CODEWITHSUNDEEP
matlabimage-processingfeature-extraction
Tariq Khalifa
Tariq Khalifa

Reputation: 152

How to save feature value from histogram of LBP image in Matlab?

I'm using Local Binary Pattern (LBP) to extract the features of group of images (500 images in Training folder and 100 Images in Test folder). Indeed, I had extracted these features successfully but I'm not sure whether they saved in correct way or not.

Here is a part of code that extract the features:

for x = 1:total_images
% Specify images names with full path and extension    
full_name= fullfile(test_set, filenames(x).name);

% Read images from Training folder
I2 = imread(full_name);
I3=I2;
m=size(I2,1);
n=size(I2,2);

for i=2:m-1
    for j=2:n-1
        c=I2(i,j);
        I3(i-1,j-1)=I2(i-1,j-1)>c;
        I3(i-1,j)=I2(i-1,j)>c;
        I3(i-1,j+1)=I2(i-1,j+1)>c; 
        I3(i,j+1)=I2(i,j+1)>c;
        I3(i+1,j+1)=I2(i+1,j+1)>c; 
        I3(i+1,j)=I2(i+1,j)>c; 
        I3(i+1,j-1)=I2(i+1,j-1)>c; 
        I3(i,j-1)=I2(i,j-1)>c;
        LBP (i,j) =I3(i-1,j-1)*2^7+I3(i-1,j)*2^6+I3(i-1,j+1)*2^5+ ...
            I3(i,j+1)*2^4+I3(i+1,j+1)*2^3+I3(i+1,j)*2^2+ ...
            I3(i+1,j-1)*2^1+I3(i,j-1)*2^0;
    end
end

featureMatrix {x} = hist(LBP,0:255); 

end

By using this code, I get LBP features of all images but I'm not sure about saving them correctly in a matrix. How to save feature value from this histogram of LBP image? I want to store this value for each image.

featureMatrix is a matrix that data will stored in. It should be consist of 500 rows, each row should has all features of each image.

Any answer will be appreciated.

Upvotes: 0

Views: 1994

Answers (1)

underscore_nico
underscore_nico

Reputation: 149

you should initialize the feature matrix before entering the outer loop (if you could know the size of LBP):

featureMatrix = zeros(total_images,size_LBP); % where size_LBP is the number of columns of LBP.

then replace featureMatrix {x} = hist(LBP,0:255); in the loop with:

featureMatrix(x,:) = hist(LBP,255);

I hope this works for you!

Upvotes: 1

Related Questions