Image Edge Detection using Sobel Operator

I am working on an assignment where I have to detect image edge using Sobel operator. It was stated that the output image must be uint8 format and has two fewer rows and columns than the input one, since the pixels in first and last columns and rows do not have enough neughbors for the required computation. I am struggling in the last part as I get same input and ouput size. I would apperciate your help and suggestions. I have attached my code for your reference.

function [edge] = edgy(cc)
cc = double(cc);
H = size(cc,1);
W = size(cc,2);


edge = cc;
sx = [-1 0 1;-2 0 2;-1 0 1];
sy = [1 2 1;0 0 0;-1 -2 -1];

for i = 2 : H-1
   for j = 2 : W-1 
       
           A = cc(i - 1 : i + 1, j - 1 : j + 1);
           a = (sum(sx .* A));
           x = sum(a);
           
           b = (sum(sy .* A));
           y = sum(b);
           
           pixValue =sqrt(x.^2+ y.^2);
           
           edge(i, j) = pixValue;
       
   end
end 
edge = uint8(edge);
end 

[João Pedro de Freitas Coelho]

function edge = edgy(figure)
cc = double(figure);
sx = [-1 0 1; -2 0 2; -1 0 1];
sy = [1 2 1; 0 0 0; -1 -2 -1];
B = cc;
for ii = 2:(size(B,1)-1)
    for jj = 2:(size(B,2)-1)
        A = cc((ii-1):(ii+1),(jj-1):(jj+1));
            x = sx.*A;
            x = sum(sum(x));
            y = sy.*A;
            y = sum(sum(y));
            pixValue = uint8(sqrt(x.^2 + y.^2));
            B(ii,jj) = pixValue;
    end
end
B = uint8(B);
edge = B(2:end-1,2:end-1);
end

[Osama Ahmed]

Thank you very much João