Finding the difference between two images algorithm.

Hello,I am trying to write code to find the difference between two similar images. My way was: make them gray (with rgb2gray) subtract and then add threshold) but it gave me nothing.
The way I am working isnt correct ? maybe I am missing something ?

3 Comments

What do you intend to do exactly? do you want to calculate the difference between two pixels with the same coordinates?
There is a lot that can go wrong here. We can guess what went wrong, but the best way is if you upload example images and the code you were using.
I want to make algorithm that can slove the game "guess the difference". You upload 2 similar photos and get white dots in the places when there is any difference.

Sign in to comment.

Answers (2)

Subtracting will clip negative values at zero - not good! Cast to double before that.
Try the built-in functions immse() or psnr().
If you want to do it yourself, check out imabsdiff().

2 Comments

I forgot to mention that I subtracted with abs(and now I tried with double).
Like this : abs(double(im1gray)-double(im2gray));
But it still giving me in white places in the pictures that are same. The built-in functions immse() or psnr() giving me error like that treshold so still I dont got the final picture I want(black background with white when there is any difference).
Are you sure you're displaying the difference image with [] - otherwise it will do that.
differenceImage = abs(double(im1gray)-double(im2gray));
imshow(differenceImage, []);

Sign in to comment.

please anyone help me ...which algorithm is best for fake currency detection

2 Comments

The best algorithm is probably classified, as it can be used to avoid counterfeit detection.
Also, by posting an answer you decreased your chances of getting effective help.
Have a read here and here. It will greatly improve your chances of getting an answer.

Sign in to comment.

Asked:

on 28 Jul 2018

Commented:

on 21 Apr 2021

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!