How can i save the features of multiple images into a single mattrix (.mat file)?

i want to save : Each rows containing each image features. Following is my syntax for the training section.

f=[];
srcFiles = dir('E:\M\*.jpg');  
for i = 1 : 8
    filename = strcat('E:\Ms\',srcFiles(i).name);

   TrainNormal = [Contrast,Correlation,Energy,Homogeneity, Mean, Entropy, RMS, Variance, Smoothness, Skewness, IDM];

end

save TrainNormal.mat

[sam CP]

can i get the syntax?

Answer 1

[Image Analyst]

You might want to make TrainNormal a 2-D vector. Before the loop

TrainNormal = zeros(8, 11);

then in the loop, assign the 11 values to a single row.

TrainNormal(i, :) = [Contrast,Correlation,Energy,Homogeneity, Mean, Entropy, RMS, Variance, Smoothness, Skewness, IDM];

then after the loop, save the variable to a .mat file:

save('TrainNormal.mat', 'TrainNormal);

[sam CP]

TrainNormal = zeros(8, 11);

Here,

8 : for number of images for feature extraction
11: number of features

But some times the number of images taken for training will be varied..So how can i store each images into each column in an automated format?

[sam CP]

i got errors by using above code..hereby i have attached the .m file

Subscripted assignment dimension mismatch.

Error in TrainedNoramalMRI (line 117) TrainNormalMRI(i,:) = [Contrast,Correlation,Energy,Homogeneity, Mean, Standard_Deviation, Entropy, RMS, Variance, Smoothness, Kurtosis, Skewness, IDM];

[Image Analyst]

Some of those values are not scalars. I didn't know before since you hadn't posted the code until now. You need to make sure you have enough elements, when you preallocate with zeros(), to contain all the elements you are concatenating together.

[sam CP]

okay

[elz tomz]

Thanks, image analyst..it works fine. You save ma time.

Answer 2

[Sagar V]

how to save the features of 100 images into .mat file to classifying using SVM