Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Holographic Inteferometry

  1. Jun 25, 2007 #1
    We are attempting here at Illinois Wesleyan to create a holographic speckle-pattern inteferometer. We have been working at it for some time and have come to a roadblock in terms of picture quality. We're using a high quality optics table so there are no table vibrations. All of the mounts are tightly and securely mounted. We are using an image multiply system and the standard image-subtraction with Labview. We have an image, but we're looking for other adjustments or suggestions to try to improve our picture. If anyone else is doing similar research we'd extremely appriaciate recommendations for better picture quality.

    Follow the link below for the current quality of our efforts. In this image we are driving a metal plate at a resonant frequency of 1542 Hz. This is the best image quality we have right now and to show what we are supposed to see, I have outlined the pattern in red.

    http://www.totalfission.net/pic.jpg
     
  2. jcsd
  3. Jun 25, 2007 #2

    Danger

    User Avatar
    Gold Member

    Welcome to PF, Fissionman.
    I gotta admit that I've never heard of one of these things before, but I have a question for you anyhow.
    Is your resolution problem related to the holography equipment itself, or to the image sensor?
     
  4. Jun 25, 2007 #3

    berkeman

    User Avatar

    Staff: Mentor

    Interesting background reading on speckle pattern interferometry:

    http://en.wikipedia.org/wiki/Speckle_imaging

    FissionMan1, how does the holographic aspect factor in? Can you provide a tutorial web link or two to the kind of setup that you're using?
     
  5. Jun 25, 2007 #4

    Danger

    User Avatar
    Gold Member

    Hi, Berkeman. Yeah, it's interesting... what little of it I can understand. This was the site that I checked up on just to know enough to ask my first question.
    http://www-staff.lboro.ac.uk/~mcjmh/SIR/speckle.html
    Do I even want to know what 'temporal phase unwrapping' is?
     
  6. Jun 25, 2007 #5

    berkeman

    User Avatar

    Staff: Mentor

    Cool link, thanks Danger. Now I see what the basic setup is, and where the holographic interference part is used. I'm going to ping an optics specialist friend of mine at HP Labs to see if he has any input.
     
  7. Jun 25, 2007 #6

    Danger

    User Avatar
    Gold Member

    That could be amusing. :biggrin:
    Until I read that line about 'under 100 nanometres' resolution, I was about to suggest switching to a shorter wavelength laser. Apparently, though, that wouldn't make much difference. (And I still don't know whether it's the beam or the camera that's causing the trouble.)
     
  8. Jun 25, 2007 #7

    berkeman

    User Avatar

    Staff: Mentor


    FissionMan1, what does the target look like in a real photograph? From the interferometric aspect of your question and setup, I'd guess that the pattern you've outlined in red is the topographic step of the target? What other target patterns have you tried?
     
  9. Jun 25, 2007 #8

    Danger

    User Avatar
    Gold Member

    What it reminds me of most is the sort of thing you see when powder is put on a vibrating plate to show where the nodes are.

    Oh, well... 6:00. Time to head home for a beer. Back shortly.
     
  10. Jun 26, 2007 #9

    NoTime

    User Avatar
    Science Advisor
    Homework Helper

    IIRC the entire image can be reconstructed out of any segment of the mixed reference beam/Image beam.
    As a practical matter, both the image sensor area and sensor pixel count would contribute to the resolution.

    Also, I vaguely remember from somewhere that the motion of the plate is instantaneously chaotic. The motion averages out to the expected pattern over a period of time. To get the expected image you would need to sum the images from several cycles.
     
  11. Jun 26, 2007 #10
    Berkeman, you're the greatest. We're so confused as of right now as of where to go.

    As for NoTime, the problem with that is that our setup is too unstable to average the images over several frames. We're on an air table and have made sure that all the instruments are locked down tight and not on stands that are too tall, but without vibrating the object we still see some vibration lines. This impacts the image when we are vibrating because there are dark lines that move over the image on the order of 2 or 3 hz. I am unable to figure out why this is. We also have not yet accounted for speckle size vs. pixel size (read below if you don't know what I mean) which could compound our problems, but the main thing seems to be stabilizing the image.

    For all of you wanting to know what this stuff is, check out Thomas Moore's paper in AJP on building a simple speckle pattern interferometer. That is the paper we are working off of for my research on vibrating instruments. Although at the moment, we're using a circular plate in order to calibrate the stuff.
     
  12. Jun 26, 2007 #11
    Also, berkeman, the target is simply a flat circular plate which I outlined in red. The red curves are the boundaries of the 6-pointed pattern which we see at this frequency resonance. Basically, at every resonance of the object, it produces holographic patterns that we record (this in time will be used for some other applications we're working on). At different resonances we achieve different types of patterns. If you are interested, read the AJP article I referenced previously.
     
  13. Jun 26, 2007 #12

    berkeman

    User Avatar

    Staff: Mentor

    Can you maybe use a Pockels cell instead of the vibrating plate? Does the Pockels cell produce varying phase delay without any moving parts?
     
  14. Jun 26, 2007 #13
  15. Jun 26, 2007 #14
    From what I am seeing, Pockel cells are the vibrating object that we would put in front of the laser. Our goal is to be able to study many different kinds of things, not just pockel cells. But I really don't know what one of those are anyways, lol.
     
  16. Jun 26, 2007 #15

    Danger

    User Avatar
    Gold Member

    Could it be a strobe effect caused by the interaction of the vibrational frequency with the frame rate of your image capture?
     
  17. Jun 26, 2007 #16
    If it were a strobe effect it would affect the entire picture each time it happens, but unfortunately this is not the case. They are vertical lines that scroll across the screen.
     
  18. Jun 26, 2007 #17

    berkeman

    User Avatar

    Staff: Mentor

    If it were a beating effect, the lines would be horizontal, I would think. Unless the raster scan of your video system is rotated 90 degrees for some reason.
     
  19. Jun 26, 2007 #18
    We do get horizontal lines as well, but thats when we are not purposefully vibrating the object.
     
  20. Jun 26, 2007 #19

    Claude Bile

    User Avatar
    Science Advisor

    Have you tried imaging the beams using a different camera, or even just onto a sheet of paper and viewing them by eye? That should give a fairly strong indication as to whether the problem lies with the camera, or the intermediate optics.

    Also, how stable is the optimised setup? Is the image quality liable to plummet if someone very slightly misaligns one of the optics?

    Finally, Pockel's cells are crystals that have a refractive index that varies with an applied E-field. One can therefore modulate the refractive index of such a crystal (in time), by modulating the appled voltage.

    Claude.
     
  21. Jun 26, 2007 #20

    Danger

    User Avatar
    Gold Member

    Drat! I'll keep paying attention to this, but I'm afraid that my limited (ie: nonexistent) exposure to this subject is a tad disadvantageous.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?



Similar Discussions: Holographic Inteferometry
  1. Holographic dispersion (Replies: 6)

  2. Holographic Me (Replies: 7)

  3. Holographic projector (Replies: 4)

Loading...