help with infrared

[ifurlong]

I got the Hoya R-72 infrared filter. It seems to have an area in the center that is much brighter that the rest and it tapers off a lot twordes the edges, I thought that it was going to be black and white instead of red, although the red is not a problem. Did I ret the wrong filter?

[Icebreaker]

I had this problem when I shot with a Minolta DiMage 7i. In that instance it was due to internal reflections. Now It may not be the same here, but it sounds very similar.

I recently bought a Digital Rebel, has anyone tried infrared on it yet??? I would be interested in hearing about your sucess or failure

Hope this help, and thanks to any response

[MediaMagic]

ifurlong wrote:
I got the Hoya R-72 infrared filter. It seems to have an area in the center that is much brighter that the rest and it tapers off a lot twordes the edges, I thought that it was going to be black and white instead of red, although the red is not a problem. Did I ret the wrong filter?

I have the R-72 as well and the center does have the hot spot. the R-72 allows a bit of red light to pass, leading to the reddish image, where the Tiffen 89? can't remember the model has a lower wavelength cutoff.

Here is my first goof around with the false color stuff. You can clearly see the "hot spot"...

http://images.fotopic.net/?id=1249689&outx=600&noresize=1&nostamp=1

I have no idea how to correct this. I've tried messing with the curves a bit and even selecting the perceived hot area and toning it down, but not with any real satisfying results so far. Maybe someone else knows how to get rid of the center hot spot?

[dwbrant]

Hi

The problem you're seeing is likely to be due to one of the following two instances:

1) Most likely -- the lens you are using is guilty of producing the "hot spot" in the middle. There are several Canon and non-Canon brand lenses which have a coating which cuts down the transmission of infrared-spectrum light... I posted a list on this forum of the lenses, here's a link (you might have to go to the top of the page for the list):

http://photography-on-the.net/forum/showthread.php?t=18698#103702

Hope the link works. For your convenience, I am pasting the list as currently known below, please follow the link for further information.

========/snip/=========

Good

Canon EF 17-40mm f/4 L
Canon EF 24-70mm f/2.8 L
Canon EF 28-135mm/3.5-5.6 IS
Canon EF 28mm f/2.8
Canon EF 50mm f/1.8 MKI
Canon EF 50mm f/1.8 MKII
Canon EF 70-200mm f/4L
Canon EF 75-300mm/4.0-5.6 IS
Canon EF 135mm f/2L
Canon EF 100-400mm4.0-5.6 IS L
Canon EF 100mm f/2.8 macro


Bad

Canon EF 16-35mm f2.8 L
Canon EF 20mm f/2.8
Canon EF 35mm f/2
Canon EF 24-85mm f/3.5-4.5
Canon EF 28-70mm f/2.8L
Canon EF 28-105mm f/3.5-4.5
Canon EF 35-80/4.0-5.6
Canon EF 50mm/f1.4
Canon EF 50mm/f2.5 Macro
Canon EF 70-200mm f/2.8L IS
Canon EF 85mm f/1.8
Tamron 70-300mm f3.x-xxx Macro

========/snip/==========

2) Another possibility is that you're seeing a type of lens flare -- I suggest you try shooting at several different angles to the sun, 90 degrees seems to work best.

By the way, the red color you are seeing.. is probably because you're using automatic white balance -- if you use custom white balance and set it to a pic taken with the Hoya filter on, you should get sort of brown skies and vegetation should be lighter colored.

There is a "fix" for the hotspot. Please follow this link for the method posted on dpreview:

http://forums.dpreview.com/forums/read.asp?forum=1019&message=5439017

Good luck! Beautiful IR shots are possible with the Canon 10D and I'm sure the 300D as well.

[dwbrant]

In the case of the D7i, the hotspot was due to the existince of a stronger IR-pass filter just in front of the CCD. The D7 was the only model of that line that was capable of good IR pics, and boy did it work (take a look at http://public.fotki.com/dwbrant/infrared_photos/ for examples from the D7).

I'm betting the 300D is just as capable as the 10D of IR pics, do you still have the filter? If you do, give it a shot. I bet there's somebody who has already tried it, if not here, try the forums at dpreview:

http://www.dpreview.com/forums/forum.asp?forum=1031

Good luck!

[mattchase]

Here are a few more lenses to add to the list, tested on a 10D with a Kodak Wratten 89b IR filter.

Good
Canon 17-35 2.8 L

Bad
Sigma 70-200 2.8 ex
Tamron 28-75 2.8 xr di

[MediaMagic]

Thanks for the tips and links. It looks like ifurlong and I both have "good" lenses to try with. Back to the practice area! ifurlong, what are you shooting the IR with?

I agree I've seen some excellent IR shots from the 10D. I've also heard that one must slightly front focus visually before placing the filter. That's tough to do. I still have yet to get one in acceptably sharp focus. I suppose once you get it right, you'll just know what you're looking for after that. But, I stopped trying because of that center issue. Looking foward to trying it again with the new information.

David

[Icebreaker]

Today I went and bought a step down ring so I could try the Hoya R72 filter on my new Digital Rebel. I tried it on my EF-S 18-55mm 3.5-5.6 and EF 75-300mm 4-5.6. I am pleased to report NO HOT SPOT. so now I'm going to order a 58mm filter to try. I'll post another report after I've had a chance to test it.

LDW

[dwbrant]

MediaMagic wrote:
Thanks for the tips and links. It looks like ifurlong and I both have "good" lenses to try with. Back to the practice area! ifurlong, what are you shooting the IR with?

