[Pavel]

Dear friends, Hearing your recommendations to abandon (postpone or consider options) fisheye or telephoto and think instead of IR photography got me thinking and experimenting a bit. NIK software offers a customizable IR filter. I experimented with it on several of my photos and I posted a photo where I blended BW with faw IR on flickr.

Now here is my question: As far as I could tell, the filter added some grain (you can add more or remove the grain effect) and increased sensitivity wherever image showed red, yellow and green and reduced sensitivity wherever there was blue or purple or magenta. I recognize that IR operates at a different wavelength than red, yellow and green, but i wonder whether the actual effect of using IR creates an image very similar to that I observed with NIK. In the end, IR radiation has to be mapped into visible spectrum for us to see it. If the sensor maps it into red yellow and green, than using software could emulate the effect fairly well I think. I wonder whether those of you that are experienced with IR ever tried to reproduce the effect with regular camera & filter. Also if you happen to have an identical image taken both as IR and color, I would be grateful if you could make a low res small version available to me for comparison. I will play with it in NIK a bit and I will be curious if we could see a difference.

I will be grateful for any comments. Basically, I am thinking that the easiest way to do IR is in software, if you can reproduce that aspect of IR, which makes it so special. Forgive me if this is a rant. I am not against IR, I just do not know enough about it and i wish to learn.

Pavel

[Kombisaurus]

Hi Pavel,

You're correct that in order for us to see IR photos, that part of the spectrum must be mapped to visible light.. but that doesn't mean there is a correlation between visible colours and IR colours any more than there is between two different colours in the visible spectrum (ie Green and Blue). So any software IR emulation is really just based on guesswork to provide a similar look to IR. A real IR photo might look completely different.
The reason you see green tree foliage appear as white in IR photos is not because the colour green reflects more IR light, it is because tree foliage reflects IR light. You could park a green car beside a green tree and they may appear identical in a faux emulated IR photo, but completely different colours in a true IR photo.
I remember taking some test IR photos around the house and was very surprised to see that a black jumper I was wearing appeared completely white in the IR photo, while most other black things in the room appeared black. It seems the jumper absorbed visible light yet reflected all IR light.
The emulation software generally uses an algorythym based on how the most common materials seen in IR photography would appear (ie blue skies and water, and green plants, etc). Because as humans we don't really know what colour many other materials are supposed to be mapped to, we just accept them to be correct.

But, does it actually matter if an IR photograph is "true" or not? The reason most people shoot IR photos is not out of scientific interest in IR light, but simply because it allows for some beautiful photographs. If you are happy with the photographs from the software, then that's all that matters in the end.

Personally I prefer the flexibility of using a channel mixer in weird and creative ways instead of a dedicated IR plugin when I'm after a particular aesthetic look. When I shoot IR (which is rare) it's generally because I am actually curious about how the world looks in the IR spectrum, so the emulation is no good.

[Pavel]

Thank you Adrian. I noticed that the color mapping of the IR filter in NIK is more sophisticated than I described. I was surprised for example, that the green of the pine trees remained dark, while the green leaves of the tree right next to it went white. On big stems of grass, there is a fairly sharp transition from almost white (zone IX my guess) derived from sunlit green (about Zone VII to VIII i am guessing) to dark (perhaps zone II) from less well lit grass (my guess zone V).

Adrian, I tried to check out your site, but it does not seem to be populated with photos or at least i could not figure out how to access the photos.

Thank you for writing this very informative response to my question. I think I understand a lot better. I think that for now, I will experiment with the software filter and leave the hardware purchase for some later date. I am swayed by the low cost (no cost actually - I do have the software) and the convenience of capturing the images (no need to carry extra camera body, no need for a tripod most of the time, no need to put on and take of the filter each time i wish to refocus/recompose, multiple use of the same image - color, straight BW, faw IR.). I will be interested to hear why my decision does/does not make sense.

Pavel

[enviroguy]

OK, Pavel, we have had some very interesting technical discussions regarding IR photography. I have seen your "faux"?*! IR image on your Flickr site and believe that it is dramatic to a suitable degree with very nice tones and effective composition. Perhaps you should display it here on Shuttertalk for all to see easily what your creation is all about.
With all the possibilities of the Nik IR conversions, together with the many other useful and fun filters within Nik, you should have thousands of hours of fun and experimention without resorting to any hardware conversions or filters. I agree that is the way to go, but I know that you're never satisfied.
BTW, thanks for introducing me to the world of Nik. My mind is exploding with the possibilities.

