How can I convert grayscale image to a binary image without using a toolbox function?

I want to create a binary image from a gray scale image, using a specific threshold value of 0.2, but without using im2bw(), which is in the Image Processing Toolbox. How do I do it?

Answer 1

[Image Analyst]

Just threshold:

binaryImage = grayImage > 0.2;  % or < or <= or >= (whatever you want).

No toolbox needed.

[maha D]

function y = myrgb2gray(x);
x = imread('rgb.TIF');

R=x;G=x;B=x;
y = 0.299*R + 0.59*G + 0.111*B;
y = rgb2gray(x);  %

This how i convert rgb to gray but how can i convert it to binary. i didn't get what you wrote.

[Image Analyst]

This will convert (the badly-named) y to binary:

y = y > 0.2;

Just try it and see.

[maha D]

when i run the program y=y>0.2; it's giving me wrong answer, it showed me the output of 0.1 not 0.2,also when i tried to change the threshold it's giving me wrong output... sorry for disturbing.

[Image Analyst]

If the original image has a range of uint8, meaning 0-255, then you need to use the gray level corresponding to that. In other words, your threshold needs to be 0.2 * 255 = 51 rather than 0.2

y = y > 51;

im2bw() uses normalized threshold levels, so if you're not using a normalized image you have to not use a normalized threshold value and use a value in the range of your original image, like in the range 0-255.

[maha D]

it worked thank you so much .

[Machbah Uddin]

very nice

[Alexander Motler]

very very nice

[pavani]

I converted my gray image to binary using following statement.

Binary = gray >= 51;

Now i want the gray image back. how to restore it? can any one please tell me.

Thank you in advance

[Walter Roberson]

uint8(Binary).*uint8(255)

[Image Analyst]

Don't use gray as the name of your variable. It is the name of a built-in function. What do you mean by "back"? What happened to it. You still have (the badly named) "gray" in your functions workspace, right? So just use it. You didn't say "clear gray" did you? So it should still be there.

[DGM]

To be clear, let's assume an empty workspace:

inpict = imread('cameraman.tif'); % a uint8-class grayscale image
mask = inpict >= 128; % a logical-class binary image

At this point there are two variables in the workspace: inpict and mask. You can't back-calculate inpict from mask, but you don't have to, since it's still in memory. Just use inpict.

Answer 2

[Ali nafaa]

x = imread('cameraman.tif');
figure,imshow(x);

[r,c] = size (x);
output=zeros(r,c);

for i = 1 : r
    for j = 1 : c
        if x(i,j) > 128
            output(i,j)=1;
        else
            output(i,j)=0;
        end
    end
end

figure,imshow(output);

[DGM]

There's no need for the loops. The appropriate answer was given years prior.

inpict = imread('cameraman.tif'); % a uint8-class grayscale image
mask = inpict >= 128; % a logical-class binary image

... or if inpict is not needed for anything, you can avoid the intermediate result:

mask = imread('cameraman.tif') >= 128; % a logical-class binary image

Display them however you choose.