
 tGLCM = normalizeGLCM(glcm(:,:,p));
else 
 tGLCM = normalizeGLCM(glcm);
end

% Get row and column subscripts of GLCM.  These subscripts correspond to the
% pixel values in the GLCM.
s = size(tGLCM);
[c,r] = meshgrid(1:s(1),1:s(2));
r = r(:);
c = c(:);

% Calculate fields of output stats structure.

for k = 1:numStats

    name = requestedStats{k};

    switch name
     case 'Contrast'
      stats.(name)(p) = calculateContrast(tGLCM,r,c);

     case 'Correlation'
      stats.(name)(p) = calculateCorrelation(tGLCM,r,c);

     case 'Energy'
      stats.(name)(p) = calculateEnergy(tGLCM);

     case 'Homogeneity'
      stats.(name)(p) = calculateHomogeneity(tGLCM,r,c);
    end
  end

end

• How can I rewrite this code in my matlab version? *

[Edgar Corzo]

Of course! I called graycomatrix as you can see below...

However, I got the same error.

function stats = GraycoProps(varargin) %%A lot of comments

    * *I = imread('I:\pos\TESIS\...\PSM2-74538-0.038.TIF');

GLCM = graycomatrix(I,'G',[1 256],'Numlevels',256,'Offset',[0 1;-1 1;-1 0;-1 -1]);* * *

allStats = {'Contrast','Correlation','Energy','Homogeneity'};

[glcm, requestedStats] = ParseInputs(allStats, varargin{:}); %THE ERROR IS HERE**

% Initialize output stats structure.

numStats = length(requestedStats);

numGLCM = size(glcm,3);

empties = repmat({zeros(1,numGLCM)},[numStats 1]);

stats = cell2struct(empties,requestedStats,1); ... ...

To explain me better, for my output I need to get a kind of file, may be .txt or .xls, with the following information:

gray-levels 'Contrast' 'Correlation' 'Energy' 'Homogeneity'

   0         ValueCon(0)  ValueCor(0)   ValueEne(0)   ValueHom(0)
   1
   2
...
  255        ValueCon(255)  ValueCor(255)  ValueEne(255)   ValueHom(255)

In order to generate a plot (X,Y,Z), for example contrast vs correlation vs gray-level.

Thank you in advanced.

Answer 1

[Image Analyst]

You need to call graycomatrix() to get the glcm matrix. See attached demo, below in blue text.

[Edgar Corzo]

Thank you for your response

Of course! I called graycomatrix as you can see below...

However, I got the same error.

function stats = GraycoProps(varargin) %%A lot of comments

    * *I = imread('I:\pos\TESIS\...\PSM2-74538-0.038.TIF');

GLCM = graycomatrix(I,'G',[1 256],'Numlevels',256,'Offset',[0 1;-1 1;-1 0;-1 -1]);* * *

allStats = {'Contrast','Correlation','Energy','Homogeneity'};

[glcm, requestedStats] = ParseInputs(allStats, varargin{:}); %THE ERROR IS HERE**

% Initialize output stats structure.

numStats = length(requestedStats);

numGLCM = size(glcm,3);

empties = repmat({zeros(1,numGLCM)},[numStats 1]);

stats = cell2struct(empties,requestedStats,1);

for p = 1 : numGLCM

if numGLCM ~= 1 %N-D indexing not allowed for sparse.

 tGLCM = normalizeGLCM(glcm(:,:,p));
else 
 tGLCM = normalizeGLCM(glcm);
end
% Get row and column subscripts of GLCM.  These subscripts correspond to the
% pixel values in the GLCM.
s = size(tGLCM);
[c,r] = meshgrid(1:s(1),1:s(2));
r = r(:);
c = c(:); 
% Calculate fields of output stats structure.

for k = 1:numStats

    name = requestedStats{k};
    switch name
     case 'Contrast'
      stats.(name)(p) = calculateContrast(tGLCM,r,c);
     case 'Correlation'
      stats.(name)(p) = calculateCorrelation(tGLCM,r,c);
     case 'Energy'
      stats.(name)(p) = calculateEnergy(tGLCM);
     case 'Homogeneity'
      stats.(name)(p) = calculateHomogeneity(tGLCM,r,c);
    end
  end
end

To explain me better, for my output I need to get a kind of file, may be .txt or .xls, with the following information:

gray-levels 'Contrast' 'Correlation' 'Energy' 'Homogeneity'

   0         ValueCon(0)  ValueCor(0)   ValueEne(0)   ValueHom(0)
   1
   2
...
  255        ValueCon(255)  ValueCor(255)  ValueEne(255)   ValueHom(255)

In order to generate a plot (X,Y,Z), for example contrast vs correlation vs gray-level.

Thank you in advanced.

Answer 2

[Edgar Corzo]

Thank you for your response

of course! I called graycomatrix as you can see below...

However, I got the same error.

function stats = GraycoProps(varargin) %%A lot of comments

• * *I = imread('I:\pos\TESIS\...\PSM2-74538-0.038.TIF');

GLCM = graycomatrix(I,'G',[1 256],'Numlevels',256,'Offset',[0 1;-1 1;-1 0;-1 -1]);* * *

allStats = {'Contrast','Correlation','Energy','Homogeneity'};

[glcm, requestedStats] = ParseInputs(allStats, varargin{:}); %THE ERROR IS HERE**

% Initialize output stats structure.

numStats = length(requestedStats);

numGLCM = size(glcm,3);

empties = repmat({zeros(1,numGLCM)},[numStats 1]);

stats = cell2struct(empties,requestedStats,1);

for p = 1 : numGLCM

if numGLCM ~= 1 %N-D indexing not allowed for sparse.

 tGLCM = normalizeGLCM(glcm(:,:,p));
else 
 tGLCM = normalizeGLCM(glcm);
end
% Get row and column subscripts of GLCM.  These subscripts correspond to the
% pixel values in the GLCM.
s = size(tGLCM);
[c,r] = meshgrid(1:s(1),1:s(2));
r = r(:);
c = c(:); 
% Calculate fields of output stats structure.

for k = 1:numStats

    name = requestedStats{k};
    switch name
     case 'Contrast'
      stats.(name)(p) = calculateContrast(tGLCM,r,c);
     case 'Correlation'
      stats.(name)(p) = calculateCorrelation(tGLCM,r,c);
     case 'Energy'
      stats.(name)(p) = calculateEnergy(tGLCM);
     case 'Homogeneity'
      stats.(name)(p) = calculateHomogeneity(tGLCM,r,c);
    end
  end
end

To explain me better, for my output I need to get a kind of file, may be .txt or .xls, with the following information:

gray-levels 'Contrast' 'Correlation' 'Energy' 'Homogeneity'

   0                ValueCon(1)          ValueCor(1)          ValueEne(1)             ValueHom(1)
   1
   2
...
  255              ValueCon(255)         ValueCor(255)       ValueEne(255)          ValueHom(255)

In order to generate a plot (X,Y,Z), for example contrast vs correlation vs gray-level.

Thank you in advanced.