Matlab image spatial resolution, change pixel

Hi, how can ı change 8 bit image's pixel numbers like these? Help please

Answer 1

[Image Analyst]

Try this. It will make sure the output image is exactly the same size as the input image.

inputImage = imread('peppers.png');
[inputRows, inputColumns, numColors] = size(inputImage)

inputRows = 384

inputColumns = 512

numColors = 3

for ii = 1:6
    subplot(2,3,ii);
    outputImage = imresize(inputImage,2^-(ii-1));
    [rows, columns, numColors] = size(outputImage);
    fprintf('After resizing once it is %d rows by %d columns\n', rows, columns);
    % Resize output image again to match the input.
    outputImage = imresize(outputImage, [inputRows, inputColumns], 'nearest');
    % Update size.
    [rows, columns, numColors] = size(outputImage);
    fprintf('    After resizing twice it is again %d rows by %d columns\n', rows, columns);
    imshow(outputImage);
    % Show title with new size.
    caption = sprintf('%d rows by %d columns', rows, columns);
    axis('on', 'image')
    title(caption)
end

After resizing once it is 384 rows by 512 columns

    After resizing twice it is again 384 rows by 512 columns

After resizing once it is 192 rows by 256 columns

    After resizing twice it is again 384 rows by 512 columns

After resizing once it is 96 rows by 128 columns

    After resizing twice it is again 384 rows by 512 columns

After resizing once it is 48 rows by 64 columns

    After resizing twice it is again 384 rows by 512 columns

After resizing once it is 24 rows by 32 columns

    After resizing twice it is again 384 rows by 512 columns

After resizing once it is 12 rows by 16 columns

    After resizing twice it is again 384 rows by 512 columns

Answer 2

[Voss]

You can try imresize()

im = imread('image_1.png');
for ii = 1:6
    subplot(2,3,ii);
    imshow(imresize(im,2^-(ii-1)));
end

[akevg akevg]

thank you so much . it worked for me

[Voss]

Since it worked, do you mind accepting the answer? I appreciate it!

[Image Analyst]

@akevg akevg Note that the output image size is different than the original. With my code below, the image size is the same, it's just blockier looking. Do you want the output image size to be the same or smaller?

[akevg akevg]

@Image Analyst ı need same size

[akevg akevg]

@_ ı accepted now. thanks

[Image Analyst]

Then you accepted the wrong answer. Watch this:

im = imread('peppers.png');
for ii = 1:6
    subplot(2,3,ii);
    outputImage = imresize(im,2^-(ii-1));
    imshow(outputImage);
    axis('on', 'image')
    [rows, columns, numColors] = size(outputImage)
    caption = sprintf('%d rows by %d columns', rows, columns);
    title(caption)
end

rows = 384

columns = 512

numColors = 3

rows = 192

columns = 256

numColors = 3

rows = 96

columns = 128

numColors = 3

rows = 48

columns = 64

numColors = 3

rows = 24

columns = 32

numColors = 3

rows = 12

columns = 16

numColors = 3

[Voss]

I appreciate that you accepted my answer. However, if you want the same image size, go with @Image Analyst's answer and accept it instead.

[akevg akevg]

@Image Analyst ı fixed thanks

[DGM]

Or you could just do

inpict = imread('peppers.png');

k = 16;
outpict = imresize(imresize(inpict,1/k,'bilinear'),k,'nearest');

imshow(outpict)

[size(inpict); size(outpict)]

ans = 2×3

384 512 3 384 512 3

Which is far simpler and much faster than using blockproc() for any moderately large image.

Of course, it depends how much control one wants over exactly which pixels are contributing to each block and how exactly they're being weighted. I'm assuming that the goal here has no technical requirements, as none were given.

Answer 3

[Image Analyst]

Use blockproc(). Solution attached.

[akevg akevg]

thank you so much