View Full Version : New problem with Digital still cameras during tapi


Mike Rehmus
June 27th, 2003, 08:24 PM
Looking at the last wedding footage I found that the newest high-resolution digital cameras (or at least the one they used at this wedding) now project an pattern in IR light onto the subjects. Our video cameras are great IR detectors so now, before each flash, one sees a picket fence pattern of IR light on the subject for far longer than the flash duration. Impossible to eliminate this.

I will have to see if I can find a dichroic filter that can limit bandpass to visible light or just cut out IR light.

Peter Moore
June 27th, 2003, 08:57 PM
Wow that's amazing. What kind of camera was being used?

Mike Rehmus
June 27th, 2003, 09:54 PM
I think it was one of the Nikons with the 35mm body. But I"m not certain. There were about 3 camers in use during the wedding and reception.

Or it may have been one of the new Sony's with their new autofocus system that they've been bragging about. I'll have to ask the two photographers what they were using.

Really ugly effect.

Eric Lian
June 27th, 2003, 10:54 PM
True IR is invisible to the eye. What you're seeing is a very real and visible red light on the subject. As far as the camera being an IR detector, if you took your eye away from the eyepiece, your naked eye would see the same grid.

The "IR" grid isn't new technology. I had it on my Canon A2s which are more than several years old. The ultimate advantage, at least from the photographers seat, is you can auto-focus in low-light or no-light conditions.

I'd be surprised if you could find a video filter to combat the problem.

What's the difference? Walk into a dark bathroom with your TV remote control and press the button. See any light being shot out from the sensor? Can you bathe the walls with a red glow? No. True IR is invisible to the eye.

Now, take my Canon camera and walk into the same bathroom. Press the shutter release half-way down and the bathroom is lit up like a house-of-ill- repute. Yes, there is an IR component, but you can't see it. The camera needs visible light (the grid) to calculate focus distance. They could have used white light like some of the cheaper cameras, but red light is far less annoying - except to videographers. ;)

James Emory
June 28th, 2003, 02:13 AM
Eric,
I'm glad you cleared that up because I too was wondering how his camera was seeing infrared light and why he may not have seen it with his own eyes. I have tried that trick with a remote control and a night vision security camera that produces its own IR beam to see in the dark. The IR beam is so bright, you could literally use any remote as a flashlight. This red grid that is mentioned reminds me of Predator's targeting system.

Adrian Douglas
June 28th, 2003, 05:26 AM
So how come I can see my so called IR remote control beam when I look through the viewfinder of my camera, but cannot see it with my naked eye? What's up with that?

Glen Elliott
June 28th, 2003, 08:13 AM
I shoot stills with a Sony F717 and it has an option for holographic laser assisted focus. It shoots a lazer grid pattern on the subject to help gain focus in dark environments. The laser looks like a red pattern- maybe the same thing Mike is seeing. I honestly don't understand how our cameras can see IR unless they have some sort of night-shot option which makes the lense sensitive to IR. I've never tested using a tv remote or whatever to see if I can actually see the IR beam through my viewfinder/lcd.

Don Parrish
June 28th, 2003, 10:23 AM
If you look at a remote control that is IR you cannot see it, look at that same remote control thru the XL1S and you can see it, these cameras in fact do pick up IR, I have seen it myself.

Eric Lian
June 28th, 2003, 11:09 AM
The XL-1s can see invisible light? That's news to me. I own an XL-1 (not the "s")
and that camera can't see IR. Must be a new feature. If someone could post a brief video clip of this feature, I would sure like to see it.

Regardless of whether the XL-1s can see IR, the red grid Mike was seeing in his original post was "visible light" used as a targeting mechanism for the auto-focus. As James said, just like Predator. :)

Boyd Ostroff
June 28th, 2003, 11:32 AM
Here's a very interesting website (http://www.cliffshade.com/dpfwiw/ir.htm) that discusses infrared photography with digital cameras. Lots of good technical information there. Many cameras have internal filters to remove IR actually. Using the technique they suggest, I took my VX-2000 and pointed a TV remote control at it. I could clearly see the flickering infrared emitter, which is invisible to the naked eye.

If you check out the IR galleries at the end of that webpage there are some beautiful stills. When I get a little time I plan to get an IR filter and experiment around a bit with my digital still and video cameras.

Ken Tanaka
June 28th, 2003, 11:34 AM
Yes, CCD's certainly do "see" some of the light that falls within the infrared portion of the spectrum. Try Don's simple experiment if you remain unconvinced.

The "picket fence" that Mike reported seeng is, however, most likely related to auto-focus systems such as Glen's camera. These are more red laser-like devices, not infrared.

Mike Rehmus
June 28th, 2003, 11:34 AM
Guys, I know the cause, I'm just alerting you all to the problem.

I was in a startup company that built the first digital cameras back in the late 70's. (and the first desktop scanners in the 80's) CCDs are almost twice as sensitve to IR frequencies as they are to visible light. To make a nightshot camera, all sony does is remove the IR filter that is in front of the CCD on all CCD-based video cameras. If you take one apart, you will find this bit of glass that is slightly blue-green in color. That's the IR filter.

The IR sensitivity, even with the IR filter in place is why hot bodies like candles and lights cause even greater streaking problems than their visible light output would suggest.

I'm going to be looking for a 'bandpass' dichroic filter that might be available (it will be very expensive) that might let me ignore IR. The bulk-dyed glass filters are very broad and not very good filters for this application.

Yup, you can check out any IR source with your video camera. Trust me, your XL1 can see IR. Point it at your TV control and press any button while viewing the image on your camera. You will see the bright pulsing or steady light on the end of the control.

Phil Reams
July 2nd, 2003, 09:32 PM
Hi Mike,

I could be wrong, but I believe the Edmund Scientific optics catalog has the IR bandpass (or would that be "reject"?) dichroic filter you might be looking for.

I remember seeing one a short time back while I was looking for an optical blank to grind my own telescope mirror. If I find it online, I'll post the link.

-Phil

Dan Holly
July 3rd, 2003, 09:59 AM
I have the same issue Mike explains with my Sony cybershot 2.0 MP camera and my XL1s. I've never tested it with my GL1.

The model cybershot I have only projects this type of light right before a flash....to reduce the "red eye". For this simple reason, I never use it while filming (duh =) unless the Sony does not require the flash.