I agree I've seen some excellent IR shots from the 10D. I've also heard that one must slightly front focus visually before placing the filter. That's tough to do. I still have yet to get one in acceptably sharp focus. I suppose once you get it right, you'll just know what you're looking for after that. But, I stopped trying because of that center issue. Looking foward to trying it again with the new information.

David

You know, I was also convinced when I first tried IR with the 10D that focusing would be a huge pain in the butt.

"But" I was mistaken. I read in another forum post that most people were having success with autofocus when the filter was on.

What I discovered was that yes, AF does work with the filter on... however, it seems to only be accurate when you're shooting something close. If you're shooting a landscape, you're much better off a) using ISO 100 (reduces noise in the final image), b) using a medium to "deep" f-stop (say, f9 to f22) to get good depth-of-field, c) using a cable release or timer to make the shots, and d) not trusting auto exposure (it's best to set to manual, set your aperture and focus, then make your "best guess" as to what shutter speed to use). Since the shutter speeds will inevitably be at least 4-5 seconds with the Hoya R72 on, camera shake even with a wide-angle lens will be an issue.

I included the bit about auto exposure because on my camera, when my exposure is correct according to the histogram of the image I've just taken, the camera "says" that I am horribly overexposing the shot. My procedure is to make a shutter-speed guess, then take a look at the review image and use the histogram to determine whether I got a good exposure -- there's really no better way than the histogram, you can't always trust your eyes when looking at that small screen.

[justme_dc]

Here is my first goof around with the false color stuff. You can clearly see the "hot spot"... I have no idea how to correct this. I've tried messing with the curves a bit and even selecting the perceived hot area and toning it down, but not with any real satisfying results so far. Maybe someone else knows how to get rid of the center hot spot?

If you are using Photoshop it's a piece of cake. I downloaded your image and was able to get rid of the hot spot in about 30 seconds. Here's what you gotta do....

layer/new/adjustment layer
set adjustment layer to brightness/contrast

adjust new layer until center of image is where you want it color and density wise.

use the gradient tool to make a radial gradient on the adjustment layer that is roughly the same size as the hot spot. double click adjustment layer to fine tune the B/C settings.

Stir well and serve over Ice with a festive garnish and Viola! You have the center hotspot fixed.

I have employed this trick a bunch of times with wide angle 4x5 negs. the light fall off on those can be horrific sometimes.

lemme know if this works for you. Good luck

[iwatkins]

I love the effect that filter gives. However, can anybody tell me if that is an actual infra-red image rather than just an effect.

Is it only letting through electromagnetic energy in the infra-red waveband ?

Do these filters come with a markings etc. that shows which part of the waveband is being passed ?

I ask, as although I'm not really interested for my own photography (but the effect is nice), I am interested for my day job.

Cheers

Ian

[mattchase]

These filters don't actually allow true IR to be exposed. The more accurate way to refer to this type of photography is Near Infra Red (NIR). The wavelength of each filter is a bit different, but the Hoya R72 blocks wavelengths below 720nm, while the Wratten 87 blocks below 800nm, and the Wratten 87c blocks below 850nm.

The resulting image is infra red like, but not quite the real thing compared to shooting IR film.

[iwatkins]

Thanks for that Matt. :)

Cheers

Ian

[Longwatcher]

mattchase wrote:
These filters don't actually allow true IR to be exposed. The more accurate way to refer to this type of photography is Near Infra Red (NIR). The wavelength of each filter is a bit different, but the Hoya R72 blocks wavelengths below 720nm, while the Wratten 87 blocks below 800nm, and the Wratten 87c blocks below 850nm.

The resulting image is infra red like, but not quite the real thing compared to shooting IR film.

Slight correction to the above: The Hoya R72 does peak blocking at 720nm, but it lets in some spectrum below that although not much (it is a sharp drop at the 720nm mark).

Second anything above 750nm is IR. There is Near-IR, Short-wave IR, Mid-Wave IR, and Long-wave IR. Mid and longwave are also called thermal IR. Somewhere in the mid-wave portion sensors stop picking up reflective IR and begin picking up emmisive (ie thermal) IR spectral information.

Best guess is the 10D stops at about the 780nm point due to the IR filter, The D60 at about 720nm.

Just being picky,

[mattchase]

Not picky, you're just more accurate! Thanks for the clarification, Tim.

[JMSetzler]

I would love to see some sample IR photos from the 10d if anyone has any to share.

John Setzler
Hickory, NC

[mattchase]

Hi John, follow these links.

http://www.matthewchase.com/forum/ir1a.jpg

http://www.matthewchase.com/forum/ir1c.jpg

http://www.matthewchase.com/forum/irfarm.jpg

http://www.matthewchase.com/forum/ir02.jpg

http://www.matthewchase.com/forum/ir01.jpg

http://www.matthewchase.com/forum/irb.jpg

[Longwatcher]

JMSetzler wrote:
I would love to see some sample IR photos from the 10d if anyone has any to share.


Here are a few, I have more, but they are in the adult portion of my site.

IR plant test
http://www.longwatcher.com/camera_test.html

IR model shot (red head female)
http://www.longwatcher.com/images/CRW_5338-edit.jpg

IR model shot (Black male)
http://www.longwatcher.com/images/CRW_5855-edit-IR.jpg

People are tricky, they have to hold still for at least 4 seconds, so not exactly candid in nature. also minor movement tends to cause some bluring, but not bad.

[dwbrant]

Ok I have an addition to the list (at the top of the thread):

good: Sigma 20mm f1.8

bad: Tamron 19-35mm f3.5-4.5