How can I use a grayscale image as a input in this piece of code?

function RedCellCount(dataImg)
  data = imread(dataImg.Key{1});
  diff_im = imsubtract(data(:,:,1), rgb2gray(data));
  diff_im = medfilt2(diff_im, [3 3]);
  diff_im = imbinarize(diff_im,0.18);
  diff_im = bwareaopen(diff_im,300);

    bw = bwlabel(diff_im, 8);

    stats = regionprops(bw, 'BoundingBox', 'Centroid');

    fileName = dataImg.Key{1};
    display(fileName)
    display(length(stats));

I'm trying to load grayscale images in this piece of code from a folder whose path is in another .m file. This piece of code works perfectly fine for RGB images. but when i'm trying to use grayscale images its popping up with these errors:

   >> RedCellCount
      Not enough input arguments.

      Error in RedCellCount (line 2)
      data = imread(dataImg.Key{1});

and while using mapreduce() for grayscale images in another function, I got this error.

    >> MAP must be a m x 3 array.

I've tried everything to make it run but the program is not accepting grayscale images as it is 1-D whereas RGB images are 3-D matrices.

I'm not able to understand why imread() is working for RGB images but not for grayscale images.

    data = imread(dataImg.Key{1});

Any help will be highly appreciated. Thanks in advance.

[Geoff Hayes]

cybertetrahedron - what is the error that you are observing? From your post, it looks like you are calling RedCellCount as

 >> RedCellCount

and not providing any input arguments and so the error message makes sense.

How are you trying to run this function with the grayscale image? The input to RedCellCount assumes that the filename is obtained from

 dataImg.Key{1}

[FS]

Mr. Hayes- Thanks for making a comment. I've already mentioned that input is fetched from another matlab file which calls this function

    function RedCellCount(dataImg)

As I already clearly stated that this function is processing the conversion of image into binary by using imbinarize() and filtering it by medfilt2() successfully but it just works with an RGB image. Kindly rewrite the function(if possible) and post it using any other alterative and your own logic. If possible kindly drop your mail so that I send you the entire code. You can connect with me at [e-mail]

[Geoff Hayes]

FS - how about you show the full error message when your other function tries to pass the grayscale image into RedCellCount? If it is already an image (like you say) then you cannot pass it to RedCellCount since it is assuming that you are passing in a struct that has a filename. You will need to adapt the code to the new (grayscale image) input parameter.

[Image Analyst]

Before it crashes, put this on it's own line without a semicolon:

dataImg.Key{1}

What do you see in the command window? A string that contains a valid filename?

Answer 1

[Julie Kraus]

You can make the grayscale image into an RGB image by using

 NewImg=repmat(GreyImg,[1,1,3])

If you get an error due to the fact that it is using decimal points as opposed to intigers 1:256, before you run repmat

 GreyImgNew=round(256*GreyImg)

[FS]

Thanks for the suggestion. But how am I going to use it when

    data = imread(dataImg.Key{1});

is not able to read grayscale images?. Can you please implement this function using your own logic for grayscale images?