search

soft-matter / trackpy-examples

sample images, examples, and speed tests for trackpy
Other
21 stars 42 forks source link

Proposition for the gallery of examples: bubble velocity field in foams. #9

Closed sciunto closed 9 years ago

sciunto commented 9 years ago

Hi guys,

As I shown to Dan, I used trackpy to track bubbles in a 2D foams. To detect the position of bubbles, I need to use scikit-image directly. Would you be interested in an example showing that?

Is there any specific guidelines I should be aware of before starting?

Best.

tacaswell commented 9 years ago

I think that would be a good thing to have. One of the things about trackpy I want to make clear is that the finding the stuff step and the tracking the stuff step are completely independent (so much so, that the first round of trackpy didn't even have identification code in it).

I would try to keep the data volume sane (5 frame should make your point) and, if you are willing to put in the effort, write it at a level that new grad-students could make sense of it.

sciunto commented 9 years ago

Hi,

I pushed a proof of concept there: http://www.fast.u-psud.fr/~boulogne/Trackpy2.html @tacaswell @danielballan I would be happy to have your comments.

sciunto commented 9 years ago

PS: we submitted a paper in which we used this method. :)

tacaswell commented 9 years ago

Cool!

If you are publishing a paper with this, could you cite all of the libraries you use? Trackpy does not have a paper yet (in part because I am a bad academic, but @danielballan and @nkeim are working on fixing that) just a release DOI, but matplotlib (http://matplotlib.org/citing.html) and the rest of the stack (http://www.scipy.org/scipylib/citing.html) do have papers to cite. There are a number of people trying to find money/have academic careers built around maintaining and developing the scientific stack and the for better or worse, citations are academic currency.

danielballan commented 9 years ago

This is really nice. Thanks for taking the trouble to write it up. I can't think of anything I would have done substantially differently here; I have no revisions to suggest. If you open a PR with this notebook, we'll add it!

You might add a sentence or two at the beginning or the end about who you are and what your scientific purpose was. It's nice to show that trackpy isn't only used by its core devs.

Dan

On Sun Nov 09 2014 at 12:06:34 PM Thomas A Caswell notifications@github.com wrote:

Cool!

If you are publishing a paper with this, could you cite all of the libraries you use? Trackpy does not have a paper yet (in part because I am a bad academic, but @danielballan https://github.com/danielballan and @nkeim https://github.com/nkeim are working on fixing that) just a release DOI, but matplotlib (http://matplotlib.org/citing.html) and the rest of the stack (http://www.scipy.org/scipylib/citing.html) do have papers to cite. There are a number of people trying to find money/have academic careers built around maintaining and developing the scientific stack and the for better or worse, citations are academic currency.

— Reply to this email directly or view it on GitHub https://github.com/soft-matter/trackpy-examples/issues/9#issuecomment-62310519 .