How to Create function to invert Image colors

300 views (last 30 days)
Krish Desai
Krish Desai on 8 Oct 2015
Edited: DGM on 26 Nov 2022
I need to write a function that inverts the colors (negative effect) of an image. I know that I need subtract each RGB color value from 255, but I do not know how to write function to do this. The following is my best guess:
% Invert image
s = input('What is the value of the threshhold (from 0 to 255) ');
Inverted = invert(current_img); % create your own function for inverting
%for current_img
current_img(:,:,1)= current_img(:,:,1)-s;
current_img(:,:,2)= current_img(:,:,2)-s;
current_img(:,:,3)= current_img(:,:,3)-s;
  1 Comment
Adam
Adam on 9 Oct 2015
If you are just inverting the colours where does 's' fit in? Your code doesn't do any subtraction from 255 despite you saying that is what you need.

Sign in to comment.

Answers (3)

Deepanshu Mehta
Deepanshu Mehta on 16 Aug 2019
Or you can simply use
img=imread('image.png');
y=255-Img;
imshow( y );
Thanks me later...
  2 Comments
Aaditya Aaditya
Aaditya Aaditya on 28 Feb 2022
ok, but this is image negation, not inversion, unless image is black & white.
Rik
Rik on 28 Feb 2022
@Aaditya Aaditya Can you explain what should happen instead? I don't see how this wouldn't be an inversion. Looks like one to me.
im=imread('peppers.png');
subplot(1,2,1)
imshow(im)
subplot(1,2,2)
imshow(255-im)

Sign in to comment.


Robert Dylans
Robert Dylans on 9 Oct 2015
A simple invert of image 'x' into image 'y':
x=imread('image.jpg');
y(:,:,:)=255-x(:,:,:);

DGM
DGM on 25 Oct 2022
Edited: DGM on 26 Nov 2022
If you can be certain of the class of incoming images, you simply subtract the image from the nominal white value associated with that class. That said, assuming that all images are always uint8 is a great way to cause problems. Unless you know that the incoming images are always uint8, then expect that the nominal white value is not always 255.
As @Robert Dylans points out, Image Processing Toolbox (IPT) has a class-agnostic tool that does this quite safely. If your images are correctly-scaled for their class, you can just use imcomplement().
If you don't have IPT, you can use the attached file from MIMT. It should not be dependent on MIMT or IPT.
% some images of varied class
Auint8 = imread('peppers.png');
Aint16 = im2int16(Auint8);
Adouble = im2double(Auint8);
% invert them all
Buint8 = iminv(Auint8);
Bint16 = iminv(Aint16);
Bdouble = iminv(Adouble);
% create a montage
montage({Auint8 Buint8; Aint16 Bint16; Adouble Bdouble})
Open that file and see how it does the inversion for all the different image classes. It's a bit more succinct than what's been described.

Categories

Find more on Programming in Help Center and File Exchange

Community Treasure Hunt

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

Start Hunting!