Is it possible to subsample an image by deleting data then interpolating to create a blurred image, while keeping the resulting image the same size as the original image?

Is it possible to subsample an image by deleting data then interpolating to create a blurred image, while keeping the resulting image the same size as the original image?

The code I am using below does the subsampling and interpolation but changes the image size. It looks smaller. See a part of my code below.

originalImage = imread(imagePath);
zeroPaddedImage = zeros(size(originalImage), 'like', originalImage);
zeroPaddedImage(:, 1:subsamplingFactor:end) = originalImage(:, 1:subsamplingFactor:end);
blurredImage = imresize(zeroPaddedImage, 1/subsamplingFactor, 'bicubic');

Answer 1

[Matt J]

siz=size(originalImage);

tempImage=originalImage( 1:subsamplingFactor:end,  1:subsamplingFactor:end, :);

blurredImage = imresize(tempImage, siz(1:2), 'bicubic');

[Tevin]

I did the code below and got the error following in italic.

zeroPaddedImage = zeros(size(originalImage), 'like', originalImage);

siz=size(originalImage);

originalImage=originalImage( 1:subsamplingFactor:end, 1:subsamplingFactor:end);

blurredImage = imresize(zeroPaddedImage, siz, 'bicubic');

When the second argument is a 1-by-3 vector, IMRESIZE interprets it

as an M-by-3 colormap. If you intend to specify an output size, use

the syntax: imresize(A,___,'OutputSize',SZ) where SZ is a 1-by-2

vector.

Error in matlab.images.internal.resize.resizeParseInputs (line 28)

parsePreMethodArgs(varargin, method_arg_idx, first_param_string_idx);

Error in imresize (line 153)

params = matlab.images.internal.resize.resizeParseInputs(args{:});

Error in Subsample_interpolate (line 33)

blurredImage = imresize(zeroPaddedImage, siz, 'bicubic');

[Matt J]

Did you try what the error message recommends?

[Catalytic]

When the second argument is a 1-by-3 vector

Why is siz a 1-by-3 vector? Is this an RGB image? In your original post, you are indexing it like a 2D image-

zeroPaddedImage(:, 1:subsamplingFactor:end) = originalImage(:, 1:subsamplingFactor:end);

[Tevin]

It is a coloured image. I tried to use the error message to correct my code but it did not work. I do not understand exactly what I am supposed to do.

I want to know if I have an image with missing data is one direction ( I subsampled and replaced those values with zeros). I wanted to interpolate to refill the missing data and smoothen the image. Is that possible?

[Matt J]

siz=size(originalImage);

originalImage=originalImage( 1:subsamplingFactor:end,  1:subsamplingFactor:end, :);

blurredImage = imresize(zeroPaddedImage, siz(1:2), 'bicubic');

[Tevin]

Thank you. When i do this, the blurredImage is black.

[Matt J]

You should attach the image in a .mat file, so that it can be examined directly.

[Image Analyst]

What does this show?

figure
imshow(blurredImage, []); % Use [] to scale image.
maxGL = max(blurredImage(:))
minGL = min(blurredImage(:))

[Tevin]

%Please see a part of my code

% Create a zero-filled image of the same size as the original
siz = size(originalImage);
zeroPaddedImage = zeros(size(originalImage), 'like', originalImage);
zeroPaddedImage(:, 1:factor:end, :) = originalImage(:, 1:factor:end, :);

blurredImage = imresize(zeroPaddedImage, siz(1:2), 'bicubic');

I have attached a file with the original image, the subsampled and the interpolated image. The interpolated image looks like the subsampled one. I want to fill the gaps by interpolating so it seemed like a blurry or noisy image, rather than a having gaps. Can you look at where my code is going wrong?

[Matt J]

originalImage=load('Images').originalImage;
subsamplingFactor=20;

siz=size(originalImage);

tempImage=originalImage( 1:subsamplingFactor:end,  1:subsamplingFactor:end, :);

blurredImage = imresize(tempImage, siz(1:2), 'bicubic');

montage({originalImage,blurredImage})