What is the possible solution to this problem converting to a gray scale image?

I am facing this error while running the code for background image subtraction.

>> RemoveBackground('a.jpg','ab.jpg')
??? Error using ==> rgb2gray>parse_inputs at 82
MAP must be a m x 3 array.

Error in ==> rgb2gray at 35
X = parse_inputs(varargin{:});

Error in ==> RemoveBackground at 5
I1=rgb2gray(I);

The complete code is as follows:

function [im]=RemoveBackground(I,Ib)
% Ib=background image
% I=input image
% im=background removed image
I1=rgb2gray(I);
Ib1=rgb2gray(Ib);
Id=imsubtract(Ib1,I1);
Id=Id;
Id=bwareaopen(Id,500,8);
Id=imfill(Id,'holes');
Id=uint8(Id);
[r c]=size(Id);
for i=1:r
    for j=1:c
        if Id(i,j)==255
            Id(i,j)=1;
        end
    end
end
if size(I, 3) > 1
  I = rgb2gray(I);
end
im(:,:,1)=I(:,:,1).*Id;
im(:,:,2)=I(:,:,2).*Id;
im(:,:,3)=I(:,:,3).*Id;

figure,imshow(im)

Answer 1

[Walter Roberson]

No, the problem is that you are passing the name of an image rather than the content of the image. You are processing strings, not images.

Answer 2

[Image Analyst]

You forgot to attach your images. Why are you trying background subtraction? It's usually for modalities like radiology and fluorescence, and not for things like digital RGB cameras. What are you using?

Your problem is you're trying to call it on a gray scale image instead of a color image.

Try this:

[rows, columns, numberOfColorChannels] = size(I);
if numberOfColorChannels == 3
    % If it's color, convert it to gray level.
    I1 = rgb2gray(I);
else
    I1 = I; % Already grayscale.
end

[rows, columns, numberOfColorChannels] = size(Ib);
if numberOfColorChannels == 3
    % If it's color, convert it to gray level.
    Ib1 = rgb2gray(Ib);
else
    Ib1 = Ib; % Already grayscale.
end

[Innovative]

Further I am taking image from Laptop webcam!

[Innovative]

After adding the code, now it is giving me this error

[Image Analyst]

Well what do you think might be the problem? It's saying that the x and y size of your images don't match. Did you crop one, or get one from a different camera or other source?

[rows, columns, numberOfColorChannels] = size(I);
[rowsb, columnsb, numberOfColorChannelsb] = size(Ib);
if rowsb ~= row
    errorMessage = sprintf('Number of rows does not match.\nImage had %d rows, while Imageb has %d rows.', rows, rowsb);
    uiwait(warndlg(errorMessage));
    return;
end
if columnsb ~= columns
    errorMessage = sprintf('Number of columns does not match.\nImage had %d columns, while Imageb has %d columns.', columns, columnsb);
    uiwait(warndlg(errorMessage));
    return;
end
% Both rows and columns match if you get here.
Id = imsubtract(I - Ib);

[Innovative]

I just captured from laptop webcam. I am working on it. Let's c

[Image Analyst]

If they both were captured by the same webcam, then they should be the same size. You probably did something to one of them to change its size. I suggest you learn how to use the debugger to step through your code and figure out when/where/why the snapped image changes size.

[Walter Roberson]

You need to pass the content of the images, not the name of the images!!

[Image Analyst]

Like this:

a = imread('a.jpg')
b = imread('ab.jpg');
RemoveBackground(a,ab);

[Innovative]

Thank You Walter Roberson for the suggestion and Thank you Image Analyst for the syntax. Both suggestion and syntax worked and My code got debugged!!

The basic entity where I were going dumb were these commands. im(:,:,2)=I(:,:,2).*Id; im(:,:,3)=I(:,:,3).*Id;

I just commented them , ran the code with this : im(:,:,1)=I(:,:,1).*Id; and followed the syntax mentioned by Image Analyst and My problem got solved.

Answer 3

[Innovative]

Thenk You very much Image Analyst and Walter Roberson for Thy Timely HELP !!