views:

289

answers:

4

I have a degree in computer science and I have taken the following math courses.

  • Calculus I
  • Calculus II
  • Discrete Mathematics and Number Theory
  • Linear Algebra
  • Probability
  • Logic
  • Automata Theory

What other courses should I take in order to prepare for studying wavelets, with a focus of implementing wavelet transforms?

EDIT:

Looks like this was closed for not being "programming related". That is wrong!

Wavelet transform is a very common image processing technique, it's used in H.264 and JPEG2000. Is image processing beyond the scope of StackOverflow?

+2  A: 

Linear algebra and calculus may help you there, but not much else. You'll also want to look at complex analysis and differential equations.

Chris Simmons
+7  A: 

On top of what you've got there already, I would recommend signal processing or some similar course that covers Fourier transforms and the like. Besides being useful as a foundation for wavelets, Fourier theory will give you a new way of looking at data that is often useful. Wavelets will probably be part of the curriculum for more advanced signal processing courses.

Martin B
+1 for Signal Processing. Understanding aliasing and other SP concepts help!
kigurai
+2  A: 

It sounds to me like you should just start learning about wavelet transforms and then figure out gaps along the way. They're not that involved. Fourier transforms etc are just an example of an orthogonal basis that is part of linear algebra.

Peter
A: 

Depends whether you want to learn about discrete or continuous wavelet transforms. If discrete then you'll need basic Fourier theory, linear algebra and complex number theory. If continuous then you'll need advanced Fourier theory and stationary phase approximations.

If you want to do research then I'd recommend learning both discrete and continuous. Most people only know one or the other in detail and it is seriously stifling research. There is a lot of opportunity for cross pollination here.

Jon Harrop