Hi, I want a 2d array of rows and columns filled with floating point numbers (matrix) which can be converted to a grayscale image.

Should i use the mat2gray function of matlab ? I want to program my own image and fill it with numbers.

[jgg]

Yeah, that will totally work.

 A = repmat(0:0.01:1,10,1)
 i = mat2gray(A)
 imshow(i)

Creates a new nice grayscale gradient. Each element of the matrix is a pixel with colour intensity which is converted to between 0 and 1. So, you could draw a nice image by just fiddling with the numbers if you're good at pixel drawing.

[Anvinder Singh]

Thanks Jegg, So A = repmat(0:0.01:1,10,1) plots values from 0 to 1 with a gap of 0.01. I could not understand what is purpose of 10,1 in the end ? How Is that the number of rows and columns ?

[Walter Roberson]

repmat(A,B,C) makes B vertical copies and C horizontal copies. When A is a row vector that means you would end up with B rows.

[Anvinder Singh]

Thanks Walter. So is it possible to use multiple loops in the same matrix to have different pixels ? I mean instead of one repmat i need multiple repmats for the same matrix A.

Answer 1

[Cam Salzberger]

Hello Anvinder,

I understand that you are looking to display a grayscale image based on values from a matrix. As jgg said, "mat2gray" helps in ensuring that the matrix values get scaled appropriately (between 0 and 1) to display the full range in an image.

You don't need to use "repmat" to create the grayscale image, that was just an example. Here's an example that uses loops to allow for calculating each pixel value:

nRows = 10;
nCols = 15;
A = zeros(nRows,nCols);
for r = 1:nRows
    for c = 1:nCols
        A(r,c) = r+c/2;
    end
end
I = mat2gray(A);
imshow(I)

I hope this helps!

-Cam