Want updates? Opt out any time

« Touring vs tunnel vision | Main | Brittleness and robovibes »

Segmentation and decomposition

Day 4 of the SEG Annual Meeting in Las Vegas was a game of two halves: talks in the morning and workshops in the afternoon. I caught two signal processing talks, two image processing talks, and two automatic interpretation talks, then spent the afternoon in a new kind of workshop for students. My highlights:

Anne Solberg, DSB, University of Oslo

Evan and I have been thinking about image segmentation recently, so I'm drawn to those talks (remember Halpert on Day 2?). Angélique Berthelot et al. have been doing interesting work on salt body detection. Solberg (Berthelot's supervisor) showed some remarkable results. Their algorithm:

  1. Compute texture attributes, including Haralick and wavenumber textures (Solberg 2011)
  2. Supervised Bayesian classification (we've been using fuzzy c-means)
  3. 3D regularization and segmentation (okay, I got a bit lost at this point)

The results are excellent, echoing human interpretation well (right) — but having the advantage of being objective and repeatable. I was especially interested in the wavenumber textures, and think they'll help us in our geothermal work. 

Jiajun Han, BLISS, University of Alberta

The first talk of the day was that classic oil industry: a patented technique with an obscure relationship to theory. But Jiajun Han and Mirko van der Baan of the University of Alberta gave us the real deal — a special implementation of empirical mode decomposition, which is a way to analyse time scales (frequencies, essentially), without leaving the time domain. The result is a set of intrinsic mode functions (IMFs), a bit like Fourier components, from which Han extracts instantaneous frequency. It's a clever idea, and the results are impressive. Time–frequency displays usually show smearing in either the time or frequency domain, but Han's method pinpoints the signals precisely:

That's it from me for SEG — I fly home tomorrow. It's tempting to stay for the IQ Earth workshop tomorrow, but I miss my family, and I'm not sure I can crank out another post. If you were in Vegas and saw something amazing (at SEG I mean), please let us know in the comments below. If you weren't, I hope you've enjoyed these posts. Maybe we'll see you in Houston next year!

More posts from SEG 2012.

The images adapted from Berthelot and Han are from the 2012 Annual Meeting proceedings. They are copyright of SEG, and used here in accordance with their permissions guidelines.

PrintView Printer Friendly Version

EmailEmail Article to Friend

Reader Comments (4)

Hi  Matt
That indeed looks like a great result on the salt picking. And in 3D? Awesome, thanks for reporting!

I agree there is tremendous potential for applying image processing methods in seismic processing and interpretation.

A couple of years ago I started playing with Otsu segmentation to separate multiples in tau p domain followed by ant tracking-like techniques to auto pick a mute line. I got some ideas and made some progress but never finished up.

It would be fun to organize a workshop or something like that...

November 9, 2012 | Unregistered Commentermatteo

I have avoided Empirical Mode Decomposition due to its obvious shortcomings. Han's paper suggests that it may have evolved enough now to be worth a look. Doesn't seem terribly complicated.

November 12, 2012 | Unregistered CommenterStewart Trickett

@Matteo: Workshop — definitely! Especially if it's the kind where people actually work. I especially like these volume-based, true 3D methods, that get away from picking surfaces for horizons and faults.

@Stewart: I gather the shortcomings are: it's expensive; you don't end up in the frequency domain; there's some mode mixing (one 'scale' getting mixed up with another); and it's not perfectly reversible. Maybe there are others? Han did a nice job of addressing these problems. He has a manuscript in with Geophysics, I think. I asked about code too, but no dice.

November 13, 2012 | Registered CommenterMatt Hall


The mode mixing was my biggest beef. A single mode didn't necessarily represent a single feature of the seismic.

I never worry much about speed. If it's a useful method then we always find some way around that. And the cost of processing is an order of magnitude less (at least) than the cost of acquisition, so moving a few more electrons around to get a better product is almost always worth it. And if it makes the processor's task easier then it saves money, as wetware is more expensive than hardware.

November 13, 2012 | Unregistered CommenterStewart Trickett

PostPost a New Comment

Enter your information below to add a new comment.
Author Email (optional):
Author URL (optional):
Some HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <code> <em> <i> <strike> <strong>