Blogs

920nm IR filter for penetrating the mist

Posted by Melih Karakelle on January 4, 2011 at 3:09am
Hi Guys,
We are trying to penetrating the mist with the camera that sensitive on IR spectrum.
The video from last days of 2010. The weather was sunny but mist covering the detail of the islands. We tested 720nm, 850nm and 950nm IR photography lenses and 920nm was the best one.  
920nm looking useful but camera sensitivity reducing in this spectrum and need more light.
I will try same setup on misty day on shorter range and will share the results of the experience, maybe it helps someone.
Chers and happy new year
Melih
Tags: 920nm, ir, mist, spectrum
E-mail me when people leave their comments –
Follow

You need to be a member of diydrones to add comments!

Join diydrones

Comments

  • Lew Payne January 6, 2011 at 7:29pm

    This Rohm video encoder IC can "see" through fog, without the use of IR filters which reduce Lux.  How well a job does it do?  Check out this impressive video!  What are some of the techniques they use?  Well, you may get a hint by reading this Princeton article on how noise gets turned into information.

     

    Note: Use at your own risk.  Void where prohibited.

  • Developer
    John Arne Birkeland January 5, 2011 at 10:08am

    The problem is how manufactures specify lux without relevant data to specify video quality. This makes the lux value useless. Using slow shutter and multi exposure techniques you can pretty much get any lux rating you want, but the result is useless for motion footage. For a proper lux rating you also need to know the shutter speed and IRE rating.

  • Sebastian Gralla January 4, 2011 at 3:26pm

    impressive camera...

    nice work!

  • T3
    Krzysztof Bosak January 4, 2011 at 7:18am
    resolution... not the same ballpark.
  • evilmacaw January 4, 2011 at 6:23am
    check out the this camera it is an analog CCTV type but see`s at extremely low light levels  so with your filter it should be a nice bright picture not digital but possibly workable
  • T3
    Krzysztof Bosak January 4, 2011 at 5:47am

    "wouldn't active camera stabilization relax the quick exposure time somewhat?"
    Of course not. Not in the 1/200 of a second timescale.
    You would need 18bit servo precision, some 10KHz imu and similar servo update rate,
    all this should be moving a part that weights 500g or more and is subject to 20G accelerations during landings.
    Camera stabilisations makes parallel photos, this is the key point in map stitching,
    or else the whole process goes much worse, in the bad day the results could become completely skewed.

    So you need bright, thus large and heavy lenses. And a compact plane. This analysis rules out 90% compact cameras with most modern sensors (yet more optical distortion) and some sub-DSLR sized digitals.

  • Michael Zaffuto January 4, 2011 at 5:38am
    @Krzysztof, wouldn't active camera stabilization relax the quick exposure time somewhat?
  • T3
    Krzysztof Bosak January 4, 2011 at 4:42am

    Are you able to keep exposition times below 1/200s?

This reply was deleted.