Convert complex double to double type in Matlab

Question

I am doing a feature extraction using glcm method. The glcm output are in 'struct' type, while I need the output in double type.

So, I've tried to convert it using several code showed below.

To get Fetest1 code:

srcFile = dir('D: datatest\*.png');
fetest1 = []; %or fetrain1
for b = 1:length(srcFile)
    file_name = strcat('D:datatest\',srcFile(b).name);
    B = imread(file_name);
%   [fiturtest] = feature_extractor (B);
    [g] = glcm (B);
    [g] = struct2cell (g);
    [fiturtest] = cell2mat (g);  %fiturtrain
%   [c] = CobaDCT (A);
%   [fitur] = cobazigzag(c);
%   arr(:,a) = fitur;
    fetest1 = [fetest1 fiturtest];  %fiturtrain
%   vectorname = strcat(file_name,'_array.mat');

end
 save ('fetest1.mat','fetest1'); %fetrain1

To get Fetrain1 code:

srcFiles = dir('D:datatrain\*.png');
fetrain1 = [];
for a = 1:length(srcFiles)
    file_name = strcat('D:datatrain\',srcFiles(a).name);
    A = imread(file_name);
    [fiturtrain] = feature_extractor (A);
%   [c] = CobaDCT (A);
%   [fitur] = cobazigzag(c);
%     fiturtrain (:,a) = fiturtrain ;
    fetrain1 = [fetrain1 fiturtrain];
%   vectorname = strcat(file_name,'_array.mat');
end
 save ('fetrain1.mat','fetrain1');

The output of the whole process is fetrain1 and fetest1 variables. I run the same code to get fetest1 and fetrain1, but fetest1 is in 'double' type, while fetrain is in 'complex double' type.

and

I need to convert fetrain1 from 'complex double' type into 'double' type, so I can use the variable for the next step. Training step using Neural Network method.

Any suggestion would be very appreciated.

The absolute value of a complex number is returned by `abs`. The real part and imaginary part are returned by `real` and `imag`, respectively. The _real_ question is: what would you choose, and why? — buzjwa, May 19 '16 at 10:44
I need your help for Arabic character segmentation, can I have your email? — Hamed, May 20 '16 at 14:10

score 0 · Answer 1 · answered May 19 '16 at 11:04

If you are estracting features by the GLCM method you can do:

srcFile = dir('D: datatest\*.png');
fetest1 = []; %or fetrain1

for b = 1:length(srcFile)
    file_name = strcat('D:datatest\',srcFile(b).name);
    B = imread(file_name);

    imG = rgb2gray(B);

    % create gray-level co-occurrence matrix from image
    gimg = graycomatrix(imG);

    % extract properties of gray-level co-occurrence matrix
    stats = graycoprops(gimg, 'all');

    features = [stats.Energy stats.Homogeneity stats.Correlation stats.Contrast];

    fetest1 = cat(1, fetest1, features);
end

thanks for the suggestion. But, I've already have GLCM function, the original source code is from here http://www.mathworks.com/matlabcentral/fileexchange/22187-glcm-texture-features What I need now, is to get the appropriate output. fetrain1 in 'double' type. So, I can use it in the next step. — Ana Ain, May 19 '16 at 18:28

Convert complex double to double type in Matlab

1 Answers1