Error { using * Inner matrix dimensions must agree.} in EigenfaceCore.m ==> Eigenfaces = A * (L_eig_vec); % A: centered image vectors in main.m ==> [m, A, Eigenfaces] = EigenfaceCore(T);

This code was running well when the training image names were in sequential order, but i change their names into k132235_1.bmp , k132235_2.bmp etc

error occured !!!!

My code in EigenFaceCore.m

function [m, A, Eigenfaces] = EigenfaceCore(T)

m = mean(T,2);

Train_Number = size(T,2);

display(Train_Number);

A = [];  

for i = 1 : Train_Number

      temp = double(T(:,i)) - m; % Computing the difference image for each image in the training set Ai = Ti - m

      A = [A temp]; % Merging all centered images

end

L = A'*A; % L is the surrogate of covariance matrix C=A*A'.

[V, D] = eig(L); % Diagonal elements of D are the eigenvalues for both L=A'*A and C=A*A'.

L_eig_vec = [];

for i = 1 : size(V,2) 

      if( D(i,i)>1 )

          L_eig_vec = [L_eig_vec V(:,i)];

      end

end

Eigenfaces = A * (L_eig_vec);

[waqas siddiqui]

SIZE OF A = 10304*10 SIZE OF L_eig_vec = 0 0 HOW

Answer 1

[Walter Roberson]

Your L matrix might possibly be positive-definite so its eigenvalues might be non-negative, but they are not restricted to being greater than 1, so it is possible that none of your D elements are > 1, so you might not be putting anything into L before you do the multiplication.

[waqas siddiqui]

when i am subtracting my T from m it results all 0's how this can be ?

m = mean(T,2); // mean has positve values

Train_Number = size(T,2); // positive

A = [];

for i = 1 : Train_Number

      temp = double(T(:,i)) - m; // this returns 0 even though T is also positive

      A = [A temp]; % Merging all centered images

end