Regards.....Dennis

[Toad]

I am swayed by the low cost (no cost actually - I do have the software) and the convenience of capturing the images (no need to carry extra camera body, no need for a tripod most of the time, no need to put on and take of the filter each time i wish to refocus/recompose, multiple use of the same image ...

Interesting discussion, and explanation from Adrian. Having suggested IR, I now feel some obligation to comment

Like Adrian, I am not convinced that software filters will have the same effective as real IR - but because IR photography is largely serendipity anyway, I am not sure that is an issue. I have looked at quite a bit of IR photography, and I think that I can spot a real IR image from a software created image immediately. I can't put my finger on the exact reason, but photos done with real IR have a surreal quality to them that I have not seen captured with software filters. The difference in light wavelengths and spectrum is visible - to me at least.

All the arguments presented for using software instead of a dedicated camera body are good ones.

However...

Digital cameras typically have taken IR photos by putting an R72 or equivalent filter on the front of the lens which blocks visible light from reaching the camera sensor but allows IR wavelengths to pass through. Because the sensor's built-in IR filter is still in place, this forces you to use very long exposures on a tripod and does not allow you to compose using your view finder because visible light does not reach the camera's mirror.

Some modern IR conversions of camera bodies do it differently. This type of conversions removes the sensor's IR blocking filter and replaces it with a clear / IR sensitive filter (R72 equivalent). You need to replace the removed filter because just removing the IR filter without replacing it will make the camera "short sighted". Because the R72 (opaque to visible light) is behind the mirror, and because the camera does not need to overcome the usual IR blocking of the removed filter, this method allows you to compose using your viewfinder and take IR photos at near normal shutter speeds - reducing the need for tripods and very long exposures. This seems like the way to go IMO.

I have been in the market for an inexpensive Nikon camera body to convert - something like a D70 would do very nicely. I have also been looking at this conversion service to do the work. I may just buy their Standard IR filter and do the work myself.

Anyway - interesting discussion.

[EnglishBob]

I've played with software filters, and while they can give some nice looking images, they don't have the qualities I like in IR images.

Metals tend to show dark in IR, plants very light, usually white, a light blue sky will be jet black a lot of the time.

[Irma]

Following closely this thread as I like very much IR images...

Actually my love for this kind of photography started when I saw Adam's IR pictures, he has beautiful pictures, but I have never done anything myself.

When I wanted to convert my panasonic to take IR pictures, I couldn't find anyone I could trust to do the job.

Pavel, I think in this site you can see a picture taken in color and how to see it in IR... I understood this was something you wanted to see?

http://www.lifepixel.com/digital-infrared/samples.html

Anyway the site has very interesting information and even how you can convert your camera yourself, with pictures and instructions for different cameras.

http://www.lifepixel.com/tutorials/

[matthew]

Since digital photography is a mathematical exercise anyway, there's nobody to say what it should or shouldn't look like. There's lots of different things that digital photographs can be made to look like - polaroid, tri-x, and kodachrome come to mind. I see the 'IR look' as another option.

Adrian has already used a great example of when IR photography will look different from an IR conversion, and in an urban setting it will be even less predictable. Black fleece and black jeans will look very different. A tungsten bulb in a a room that's lit by fluorescents will stand out like a spotlight, but different colours of light won't matter at all. Rust is the same colour as steel, and some paints show different tones while others won't. There's a sign near my work, painted in red and grey, with a sticker on it that has faded into oblivion. To see it, it's an old painted sign that says "Post No Bills". To see it in IR, it's a blank piece of wood with a sticker on it.

Also if you happen to have an identical image taken both as IR and color, I would be grateful if you could make a low res small version available to me for comparison. I will play with it in NIK a bit and I will be curious if we could see a difference.

I'll see if I can shoot something this coming week. There's not a lot of foliage nearby, and sun is a hard thing to find as well, but I'll pack a pair of P&S cameras just in case. One's a Canon SX20 that shoots in colour, and the other is my Panny FZ18 that's had its IR-cut filter removed. I've added an 87C (B+W 093) cut filter to it, so it's not letting any visible light in these days.

