These are simply statistical measures for the pixel distribution within the image. Personally, I have not found a use for them. I prefer to use the histogram as a visual aid to help me adjust levels in RGB and their respective separate channels while I look at the image to see the results. I prefer not to 'edit by the numbers', relying instead on my aesthetic sense of how the image should look.
I agree with Preston. I have not seen these metrics used in imaging applications. I can imagine they can be used to judge some overall characteristics of an image (high/low-key, tonal range) but I can't think of a practical use for them. I know some over-zealous programmers like to create information overload to make their software look impressive but in many cases it is meaningless in practice.
Guy
Posted by David Bostedo on December 19, 2007 at 11:53 AM
I would guess that they might be useful for graphic designers, who can pick or choose colors and brightnesses at will. Even then, though, I don't know how useful it would be. I've written a few image processing algorithms for classes, and even within the algorithms, I don't think these numbers are very meaningful. It's possible to use them to identify, for instance, high key vs. low key and apply different algorithms based on that.
But even then, displaying them to the user doesn't really serve a purpose. I think Guy nailed it in the fact that it may just look impressive to some people.
Discussion
3 repliesThanks,
Murali
These are simply statistical measures for the pixel distribution within the image. Personally, I have not found a use for them. I prefer to use the histogram as a visual aid to help me adjust levels in RGB and their respective separate channels while I look at the image to see the results. I prefer not to 'edit by the numbers', relying instead on my aesthetic sense of how the image should look.
-PB
Guy
But even then, displaying them to the user doesn't really serve a purpose. I think Guy nailed it in the fact that it may just look impressive to some people.