What is "real" IR-colours..?

[Rune]

Advertisement (gone after registration)

Hi!

 

I am intrigued by the possibilities to take IR-photos with the M8(.2), and I have been trying out different approaches… and been thinking…

The enclosed pictures are made with Custom White Balance of a white sheet of paper in the sun with the IR filter mounted. The IR filter is an original filter from Leica, seems old by the lettering, but no other indication of what it is. Looking through it against a bright light, it is almost completely black, but with a very dark red tint which makes me believe it may be in the region around 700-720 nm, like a R72 maybe.

(By the way, it is Ø43mm, and the lens I used is a 50mm Summilux from 1971)

 

My big philosophic “problem” is, how can we know what “real” IR colours look like? I mean, it is (mainly..) outside what is visible for our eyes… is there any flaws with my thinking/approaching, using Custom WB?

I have another very cheap IR filter I bought on eBay, I think it says 950 nm, and the colours with this Custom WB approach is completely different, very magenta/purple/pink (and unsharp… so it may not be the best quality..)

 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

 

Any thoughts on this subject is much appreciated!

 

Best regards,

Rune

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[Michael Geschlecht]

Hello Rune,

 

Welcome to the Forum.

 

Actually, unlike some other creatures, most people cannot "see" Infrared Colors. Infrared colors are out of the spectrum that the cones in people's eyes are sensitive to. Altho some individual people can see some wave lengths somewhat further away from the recognized visible spectrum than most other people can see.

 

That is why it is said that the Infrared portion of the spectrum is "invisible" to human eyes.

 

In the past there was E3 processable "false color" infrared film where the infrared sensitive layer of film was assigned a color visible to people & the other color layers of the film were also assigned dyes visible to people in order to use the whole spectrum of colors which are visible to most people. But this "false color" infrared film did NOT capture & display in a visible form the actual wave lengths of the Infrared image that it was capturing. It captured a not visible infrared image that it displayed in a color visible to people. And displayed all of the visible to people colors with the remaining colors. That is 1 of the reasons that it was called "false color" infrared.

 

Best Regards,

 

Michael

Edited May 29, 2018 by Michael Geschlecht

[Rune]

Hi again!

 

I have been investigating, reading and testing a bit more, and I'm well on my way to conclude that there is no such thing as "real" IR colours... on the contrary in fact, most is referring to IR as "false" colours...

Most seems to think/agree that the green channel, i.e. leaf etc should be close to white with IR filter, so therefore the Custom White Balance should be set pointing at green grass or similar instead of a white sheet of paper. Other say that White Balance does not matter at all (which is not my experience so far…)

So, with Custom White Balance set to “green grass”, I get this as an example:

 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

 

Then it seems like many people like, or find it more “IR-like”, to swap the red and blue channel, and then the result is like this:

 

 

The green channel seems more or less the same, close to white in both cases.

I am not sure what I think at the moment, some of the fun with IR are the surrealistic colours, like in “The Simpsons”, and with the red and blue channel swapped, it seems more like a regular winter landscape here in Norway…

What do you think? :-)

 

Best regards,

Rune

Edited May 30, 2018 by Rune

[Martin B]

Hi again!

 

I have been investigating, reading and testing a bit more, and I'm well on my way to conclude that there is no such thing as "real" IR colours... on the contrary in fact, most is referring to IR as "false" colours...

Most seems to think/agree that the green channel, i.e. leaf etc should be close to white with IR filter, so therefore the Custom White Balance should be set pointing at green grass or similar instead of a white sheet of paper. Other say that White Balance does not matter at all (which is not my experience so far…)

So, with Custom White Balance set to “green grass”, I get this as an example:

 

_L1011420_LowRes.jpg

 

Then it seems like many people like, or find it more “IR-like”, to swap the red and blue channel, and then the result is like this:

 

_L1011419_colours_swapped_LowRes.jpg

 

The green channel seems more or less the same, close to white in both cases.

I am not sure what I think at the moment, some of the fun with IR are the surrealistic colours, like in “The Simpsons”, and with the red and blue channel swapped, it seems more like a regular winter landscape here in Norway…

What do you think? :-)

 

Best regards,

Rune

 

If you want more colorful IR photos, you need a filter below 720 nm. Check out this line about different IR filters:

 

https://kolarivision.com/articles/choosing-a-filter/

 

The IR colors derive from regular light between 400-700nm sneaking through the IR filter which leads to the false colors after channel swapping. There is no "real" IR color. IR photos without any regular light interference are B&W after conversion. 

[Rune]

If you want more colorful IR photos, you need a filter below 720 nm. Check out this line about different IR filters:

 

https://kolarivision.com/articles/choosing-a-filter/

 

The IR colors derive from regular light between 400-700nm sneaking through the IR filter which leads to the false colors after channel swapping. There is no "real" IR color. IR photos without any regular light interference are B&W after conversion. 

 

Thank you Martin B, that was very helpful!

 

So I guess it is correct to conclude that real IR colours do not excist, and if you go far off into the spectrum, i.e. 3000nm, it will all just be light, shades between black and white... after all, the part of the electromagnetic spectrum, 400-700nm, that we perceive as "colours", means on the other hand that everything outside this range does not have colours, at least not that can make sense for us... some animals are supposed to use part of the UV range...

 

So, IR photos then... there is no "right" or "wrong", just artistic interpretations... the more "psychic", the more "cool"...  it gives you more freedom, but in a way a lot less interesting to take IR photos also...

[Martin B]

Thank you Martin B, that was very helpful!

 

So I guess it is correct to conclude that real IR colours do not excist, and if you go far off into the spectrum, i.e. 3000nm, it will all just be light, shades between black and white... after all, the part of the electromagnetic spectrum, 400-700nm, that we perceive as "colours", means on the other hand that everything outside this range does not have colours, at least not that can make sense for us... some animals are supposed to use part of the UV range...

