Check out this video. The only thing that stops you guys from doing the same is software. Surely there is one amongst us with the ability to write the software ?

PS Dont try and understand the attached picture until you have seen the video, it just wont make sence.

http://www.youtube.com/watch?v=89sz8ExZndc

Not much chop ? No one inspired ?

seems like alot of programming to me

more info needed,

could it be done using bits from a touch screen lilliput?

Very cool. This could be the future of computing input devices, right here. Was anybody else reminded of Minority Report (without the stupid-looking gloves)

Not quite a DIY project, though.

now that I know how they made that in that video, im much less impressed than when I first saw that video.
Unfortunalty for that group, that techniq has been used for quite some time..

Here is an image of my prototype frustrated total internal reflection rear-projected display.
Currently using a piece of perspex, blue CCFL light.

Fingerprint/point source is recognised through my laptop running linux, puredata, pdp+pidip using motion detection module after being placed through a binary filter on luminescense. I have since bought an infrared webcamera and am using IR sources to do the same thing as what's in the photo, except recognising infrared light emitted from fingers so as projection onto the screen will not cause confusion to the recognition system. Using a TV remote control as a point source the accuracy of pdp/pidip's motion recognition is 99% accurate around the IR source in the remote control! I am planning to step up to a CCTV camera as these are cheaper than 'handycams' and offer better mounting support. Plus, some of them go up to 60fps in 800x600 which is more than can be said for handycams and certainly webcams. Right now my output off this $7 IR webcam is about 5fps at 150xsomething really stupid, but for less than $20 this prototype has worked and given me the green lights for a proper display.

I am planning to backlight it with a white-LED short throw projector but I'm still waiting on hearing a success story for such LED projectors.

My concern with using this with our projectors is IR reflectance (i.e. stopping IR light from getting out of your projectors and causing erroneous hotspots). With a suitable filter (hot or cold mirror) it would probably work, though it might be better to use a dichroic mirror to block the specific wavelength the LED & camera use and use a dichroic filter of the same wavelength for the camera. That's assuming it's easier than tailoring the LED & camera to the hot/cold mirror.

Your progress so far sounds awesome though!

Good comments. Infrared filter material is cheap and only a small square would have to be placed over the projector lens if this was really required. It could even be salvaged from a webcam (we have done this to turn anti-IR cameras into IR-cameras)

However i think this is unnecessary for the simple reason of signal to noise ratios. (here signal is the fingerprint, noise is the backround IR radiation). I would believe that the output of a metal halide HID bulb in a projector is designed to dominate the visible spectral range as lots of IR would be inefficient to the bulb power usage. I have to steal my sisters projector and find an IR filter to check, however I would bet 10 bucks that the amount of 'noise IR' coming from the projector would be so low that it could easily be seperated from the 'signal IR'.

If it is not, then you simply add more IR leds around the plastic until the signal to noise ratio is better.

Also, in the 'visible range' photo you see there - the light was *not* baffled at the bottom (light is leaking out all in the room). The wall is a dark purple but you can obviously see blue reflecting off the wall and through the plastic - however compare that to the intensity of the light being emitted by my fingers and you will see that the signal to noise ratio it is no problem

(ps. theoretically with FTIR (baffled so 0% light leak) you could have such an intense illumination of the plastic that in a dark room it could not be seen (edges sealed) and when you touched the plastic it lit up the room - theoretically. moral of the story: if it doesnt work, shove more light into it until it does )

I also started constructing a multitouchscreen; see the PLOG

you dont need software....... you need the controller. 3M makes them. i have some if anyone is interested.....oh.......touchscreens are not leds.... theres no leds in them its a pice of glass sourounded by small wires....and there wight sensetive............ led's wow


i do repair on them........ if yall want to know how to do it.........i'll tell yall later. i get pics going. but you need plexi....comp controller, and speaker wire. ill show it in a month

The one in the demo is made with LEDS. Coheedisdead, don't be such an arrogant know it all. I am glad to see the abundance of laughing icons. The sad thing that the joke is on you. Instead of been quick to make a mockery of another person, why don't you make damn sure of your facts first.
If you bother to learn rather than make a mockery of people you can see on the guys web site who made this his own diagram. http://cs.nyu.edu/~jhan/ftirsense/index.html
Another video URL shows Jeff giving a run down of it all. http://www.ted.com/index.php/talks/view/id/65
Are you sure the roll eyes icon should be used? Jeff even mentions in his video about the software only been introduced in the last year.

These things are pretty cool, where's the plog with the guy building one??

EDIT: looks like its here: http://www.lumenlab.com/forums/index.php?s...17532&st=20

Any software as of yet? This would be cool to have just to play around with for hours and hours and hours and hours and hours and hours and hours and hours and hours and hours and hours and hours and hours and hours and hours and hours and hours and hours and hours and hours.

sweet this if something i could try with my modded web cam for IR