Trying to invert an expression

  • Context: Graduate 
  • Thread starter Thread starter Morberticus
  • Start date Start date
  • Tags Tags
    Expression
Click For Summary

Discussion Overview

The discussion revolves around inverting an expression involving a double integral of a function f(x,y) multiplied by a cosine term. Participants explore the nature of the transform and the challenges associated with inverting it to retrieve f(x,y).

Discussion Character

  • Exploratory, Technical explanation, Debate/contested

Main Points Raised

  • One participant presents the expression n(k) = ∫cos(x-y)f(x,y)dxdy and seeks to invert it to find f(x,y).
  • Another participant questions the absence of the variable k in the original integral and asks for the domain of integration.
  • The first participant acknowledges the oversight regarding k and corrects the expression to n(k) = ∫ cos(k(x-y)) f(x,y) dxdy.
  • A different participant suggests that the transform resembles a "cosine transform" and provides a link for further reference, while also mentioning the potential use of the inverse Fourier transform.

Areas of Agreement / Disagreement

Participants do not reach a consensus on the best approach to invert the expression, and multiple viewpoints regarding the nature of the transform and methods for inversion are presented.

Contextual Notes

The discussion lacks clarity on the domain of integration and the specific conditions under which the transforms are applied, which may affect the inversion process.

Morberticus
Messages
82
Reaction score
0
I've come across an expression that looks like

n(k) = ∫cos(x-y)f(x,y)dxdy

Is there a name for this transform? I would like to invert it to obtain f(x,y) but I'm not used to the 2D integral on the RHS. I tried to turn it into a Fourier transform:

n(k) = 1/2 ( ∫eixe-iyf(x,y)dxdy + ∫e-ixeiyf(x,y)dxdy)

but got stuck. Any help would be appreciated.

Thanks
 
Physics news on Phys.org
Hi, there is something of strange, where is ##k## in the integral?, you say ##n(k)## but in the right side I don't see ##k##, and what is the domain of integration ?
 
You are right! Sorry, I forgot about the k when writing down the expression(s). The correct expression is

n(k) = ∫ cos( k(x-y) ) f(x,y) dxdy

Thanks
 
I think your transform is analogous to the ''cosine transform'', I hope you can find something useful here

http://dsp-book.narod.ru/TAH/ch03.pdf

but I think (as you wrote...) it is possible to use the inverse Fourier transform...
 

Similar threads

  • · Replies 4 ·
Replies
4
Views
3K
  • · Replies 4 ·
Replies
4
Views
3K
  • · Replies 0 ·
Replies
0
Views
4K
  • · Replies 3 ·
Replies
3
Views
3K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 3 ·
Replies
3
Views
3K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 2 ·
Replies
2
Views
3K
  • · Replies 4 ·
Replies
4
Views
3K