Using Circular Polarizing Filters for Digital Infrared Photography

Discussion in 'Digital SLR' started by Wayne J. Cosshall, Aug 4, 2007.

  1. Wayne J. Cosshall, Aug 4, 2007
    1. Advertisements

  2. Can Digital HDRIR be far behind?

    Ever tried using a pair of crossed linear polarizers? As you approach 90
    degrees, the spectrum of light that makes it through apparently gets
    weighted toward the red and IR. Not an appealing prospect to add these
    to a filter stack, and the pair didn't substitute well for the IR pass
    filter, but I was amused anyway.

    Observed with unmodded Sony F828 in "nightshot" mode. Crossed linear
    polarizers stacked with IR pass filter, in an attempt to reduce the need
    for additional ND filters. (Yeah I know, I should just go get the thing

    Imagine my surprise when I saw the IR target scene not fade at all as I
    approached 90 degrees.

    Several filters have unexpected effects when used outside their intended
    spectra. A Hoya X1 *green* filter is more effective at attenuating IR
    than yer basic 2 or 3 stop ND filter.


    It Came From Corry Lee Smith's Unclaimed Mysteries.

    In a time of deception telling the truth is a revolutionary act. -
    George Orwell
    Unclaimed Mysteries, Aug 4, 2007
    1. Advertisements

  3. Interesting results. When I use my old Sony TRV9 (FS, BTW,
    in very nice low-use condition...) Mini-DV camcorder (mine is
    daylight IR enabled) with an IR filter, I also generally add a
    circular polarizer, which gives a little more control over the
    image, as you found. Some samples shot from a car are at
    David Ruether

    (see our "restaurant menu", at
    David Ruether, Aug 4, 2007
  4. Wayne J. Cosshall, Aug 5, 2007
  5. Any of the Sony 1-CCD camcorders can be modified to prevent the
    IR switch from also forcing the camera to widest aperture and too
    slow a shutter speed for daylight IR video (first installed in late TRV9s
    due to Sony's silly prudery and the myth of "x-ray" vision for the
    camera - but fortunately, mine is an earlier version, and not hobbled
    by this nonsense). One day I tried it outdoors, and it worked fairly
    well, so I added a red filter, and it worked better yet. A polarizer
    improved the results further, and finally replacing the red with an IR
    filter worked very well. I use the camera in B&W mode to lose the
    "tooth paste green" look...;-) BTW, I prefer IR in motion to stills
    (it is just plain more fun - and somehow less "hokey" looking...).
    David Ruether, Aug 5, 2007
  6. Wayne J. Cosshall

    RichA Guest

    Why use any of that junk when you can buy interference filters that
    have almost dead-stop cutoffs at any wavelength you'd like? You can
    get cutoff filters, narrow band filters or even notch filters, all
    with steep attenuation and throughput.
    RichA, Aug 6, 2007
  7. A few remarks, FWTW. There are two different influences affecting the
    final results in your IR pictures. A circular polarizer consists of two
    filters: a linear polarizer followed by a quarter-wave retarder. Neither
    of these is fully achromatic, that is, they only have a limited
    wavelength range over which they are fully effective. Outside its range
    the linear polarizer will have decreasing effect in polarizing light,
    and outside its range the retarder will convert less of the lineraly
    polarized light to circular. Both tend to start failing rather badly as
    you go into the IR, so I would expect diminished affect by rotating the
    filter in the IR. How much depends on the specific polarizer and could
    change with maker.

    For my research I had achromatic polarizers and retarders fabricated
    that worked from the UV to the IR, but they were very expensive. Normal
    ones made for cameras start losing there effectiveness in the blue and
    the red, only working well in between.

    Joseph Miller, Aug 6, 2007
  8. Wayne J. Cosshall, Aug 7, 2007
  9. The custom filters would not work with normal cameras. They are quite
    bulky, fragile, and can require careful alignment. I'm an astronomer and
    use them on very big telescopes (e.g., the Keck Telescopes) to measure
    polarization of faint astronomical objects.

    Joseph Miller, Aug 7, 2007
  10. Wayne J. Cosshall, Aug 7, 2007
  11. I actually did partly work my way through college as a professional
    photographer in an observatory, but with film and and a large darkroom.
    I'll take digital any day. We closed all the darkrooms in our
    observatory many years ago. I don't do much, if any, direct imaging, but
    primarily spectroscopy, which is of course taking images of spectra. We
    have our own CCD fabrication lab and are now working on a 4K x 4K device
    with 15 micron pixels- it's big. We have made mosaics of 8 2K x 4K
    devices that work very well (64 mp!), but each CCD cost around $100K.
    Even the CCD controllers we build can cost upwards of $100K, but that's
    part of the price of getting below 2 electrons read noise with a fast
    read out. I say all this as background to how impressed I am by what you
    can buy for a few hundred dollars. The technology in even an inexpensive
    P&S still amazes me, maybe because I know how hard to do this stuff can
    be. There's nothing like mass production to bring the cost down. Our
    problem is that almost everything we build is a one-off. There are
    probably less than 100 CCDs in the entire world, maybe less than 10,
    that we would consider suitable for a recent instrument we delivered.

    Joseph Miller, Aug 8, 2007
    1. Advertisements

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.