How to solve the Issue with subimages having odd rows and columns, block error?

Question

Mei Synn Tan 2017-4-13

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/335177-how-to-solve-the-issue-with-subimages-having-odd-rows-and-columns-block-error

回答： Image Analyst 2017-4-13

    Image = ('13100.jpg'); 
    Img = imread (Image);
    [oldx oldy z] = size(Img);
    x = oldx/4;
    y = oldy/4;
    if mod(x, 2) ~= 0
        x = x + 1;
    end
    if mod(y, 2) ~= 0
        y = y + 1;
    end
    Img2 = uint8(zeros([x*4 y*4 z]));
    Img2(1:oldx, 1:oldy, 1:z) = Img(1:oldx, 1:oldy, 1:z);
    Img = Img2;
    Image = rgb2gray(im2double(Img));
    % Initialize the edge detection masks
    H = [];
     HorizontalMask = [1 1 1;0 0 0;-1 -1 -1];
     VerticalMask   = [1 0 -1;1 0 -1;1 0 -1];
     DiagonalMask   = [0 1 1;-1 0 1;-1 -1 0];
     AntiDiagMask   = [1 1 0;1 0 -1;0 -1 -1];
    % Perform the filtering using the masks to create the edge images
    HImage = imfilter(Image, HorizontalMask,'replicate');
    VImage = imfilter(Image, VerticalMask,'replicate');
    DImage = imfilter(Image, DiagonalMask,'replicate');
    AImage = imfilter(Image, AntiDiagMask,'replicate');
    % Determine the size of each subimage dynamically
    m = size(Image, 1)/4; % Number of rows per subimage.
    n = size(Image, 2)/4; % Number of columns per subimage.
    column = 1;row = 1;counter = 1; % Variable initializatio
    for i = 1:16 % Loop through every subimage
       if counter > 4
           column = 1;
           row = row + m;
           counter = 1;
       end
       subImgH(i).img = subim(HImage, m, n, row, column);  % Get subimage for H
       subImgV(i).img = subim(VImage, m, n, row, column);  % Get subimage for V
       subImgD(i).img = subim(DImage, m, n, row, column);  % Get subimage for D
       subImgA(i).img = subim(AImage, m, n, row, column);  % Get subimage for A 
       column = column + n;
       counter = counter + 1;
    end
    % Compute the blocks and histograms
    try
        for i = 1:16 % All subimages
            HLocal(i,1:5) = 0;
            row = 1; column = 1;
            for j = 1:ceil(size(subImgH(i).img,1)*size(subImgH(i).img,2)/4)  % All 2x2 blocks
                if column > size(subImgH(i).img,2)
                    column = 1;
                    row = row + 2;
                end
                % Determine the max edge of the averages in a 2x2 area
                [M, I] = max([sum(sum(abs(subImgH(i).img(row:row+1,column:column+1))))/4 ...
                    sum(sum(abs(subImgV(i).img(row:row+1,column:column+1))))/4 ...
                    sum(sum(abs(subImgD(i).img(row:row+1,column:column+1))))/4 ...
                    sum(sum(abs(subImgA(i).img(row:row+1,column:column+1))))/4 ...                
                    Threshold]);
                IndexedSub(i).img((row + 1)/2,(column + 1)/2) = I; 
                HLocal(i,I) = HLocal(i,I) + 1;
                column = column + 2;
            end
            HLocal(i,:) = HLocal(i,:)./(ceil(size(subImgH(i).img,1)*size(subImgH(i).img,2)/4));
        end
    catch
        error('Issue with subimages having odd rows and columns, block error.')
    end

It showing me, error('Issue with subimages having odd rows and columns, block error.') How to solve it? Any mistake? Or I need to remove odd rows and columns for subimages?

    function s = subim(f, m, n, rx, cy)
    % SUBIM Extracts a subimage, s, from a given image, f. 
    % The subimage is of size m-by-n, and the coordinates of its top, 
    % left corner are (rx, cy). 
    s = zeros(m, n); 
    rowhigh = rx + m - 1; 
    colhigh = cy + n - 1; 
    xcount = 0; 
    for r = rx:rowhigh 
        xcount = xcount + 1;
        ycount = 0; 
        for c = cy:colhigh 
            ycount = ycount + 1;
            s(xcount, ycount) = f(r, c);
        end
    end