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 simulating the evolution of a liquid film through the solution of a 4th order nonlinear partial differential equations.

Of late, I began experimenting with DFT of the result that I have.

My initial film profile looks like this:

Initial film profile at time=0

And it's DFT looks like this:

DFT of initial film profile

A close-up of the DFT reveals that I have a DC value and there are 2 wiggles in X and 2 wiggles in the Y direction and that these are the major frequencies:

close-up of initial film profile's DFT

However, plotting this as a line plot shows:

Line plot - initial film profile's DFT

What is the spike at the corner? I guess it is the matrix plot from before but why the spike?

Now, at a later stage, my film rupture and looks such as this:

Ruptured film

And the DFT and the line plots look like:

DFT-rupture Line plot of DFT at rupture

So does this DFT show that "other" frequencies are now involved? And What is with the second spike in the right corner of my line plot?

share|cite|improve this question
Firs you cannot identify a 2D DFT with only one specific part of it. The greatest peak at the left corner corresponds to the DC component. Second the DFT is symmetric meaning that you will see the mirrored version of the signal with respect to $75$ according to your scale. – Seyhmus Güngören Aug 29 '12 at 19:55
@SeyhmusGüngören Sure. That I do understand, thank you though! :) So does it mean that most of the frequencies are located at the corners? There is some internal symmetry as well to the profile plots... – drN Aug 29 '12 at 20:00
up vote 1 down vote accepted

independent of the symmetry in your data you will have symmetry in DFT domain. Yes it means your data has mostly low frequencies but once again. You can plot it in many different grids of 2D DFT and can catch some different frequencies in $1D$. Meanwhile your $2D$ plot doest look like so nice. There should be another way to plot it in a better way. Just create a random matrix using the command randn $(500,500)$ and compare its DFT with your data. I think your data is smooth and doesnt have various frequencies as it seems also periodic as well.

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.