what i can do to solve this problem

IMG = imread('X_Ray_Chest_Side.png');

origin=rgb2gray(IMG);
[M,N] = size(IMG);
figure,
imshow(origin);
title('origin image');


E=ones(5,5);
E_lpf=(1/25)*E;
f_padded=double(padarray(origin,[2 2]));
g_padded=(zeros(M+4,N+4));
for i=3:M+2
    for j=3:N+2
        T=f_padded(i-2:i+2,j-2:j+2);
        conv_T=T.*E_lpf;
        g_padded(i,j)=sum(sum(conv_T));
    end
end
EEE=g_padded(2:M+1,2:N+1);
figure
imshow(uint8(EEE))


W_2=[4 8 16 32 16 8 4;8 16 32 64 32 16 8;16 32 64 128 64 32 16;32 64 128 256 128 64 32;16 32 64 128 64 32 16;8 16 32 64 32 16 8;4 8 16 32 16 8 4];
W_Gaus=(1/1936)*W_2;
g2=(zeros(m+4,n+4));
for q1=3:(m-2)
    for p1=3:(n-2)
        origin=add(q1-2:q1+2,p1-2:p1+2);
        org=double(origin);
                res=W_Gaus .*org;
        g2(q1,p1)=sum(sum(res)); 
    end
end
g2=g2(3:m+2,3:n+2);
imshow(uint8(g2));
title(' gaussian filter ');


W_composite=[0 -1 0;-1 5 -1;0 -1 0];
f2=double(add2);
g3=(zeros(m+6,n+6));
for i=1:m-3
    for j=1:n-3
        temp3=f2(i:i+2,j:j+2);
        conv_temp3=temp3.*W_composite;
        g3(i,j)=sum(sum(conv_temp3));
    end
end
g3=g3(4:m+3,4:n+3);
imshow(uint8(g3));
title('Composite laplacian filter');

[Walter Roberson]

What problem?

You have presented a bunch of undocumented code (and no copy of the input image). We must assume that the purpose of the code is to do exactly whatever it is that the code does, and that therefore the code is perfect.

[Rony ali]

this is my homework i try to write the code correctly but i cant complete and i have a mistakes and i dont know how to solve it ....th

[Rony ali]

this is the error

Index in position 2 exceeds array bounds (must not exceed 1200).

Error in (line 15)

T=f_padded(i-2:i+2,j-2:j+2);

Answer 1

[Walter Roberson]

IMG = imread('X_Ray_Chest_Side.png');
origin=rgb2gray(IMG);
[M,N] = size(IMG);
%...
g2=(zeros(m+4,n+4));

You define variables M and N, but you do not define variables m or n, but you expect to be able to use values up to m or n as indices even though whatever values for them you happen to have in your workspace might be totally unrelated to the size of the image.

However, if you were to recode m as M and n as N then you would still have a problem. You would not be calling rgb2gray(IMG) unless IMG is rgb, but when you take

[M,N] = size(IMG);

you are asking for two output variables to store the size information about the 3 dimensional array IMG. MATLAB does define the result, but the result of

[M,N] = size(IMG);

for a 3D array is not the same as

temp = size(IMG);
M = temp(1); N = temp(2);

Instead, the result is the same as

temp = size(IMG);
M = temp(1);
N = prod(temp(2:end));

for example if IMG is 1024 x 1200 x 3 (because it is RGB) then [M,N] = size(IMG) will give M = 1024, N = (1200*3)

It is recommended that in MATLAB you get in the habit of only using size() with one output, either in the form

all_sizes = size(ARRAY)

to return all of the dimensions at once, or else in the form

specific_size = size(ARRAY, dimension_number)

In this particular case, the problem would also have been avoided if you had used

[M,N] = size(origin);

Answer 2

[Rony ali]

[M,N] = size(origin);

i use it function but now what i can do about undefined values m,n

[Rony ali]

THIS IS THE NEW ERROR

Undefined function 'add' for input arguments of type 'double'.

Error in rawand (line 32)

origin=add(q1-2:q1+2,p1-2:p1+2);

>>

[Walter Roberson]

Perhaps add should be g_padded?