error !! horzcat CAT arguments dimensions are not consistent.

hi sir i am very much new to matlab ..i have two images ..i want to use "imoverlay" for the images but whenever i run it ,i get error saying horzcat arguments dimension are consistent ..i am working on retinal image analysis..belo i have uploaded my code and error.sorry for not providing sufficient info error is" horzcat CAT arguments dimensions are not consistent.

Error in ==> file1>filter_Callback at 692 C = horzcat(I2,1);"

[Matt Kindig]

What are the sizes of

I1 and I2?

[Jan]

Please take into account that problem is not urgent. The fact, that you have to submit it tomorrow does not encourage people to stop their work and care about your problem at first.

If you really want the forum to help yout, remove the "urgent" and insert a copy of the complete error message, instead of the short paraphrased version. Then we would not have to guess, where the problem occurres.

[vidya]

sorry sir :-( :-(

Answer 1

[Vivek Selvam]

Vidya, you can solve your padding problem by

1. switching vertcat and horzcat (reasoning in the comment below)
2. image is 3 dimensional matrix (the 3rd dimension for the rgb color)

Here is the fixed code:

    %Pad the columns 
    if size(I1,1) > size(I2,1) %I1 has more rows so pad I2
            pad = zeros (size(I1,1) - size(I2,1), size(I2,2),size(I2,3));
            %I2 = [I2 ; pad]; %Append the rows of zeros to the bottom of I2
            I2 = vertcat(I2,pad); % vertical concatenation as you pad rows
    else %I2 has more rows so pad I1
            pad = zeros (size(I2,1) - size(I1,1), size(I1,2),size(I1,3));
            %I1 = [I1 ; pad]; %Append the rows of zeros to the bottom of I1
            I1 = vertcat(I1,pad); % vertical concatenation as you pad rows
    end
    %Pad the columns    
    if size(I1,2) > size(I2,2) %I1 has more rows so pad I2
            pad = zeros (size(I2,1), size(I1,2) - size(I2,2),size(I2,3));
            %I2 = [I2 , pad]; %Append the columns of zeros to the left of I2
            I2 = horzcat(I2,pad); % horizontal concatenation as you pad columns
    else %I2 has more rows so pad I1
            pad = zeros (size(I1,1), size(I2,2) - size(I1,2),size(I1,3));
           % I1 = [I1 , pad]; %Append the columns of zeros to the left of I1
            I1 = horzcat(I1,pad); % horizontal concatenation as you pad columns
    end
    ovr1=imoverlay(I1,I2,[0,0,0]); % overlaying updated images

[vidya]

thank you so much sir...m relieved...but sir its not getting overlaid properly i mean to say the black image should exactly fit in the green image(exactly on the bright regions) where can i adjust the settings. it shoud be like on the image paper..in that 12th step

[Vivek Selvam]

What is the image paper?

[vidya]

sir its the paper i am implementing...i completed 10 steps of it now i am in 11 step

[Vivek Selvam]

You can check out the detailed blog on this subject here.

 %%ML Answers
 fig1 = imread('gg1.jpg');                  % original
 fig2 = rgb2gray(imread('gff.jpg'));        % gray - influence map

 % Display the original image
 figure; imshow(fig1, 'InitialMag', 'fit'); 

 % Make a truecolor all-blue image.
 [m n] = size(fig2);
 B = cat(3, zeros(m,n), zeros(m,n), ones(m,n));
 hold on
 h = imshow(B); 

 % Use our influence map as the AlphaData for the solid blue image.
 set(h, 'AlphaData', fig2)

Answer 2

[Jan]

Let me guess, where the problem is:

        pad = zeros (size(I1,1) - size(I2,1), size(I2,2));
        %I2 = [I2 , pad]; %Append the rows of zeros to the bottom of I2
        C = horzcat(I2, pad);

Here the 1st dimension of pad does not match, because it must be size(I2, 1).

Guessing is inefficient. Please do not post a question without showing the complete error message in the future. Thanks!

[vidya]

sir i did the same ..but still the error is same...