Friday, August 10, 2012

Regular Image Sensor Used for Radiation Detection

Vienna Technical University, Austria presented a way to measure Alpha, Beta, Gamma particles with a regular webcam. The webcam sensor is shielded from visible light by a 5um-thick aluminium foil and a software is used to analyse the video stream for characteristic patterns:


The poster has been presented at SIGGRAPH 2012:

"GeigerCam: Measuring Radioactivity with Webcams"
Thomas Auzinger, Ralf Habel, Andreas Musilek, Dieter Hainz, Michael Wimmer

15 comments:

  1. Wonders one can do with just a webcam! Does anyone of you know if imagers are affected by magnetic fields ( I mean can magnetic fields be detected by imagers or could they be modified for detection?)Can any modification of an imager be used for sound detection? (like SONAR)

    ReplyDelete
  2. I'm not sure about magnetic field imaging, but Plessey offers EPIC sensor for electric field visualization:

    http://www.plesseysemiconductors.com/products/epic/

    But why do you want to use image sensor for sound? What's the problem with microphones?

    ReplyDelete
    Replies
    1. To make it more compact, and possibly of higher resolution/sensitivity. Of course there are MEMS based acoustic sensors, with it's own set of problems. I am thinking in the lines of membrane/s stretched above the image sensor and illuminated with monochromatic light. The sensor reads the acoustics-generated interference pattern to later reconstruct the amplitude, phase of the acoustic wave.

      Delete
    2. As for such MEMS based microphone, Toshiba has been working on it for a while.
      http://www.toshiba.co.jp/tech/review/2005/08/60_08pdf/f02.pdf

      Delete
  3. Sure it is kind of interesting that some computer science guys rediscovered what the nuclear science guys have known for a long time. And if you decide you want to build a better radiation detector you will wind up where they have already been. Still, now maybe someone can build an iPhone app for radiation detection.

    Using image sensors for detecting radiation is almost as old as image sensor technology. Certainly as old as DRAM technology. DRAMs were very good at detecting alpha particles, for example. (ha!) CCDs give a nice response to Fe-55. I remember Martin Buehler at JPL flying around with SRAM radiation detectors in airplanes, and George Soli at JPL using our earliest CMOS APS devices for radiation detectors as well and proposing to fly them on spacecraft. I think our computer science friends should consider a little literature search.

    ReplyDelete
    Replies
    1. It's more than just radiation detection. They track the real-time dark noise statistics to find the right threshold. Then, there is an analyzing step when the particle energies are estimated. And all is software-based and done in real time.

      Delete
    2. Vlad, I think this is pretty much what the nuclear scientists do on a daily basis. But, it would be good to hear from them on this subject.

      Delete
    3. Not with a $20 web-cam! Doing nuclear science with a $20 webcam (poor-mans solution) is more technologically challenging than with a $1 million setup.

      Delete
  4. Eric is confusing things. This is mainly a software effort to identify particles: to my understanding the authors never claim to have invented a novel method to detect radiation. By going along Eric's logic, no one should be working on image processing algorithms since image sensors have already been invented. Maybe the next argument would be that no one should be flying around with SRAMs in their luggage (who would do that) because one of his pals already did it ;)

    ReplyDelete
    Replies
    1. Well, you guys are certainly more enthusiastic than I am about this software work. It just reminds me of people discovering that they can make an image sensor by taking the lid off a DRAM device. As such, I was just very underwhelmed by the poster and its lack of acknowledgment of any similar work done in the past.

      Delete
    2. Well, it's a poster, not an article. The more info you cram about other works in the same field, the less you can put about the actual work you did.

      Delete
    3. It is hardly a journal paper: it is a poster. Usually references are limited and not an absolute must.

      Delete
    4. I think posters should communicate some related work. We don't have to agree on this, but it is the academic norm and these are academic guys.
      I think they are just unaware of it rather than making a deliberate decision not to include it. I also expect many readers here are not aware of the related work either so I just mentioned work I was adjacent to at JPL.

      Delete
    5. Once we talk about history and nowadays, the interesting twist is that the modern SRAMs are more sensitive to radiation than the modern DRAM chips. So, if one makes a radiation sensor out of memory chip today, he'd better take SRAM.

      Delete
    6. Excellent application. It looks like they are using whole imaging array as a single pixel detector without imaging capability yet. In order to fully take the advantage of a modern image sensor I would suggest using a coded aperture lens with patterned metal. I'm sure they thought about this.

      http://en.wikipedia.org/wiki/Coded_aperture
      http://www.sron.nl/~jeanz/cai/
      My favorite telescope :
      http://www.sciencemuseum.org.uk/objects/space_technology/2005-33.aspx

      I guess it can give direction information of the incoming radiation! It will be very useful for the lonely guys walking around slowly, turning around themselves, with geiger counters or similar equipment in their hand ... :)

      Delete

All comments are moderated to avoid spam.