Not getting channel 1 in an Face images even though in Grey-colour

1 次查看（过去 30 天）

显示更早的评论

Kartikeya 2023-10-18

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/2035204-not-getting-channel-1-in-an-face-images-even-though-in-grey-colour

编辑： Harsha Vardhan 2023-10-18

采纳的回答： Harsha Vardhan

greyface2.jpg

在 MATLAB Online 中打开

originalImage = imread('greyface2.jpg');
% To get a 2-D grayscale image from the file, converting from RGB
% to gray scale if it's not already grayscale.
% Get the dimensions of the image.
% numberOfColorChannels should be = 1 for a gray scale image, and 3 for an RGB color image.
[rows, columns, numberOfColorChannels] = size(originalImage)
rows = 264
columns = 191
numberOfColorChannels = 3
if numberOfColorChannels > 1
    % It's already RGB.  Just put it into our rgbImage variable.
    rgbImage = originalImage;
else
    % It's not really RGB like we expected - it's grayscale (or indexed).  We'll assume it's grayscale here.
    fprintf('It is not really RGB like we expected - it is grayscale.\n');
    % Concatenate the gray scale image to be all 3 color channels.
    rgbImage = cat(3, originalImage, originalImage, originalImage);
end

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

采纳的回答

Harsha Vardhan 2023-10-18

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/2035204-not-getting-channel-1-in-an-face-images-even-though-in-grey-colour#answer_1335794

编辑：Harsha Vardhan 2023-10-18

在 MATLAB Online 中打开

Hi,

I understand that you want to check if an image is a gray scale image.

In many cases, gray colored images are stored as RGB images. When the R, G and B values of a pixel are either equal or very close to each other, that pixel gets the appearance of grayscale. In such cases, it must not inferred that the image is a gray scale image. Such an image can still contain 3 channels of R, G and B.

When you load such an image in MATLAB using imread, it will still read it as a 3-channel RGB image, because that's how the image is stored. That's why you're seeing numberOfColorChannels as 3.

To understand further, please execute the below code and observe the output.

originalImage(1,1,:)
  1×1×3 uint8 array
ans(:,:,1) =
   197
ans(:,:,2) =
   198
ans(:,:,3) =
   200

It can be seen from the above output that the first pixel - originalImage(1,1,:) has 3 channels associated with it. And the values of each channel are very close to each other. This gives an impression of greyscale inspite of having 3 channels.

If you want to ensure you're working with a true single-channel grayscale image in MATLAB, you can convert it using:

grayImage = im2gray(originalImage);

This answer has more information about the values returned by the 'size' function in the context of images - https://www.mathworks.com/matlabcentral/answers/81089-finding-the-dimensions-of-an-image#answer_939490

Hope this helps in resolving your query!

9 个评论
显示 7更早的评论隐藏 7更早的评论

Harsha Vardhan 2023-10-18

编辑：Harsha Vardhan 2023-10-18

@Kartikeya

In the screenshot you attached, the colunm 'size' displays the size of the matrix. The 'grayImage'has 365x265 in the 'size' column. It means it can be identified as a 2-dimensional matrix since it has only rows X columns.

Kartikeya 2023-10-18