[Pavel]

Thank you all for your help. For me it is not all that important that the IR software filter reproduces real IR hardware setup. I would like to experiment with the "IR effect" I see IR photos as very contrasty and grainy. The grainy part i do not like but i can simply dial out when I use NIK faux filter and the contrast is something I can generally control up or down, within reason. I find that blending with BW seems to give me desired tonality. I have no doubt that experts will immediately recognize my IR as fake, but I am not making any claims and for me IR is just another intriquing tool which is hard to predict but which can produce interesting, high impact results. I would like to understand it better, and to use the software version more often, but I am not planing to buy the hardware for the real thing, as much as i admire some truly exceptional IR photos. I do love toys and i go through periods of temporary irrational acquisitiveness, but fundamentaly I would just like to learn to be a decent photographer and the truth is, I do not need more equipment, just more knowledge and experience (I think).

Dennis, you are right of course and you helped me come to my senses - thanks.

Rob, thanks for making me think of IR. It may not have the consequences you envisioned, but I will pay a great deal more attention to the opportunity to use faux IR in post-processing, perhaps in combination with straight BW. I also appreciate a very lucid intro to IR photography.

Craig, I am no longer under illusion that I can emulate reliably IR in software. If it was so easy, everybody would be doing it, I suppose.

Irma, I read with interest the links you sent me. Unfortunately, they do not compare IR filter with a faux filter, but that is probably too much to hope for. I very much appreciate your help.

Matthew, it is generous of you to offer to provide a comparison photos. I would like to see the degree to which NIK can emulate real IR and how much trouble it is to get something that "feels" similar. If you provide me with low res photos, I will e-mail the resulting images back to you. I would like to retain a copy of the IR and faux IR images as a learning tool, but if it is not OK with you, I will look at the result, send you the results and delete copies my computer. If you let me retain copies, I will not publicly display or distribute them but I may show them on my monitor at home to visiting photographers, acknowledging the source.

[matthew]

Considering the interest in this thread, I think it would be great to be able to share the results more widely. It would be great if you'd be okay with posting the results of your processing here, or in a similar thread, and we can get our heads together for some collective wisdom. Perhaps a watermark of "photos by / processing by" wouldn't be a bad idea, just in case they escape into the wild and someone wants to offer us fabulous jobs as guest lecturers somewhere, but I'm not otherwise worried about attribution or public/private display. I do appreciate the compliment and consideration, though.

(To paraphrase one of my favourite What The Duck cartoons, the best way I've found to stop people from stealing my photos is to keep taking the kind of photos that I do.)

[Toad]

Good luck to everyone...I sincerely want to see the results of these experiments...I will add my efforts when I get my conversion done..I hope this will be a fruitful thread..

[EnglishBob]

This thread has me thinking of getting my XTi converted to full time IR..... Hmmmm.

[Irma]

I am looking at prices too... The Hoya I need for my landscape pictures (77mm) is around 150euro if I get good price...

I don't know how much it would be to convert my D70, and to get someone you really trust....:/

A fast solution and not so expensive would be to get a filter for my 50mm, it is around 40 euro.

Looking forward to seeing your pictures....

[Pavel]

OK, I will process some photos I get from you guys in NIK and pass them back to you. You can than decide whether or not to post the results and how. I feel more comfortable that way with other people's photos. A real discussion - I like that.

[matthew]

I've put up a gallery where some full-sized photos can be downloaded by anyone who's interested. The link is in this thread.

Now, these aren't the best quality photos - not even close - but they are a rough idea of what the colour scene looks like, and how it's seen by an IR camera.

The image quality of a compact camera can't match even an older SLR, but a camera with an LCD to view the images and contrast-detect autofocus solves a lot of problems that converted dSLRs have. They might be worth investigating for anyone who's dedicated but not expecting IR to be their passion.

[adam]

I have an infrared converted G2 and love it very much.
I used to bring it with me to uni, and my friends much rather played with the IR G2 than my nonIR G2 - because it gave them nicer skin

[Zig]

Seems like yous guys have covered all bases here, and I remember Matthew's successes when he got his camera "IR"'d.
My tuppenceworth: faux-IR are close but never as successful as IR achieved by a conversion of a DSLR, and even DSLR-converted IR does not quite have the nuances that IR film has/had. From memory, I used to use both Kodak and Konica IR 35mm and the results varied considerably between the pair of them: they were skewed to a different wavelength from each other and had differing ways of coping/not coping with reciprocity; varying the amount of pushing and pulling the ISO.
I'd reckon one could get darn close, though, with a converted DSLR and pp.
Of the plugs I've seen for "IR-type", Nik's and AutoFX seem the most successful, though the tweaks amount to what anyone can achieve with a bit of playing about..but I'd reckon it's well worth getting a Panasonic/G2/Canon 350D for a song then getting it to the vets for a snip!