Umm, pardon my current ignorance, but this looks like a cut down version of ATWT in many ways?
I'm sure Juan can explain it in much better terms, and maybe even correct what I'm about to say, but just because the interface looks similar doesn't mean it's a toned down version of ATWT, rather a different way to do multiscale decomposition and so on. Instead of using wavelet transforms, it uses a median transform, which offers some advantages (and disadvantages) over WTs, in particular the absence of certain artifacts that are unavoidable when doing WTs, "better" smoothing, etc.
Thank Rogelio. I've just been trying to digest some of Juan's WWW reference. I
think I'm starting to get it - as you say, just because interfaces very similar (which is great, really) the algortithm behind it is much more robust for the use of medians. Will await tutuoral info with interest. For now, will just have to play and compare....
