I think I was a little harsh in this answer. Thoughts?

@rcollyer sounds fine to me

@Szabolcs (hope this gets to you) or @rcollyer: Do you know of a MMA implementation of the "total variation regularization" links you have in this question: stackoverflow.com/questions/6502440/… for discrete data?

@tkott You know, I never looked closely at that.

@tkott I'm sorry, it was only Googling. While I find it interesting, at that time I simply didn't have time to study it.

@rcollyer, @Szabolcs thanks! I just found a matlab implementation, so I'll just translate that, thought I already have a gaussian filter implementation that is working reasonably well :)

@Szabolcs and @Fx thanks. Sometimes I get tired of answering that question, and it surfaces every couple of weeks it seems. But, I'm relieved others don't think I was being to harsh.

Average number of questions/day now dropped slightly below 10. Time for some publicity action again I'd say. Fire up your blogs, twitter, FB, LinkedIn, newsgroup etc.!

@tkott if you're looking for Gaussian filters and scale space this site of a former colleague of mine and MMA enthusiast Bart ter Haar-Romenij could be useful. Lots of courseware, books and mma code sample

S

@SjoerdCdeVries Thanks, that actually looks like a great resource for MMA + Image analysis. I actually am applying it to 2D data (i.e., {{x1,y1}..}) that I need to smoothly differentiate. I implemented the method from the link above, and it works, but slowly. On monday I'll take a stab at improving it and then probably post as a question :)

@rcollyer It was acl, not me. But I too think it's fine.

@tkott Did you see the answer I gave to the question you linked to? It's about smooth differentiation of 1D data but it should be straightforward to adapt to your needs.

Yup, I tried that first, but it wasn't quite smooth enough to be useful. I had some time to implement that total variation regularization, and play around with SparseArrays in MMA for the first time, so I just translated it from a MATLAB file.

@Szabolcs :) did not notice.

Hi @SjoerdCdeVries just stopping in to see if anything is going on. Apparently not. Talk to you later.

@rcollyer Were you referring to my answer ("It's fine") when you pinged Szabolcs?