Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I am trying to implement a computer vision algorithm, but I'm having a problem with some notation used in a article. They define an image as a set of RGB colors with an index $\mathbb{z} = (z_1, z_2, ..., z_n)$.

They then use this formula: $\langle \| z_p - z_q \|^2 \rangle$, with $\langle \cdot \rangle$ defined as the expectation operator over the whole image.

My question is, how can a color be passed to an expectation operator? As far as I understand it, an expectation operator needs a random variable. Somehow, I think I'm interpreting this wrong. Can anyone shed some light on the meaning of the $\langle \cdot \rangle$ operator?

share|cite|improve this question

It is possible that they mean empirical expectation, ie they don't really know the exact distribution of the pixels, but they define for instance $p_i=\frac{\#(z_j==i)}{N}$, where $N$ is the total number of pixels and $j\in[1,N]$, which yields the relative number of pixels valued $i$ in the entire image. Now you can define the empirical expectation as $\hat{E}[f(x)] = \sum_i p_i f(x)$ .

I do not know if this is the case here, but it occurs sometimes.

share|cite|improve this answer
up vote 0 down vote accepted

I have found the answer:

It seems like the meaning of $\langle \| z_p - z_q \|^2 \rangle$ is the color variance measured of the whole image. I dont see their logic, but it seems they meant it like that.

share|cite|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.