Re: Inexpensive infrared edge/cutoff filters

[Otto Giesenfeld <otto.giesenfeld@xxxxxxxxxxxx>]

--On den 2 november 1999, 10:23 -0500 Stephen Intille
<intille@xxxxxxxxxxxxx> wrote:

>                        Emitters-2
>                     /               \
>                    /                 \
>                   /                   \
>                   --------------------- screen
>                                       |\ 
>        filter-c1                      | \
>         ----                0         |  \
>         |   >              \|/        |   Emitters-1
>         ----                /\        |  /
>          c1                           | /
>                                       |/
> 
>                              ^   filter-c2
>                             | |
>                             | |  c2
>                              -
> 
> Consequently, what I want is a *cheap* solution to do the following: 
> 

May I suggest a slightly different solution? Replace the evenly illuminated
screens by arrays of IR-emitting LEDs. Use the cheapest LEDs you can find
that still output a reasonable amount of IR, and disregard the exact
wavelength. Preferably, they should project their light in a rather narrow
angle.

Then set the apertures of the video cameras so that images of individual
LEDs are just barely registered as 100% white. The idea is that by
recording point sources instead of diffuse sources of IR, it will be easier
to separate the interesting signal from noise (i.e. reflections off the
subject's clothes/skin).

This approach might also save you from requiring the user to turn any
strong spotlights etc off during sessions (not to mention the problems
sunlight from a window could cause). Especially using narrow-band filters
on sources that are not narrow-band to begin with should demand fairly
strict requirements on lack of ambient light.

The main problem I see is in finding cheap enough LEDs to allow the
construction of arrays that offer enough resolution. (Quick research at the
web site of an electronics supplier indicates a price of $0.06 per LED when
buying more than 500. This would make 100x100 arrays pretty expensive. But
who knows, the price of 20 000 LEDs might be $0.02 each.) An alternative
could then be back-lighting of perforated screens (having, perhaps, 5 mm
holes and a 20 mm grid spacing).

It could at least be worthwhile to make a quick test to see if the
underlying assumptions about relative light levels are reasonable.

Otto Giesenfeld

*
****
*******
******************************************************
*  To remove yourself from this list, send:          *
*         UNSUBSCRIBE INFRARED                       *
*       to                                           *
*         MAJORDOMO@xxxxx                            *
*----------------------------------------------------*
*   For the IR-FAQ, IR-Gallery and heaps of links:   *
*  http://www.a1.nl/phomepag/markerink/mainpage.htm  *
******************************************************