 

So, IR photos then... there is no "right" or "wrong", just artistic interpretations... the more "psychic", the more "cool"...  it gives you more freedom, but in a way a lot less interesting to take IR photos also...

 

You can't go with a photo digital sensor above 1600 nm for reflected IR light (btw, night vision which registers heat IR is at 30 000 nm - big difference!). You got the interpretation right - what we perceive as color, is only in the 400-700 nm range (insects perceive color very different than us since they can register also UV and IR light, but we can only suspect how they "see" it). 

 

It is very subjective what kind of IR look in photos we want to show. I personally prefer mostly the B&W look, but sometimes some false color image makes the better cut, too. I normally recommend using a 720 nm cutoff IR filter. 

[Rune]

Advertisement (gone after registration)

You can't go with a photo digital sensor above 1600 nm for reflected IR light (btw, night vision which registers heat IR is at 30 000 nm - big difference!). You got the interpretation right - what we perceive as color, is only in the 400-700 nm range (insects perceive color very different than us since they can register also UV and IR light, but we can only suspect how they "see" it). 

 

It is very subjective what kind of IR look in photos we want to show. I personally prefer mostly the B&W look, but sometimes some false color image makes the better cut, too. I normally recommend using a 720 nm cutoff IR filter. 

 

Thanks again Martin!

 

I was searching for real IR colour, and now have to settle with false color image..! Talk about ruining my day..!  

 

At least now I have the confidence to think that there is not right or wrong, just different aesthetic presentations... I also have to look more into B&W conversion... like a red filter on hormons... 

 

Btw... why do people seem to believe that swapping red and blue channel give a more "correct" IR photo... I don't get that part....

[Martin B]

 

 

Btw... why do people seem to believe that swapping red and blue channel give a more "correct" IR photo... I don't get that part....

 

This is just perception - when you swap red and blue color channels of the original IR image, the sky will look blueish which is most appealing to our eye. But if you like, you can change it in PP to any color you like (I never tried it, I always stick to blue sky). It is certainly not more correct - correct is just the registered IR reflection from objects, for example leaves which appear white in IR with white balance set to green grass. Molecules within the plant cells reflect IR well, that's why the white balance is often set to green plant/grass to make it appear white. If you take photos of metal statues in IR light, you need to custom set your white balance differently if no grass/foliage is present. 

[pico]

Rune you got some very good info. Thanks, Martin.

One tip regarding sharpness/focus. It is best to slightly near-focus with IR. On a lot of lenses there is a red dot, or 'r' on the distance scale where one would focus for infinity. You can focus at near subject and move the mark to the distance of original focus.

 

(This adjustment is especially important for our 35mm format cameras due to their narrow depth-of-focus, and less important as format size increases.)

[adan]

What Kodak did with its "false color" Ektachrome IR films was to swap all three color "channels" - or the film equivalent, the color layer chemistry. Using the traditional dyes (cyan, magenta and yellow) but combined with sensitivity to IR, Red and Green (blue was supposed to be blocked with a yellow filter on the lens).

 

Here's my 2009 explanation of how to replicate this intense "Ektachrome IR" false color by 1) shooting two M8 pictures, one filtered for normal color (IR/UV cut filter in place; or with less precision, no filter at all) and one with a dark red "IR pass" filter to capture mostly only IR. And then swapping the channels between the two images to replicate what Kodak's chemists did with the real film to change the sensitivities of the film layers vs. the dyes they produced (cyan skies, red foliage, misc other false colors).

 

https://www.l-camera-forum.com/topic/84753-getting-ektachrome-infrared-colors-out-of-the-m8/

Edited June 23, 2018 by adan

[helged]

What Kodak did with its "false color" Ektachrome IR films was to swap all three color "channels" - or the film equivalent, the color layer chemistry. Using the traditional dyes (cyan, magenta and yellow) but combined with sensitivity to IR, Red and Green (blue was supposed to be blocked with a yellow filter on the lens).

 

Here's my 2009 explanation of how to replicate this intense "Ektachrome IR" false color by 1) shooting two M8 pictures, one filtered for normal color (IR/UV cut filter in place; or with less precision, no filter at all) and one with a dark red "IR pass" filter to capture mostly only IR. And then swapping the channels between the two images to replicate what Kodak's chemists did with the real film to change the sensitivities of the film layers vs. the dyes they produced (cyan skies, red foliage, misc other false colors).

 

https://www.l-camera-forum.com/topic/84753-getting-ektachrome-infrared-colors-out-of-the-m8/

Thanks Adan - for this and plenty more insights on the forum!

[pico]

IR makes my stomach hurt unless it reveals what I cannot ordinarily see. WTF good is color IR? Someone inform me.

[adan]

If you ordinarily see red grass or pink trees or green skies - just how much of that other "grass" have you been imbibing?

 

More seriously, yes, infrared color (or any other photographic manipulation - using color filters in B&W, shooting B&W infrared, toning B&W prints, cross-processed color, multiple exposures, etc. etc. etc. etc.) can just be a parlor trick, if it only shows a different surface, but not a significantly different atmosphere/meaning/experience. A triumph of "process" over artistic vision.

 

That doesn't mean that the right photographer, with the right vision and the right subject, can't use IR color, or any of those other tools, to make a subject sing in a whole different choir.

 

Jeez, there's an artist who exhibits just down the street from my gallery who makes art out of snail trails: https://www.maiwyn.com/artists-summary/lauri-lynnxe-murphy

 

Makes my eyes hurt, too - sometimes. But not always.

Edited June 23, 2018 by adan