Blue Channel Not Separating from Image

I wrote this code to display only the blue channel of an image and both red and green channels work but the blue channel is unable to produce a result.

image_variable3 = imread('CH0.tif');
blue_channel = image_variable3; 
blue_channel(:,:,1)=0;
blue_channel(:,:,2)=0;
imshow(blue_channel);

I would get this error message and I don't know what any of it means.

Error using images.internal.imageDisplayValidateParams>validateCData

Multi-plane image inputs must be RGB images of size MxNx3.

Error in images.internal.imageDisplayValidateParams (line 30)

common_args.CData = validateCData(common_args.CData,image_type);

Error in images.internal.imageDisplayParseInputs (line 79)

common_args = images.internal.imageDisplayValidateParams(common_args);

Error in imshow (line 253)

images.internal.imageDisplayParseInputs({'Parent','Border','Reduce'},preparsed_varargin{:});

Error in Merge (line 46)

imshow(blue_channel);

I would greatly appreciate any help. Thank!

[Stephen23]

"Multi-plane image inputs must be RGB images of size MxNx3."

Please show us the output of this command:

size(image_variable3)

[Kenji]

The output for the command size(image_variable3) is

ans =

11298 12846

[Matt J]

@Kenji So isn't your question resolved? Clearly it is not MxNx3.

[Kenji]

I see. So by my understandings, in an MxNx3, the 3 indicates a length of 3 layers of red, green, and blue respectively. So since the image is not MxNx3, the image does not contain any blue altogether? A pure blue square image I found is not MxNx3 either and the blue channel can't be separated? Is there a way to convert a MxN to a MxNx3?

[Stephen23]

" So by my understandings, in an MxNx3, the 3 indicates a length of 3 layers of red, green, and blue respectively."

Simply knowing the size of the array is not sufficient information to determine the colorspace. Although sRGB might be the most common colorspace, it could be other kinds of RGB (e.g. Adobe RGB), or CIELAB (i.e. Lab), or any of various other 3-channel colorspaces. See for example:

https://imagemagick.org/script/formats.php#supported

"So since the image is not MxNx3, the image does not contain any blue altogether?"

If it is an indexed image then you need to check its colormap to know if it contains any blue.

"Is there a way to convert a MxN to a MxNx3?"

Of course there are ways to convert to sRGB Truecolor, but you have to know the image format and colorspace first:

https://www.mathworks.com/help/matlab/ref/imfinfo.html

[Image Analyst]

@Kenji did you even see any of the answers below? Scroll down.

Answer 1

[DGM]

If the image contains color content, but is returned as a MxN array when using imread() as shown, then chances are the image is indexed color.

% if it's an indexed color image
% you'll need to read the map with it
[inpict map] = imread('canoe.tif');

% if you want to treat it as RGB, you'll have to convert it
inpict = ind2rgb(inpict,map);

% you can then do whatever manipulations you want in RGB
Bonly = inpict;
Bonly(:,:,1:2) = 0;

imshow(Bonly,'border','tight')

If it's something else, then you can attach an example image. Since it's a TIF, you'll have to zip it in order to upload it.

[Stephen23]

You can find a description of indexed images here:

https://www.mathworks.com/help/matlab/creating_plots/image-types.html

Answer 2

[Image Analyst]

Your image is a gray scale image; it is NOT a RGB true color image. There is no blue channel.

You can make your code more robust by doing this:

storedImage = imread('CH0.tif');
size(storedImage) % Echo dimensions to command window.
if ndims(storedImage) == 3
    % It's an RGB image.  Get the blue channel and store in grayImage.
    grayImage = storedImage(:, :, 3);
    fprintf('Extracting blue channel of RGB image into variable "grayImage".\n');
elseif ndims(storedImage) == 2
    % It's a gray scale image.
    fprintf('Stored image is already monochrome (gray scale).\n');
    grayImage = storedImage(:, :, 3);
elseif ndims(storedImage) >= 4
    % It's a multipage gray scale image.
    fprintf('This is a multi-page TIFF image.  Decide what to do.\n');
end
imshow(grayImage, []);