Sum of RGB components

12 Ansichten (letzte 30 Tage)
Pedro Minsk
Pedro Minsk am 5 Nov. 2015
Kommentiert: Image Analyst am 6 Nov. 2015
I'm trying to sum the components of RGB of a point. Like it follows in the code. For example, if I have RGB = [230 100 150],
I want to get for RGB(1) + RGB(2) + RGB(3) the value of 230 + 100 + 150 = 480. The problem is that it returns 255. I know that 255 is the maximum value in the RGB system, but how could I fix the code so I can get the right sum?
b = squeeze(a(c, d, :)); % Value in RGB of a point of a image. 'C' and 'd' are the
% coordinates of the point
if ((b(1) + b(2) + b(3)) > 725)
K = 2.00;
elseif ((b(1) > b(2)) && (674.5 < (b(1) + b(2) + b(3))) && ((b(1) + b(2) + b(3)) <= 725))
K = 2.10;

Akzeptierte Antwort

Stephen23
Stephen23 am 5 Nov. 2015
Bearbeitet: Stephen23 am 5 Nov. 2015
Currently the numeric class is uint8, which only supports maximum value of 2^8-1 = 255. You need to convert the numeric class to single, double or uint16 (or some other integer).
What you are currently doing:
>> X = uint8([230 100 150]);
>> X(1)+X(2)+X(3)
ans = 255
Option one use sum, which converts internally to double:
>> sum(X)
ans = 480
Option two convert explicitly to uint16 or double:
>> Y = uint16(X);
>> Y(1)+Y(2)+Y(3)
ans = 480
  1 Kommentar
Pedro Minsk
Pedro Minsk am 5 Nov. 2015
It's worked! I used the option one.
Thank you so much.

Melden Sie sich an, um zu kommentieren.

Weitere Antworten (1)

Image Analyst
Image Analyst am 5 Nov. 2015
I get the idea that you are going to do this on all pixels in the image and then threshold. There are vectorized ways, like
% Sum R+G+B for all pixels in the image:
sumImage = sum(double(rgbImage), 3);
% Get a mask where the sum is more than 725.
brightPixelsMask = sumImage > 725
% and so on.
  3 Kommentare
Stephen23
Stephen23 am 6 Nov. 2015
Note that sum converts to double, so there is no need to call double:
sumImage = sum(rgbImage,3);
Image Analyst
Image Analyst am 6 Nov. 2015
You're right. I wonder why conv() or conv2() don't do that.

Melden Sie sich an, um zu kommentieren.

Community Treasure Hunt

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

Start Hunting!

Translated by