1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Visualization of detector data / other ideas for viz application

  1. Oct 14, 2013 #1
    Greetings friendly physics folks,

    I could use some advice and/or general suggestions regarding a data visualization project I'm looking to get started on. If this looks like too much reading, please skip to the TL;DR below.

    The basic project components are:

    (1) A GUI of some kind, serving as a shell for the application.
    (2) Use of at least 3 visualization techniques (methods of surface/volume rendering in application window, etc.), and 3 data analysis/mining techniques. The data analysis can be done in any common language (I'll probably use C++, Python, or some combination thereof), but the visualization component must involve rendering through VTK or Processing (likely VTK in my case, I should think).

    The instructors gave a general suggestion to think of it as building a kind of simple-but-specialized version of ParaView (just search Google images, if unfamiliar). As far as the topic is concerned, it can literally be any kind of data I can get my hands on.

    I'm the type of person for whom that range of possibilities is much too broad, but given my physics/math concentration, I know that I would likely enjoy working on something physics related. In an effort to do that, I talked to a professor of mine and have been given access to some datasets from CLEO (http://www.lepp.cornell.edu/Research/EPP/CLEO/), as well as some of the computer resources used for particle physics data analysis at my university.

    Given my lack of familiarity with particle physics, I'm uncertain whether it's even possible to meet the project requirements with this kind of data (obviously I'm concerned about the visualization components, in particular). I know that it's possible to reconstruct collision events to some degree and create those pretty CERN-esque pictures by tracing particle trajectories, finding energy deposits, etc... but I get the impression that those kinds of things are typically for public relations -- not for revealing or making intuitive any information that would be useful or enlightening to a physicist. There is seemingly lots of interesting data analysis that can be done on this type of data, but I need to be able to visualize it in some high level way, not just plot statistical analysis results on a 2-D graph.

    If anyone on PF is familiar with particle detector data analysis, I would very much appreciate any encouragement or discouragement. As it stands I'm just wondering whether it's feasible to use this data or not. (My instructors are CS people, and I'm a little intimidated by my professor).

    In addition, or just regardless, I would be interested to hear of any particular kinds of available data that you all think might work better. I know that astrophysics data in particular is much more open and available, but don't know where to start, or what kinds of things would lend themselves to 3D visualization.

    TL;DR: I need to build a data visualization application that uses VTK (3D) or Processing (2D). Data can be from literally anything I find interesting and can get access to. I'm looking for advice regarding the feasibility of using particle detector data (which I have access to), or for any other suggestions of physics-related data that might be interesting to work with. (Come to think of it, I'm really open to any kind of data that someone on PF finds interesting).

    Thanks in advance to anyone with even minor suggestions!

    [P.S. If anyone is concerned about academic dishonesty, I can assure you that our grades are determined entirely by the quality and functionality of our source code, not on the choice of dataset. The instructors have explicitly encouraged us to talk with others, especially those in our fields of interest, about finding, choosing, and looking at the important features of our data.]
  2. jcsd
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Can you offer guidance or do you also need help?
Draft saved Draft deleted