IR foreheads?

[Guest Walt]

Advertisement (gone after registration)

This topic was started quietly in the "Leica Response" thread (yes, thanks Guy and Leica), but I thought it deserved its own attention.

 

There is a striking similarity between the following three shots, the first from another forum member, the other two crops from my own photographs. The "patch" on the forehead is completely blocked and without recoverable information. In my two shots an inspection of the files shows that (forehead aside) only actual light sources within the frame were comparably blocked. The forehead area could not be recovered with an ACR "exposure" adjustment. Both photographs required a (not altogether satisfactory) burning or "patch tooling" in Photoshop.

 

As suggested in the earlier thread, I think this may be an IR problem: the scalp and brain both have a very high blood flow and I speculate that the camera has recorded emitted IR from the scalp and frontal cortex. In neither of my shots was there a significant amount of light shining on this area compared to other parts of the body or other objects in the scene.

 

As a BW photographer, I had imagined being free of the IR filter issue, but now I'm beginning to wonder. Any ideas or experience on this subject?

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!

[Advertisement]

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

[Irakly Shanidze]

hmmm... plunging their heads in a bucket of ice before the shoot would surely do the trick.

in reality it is much less feasible than just using a "recovery" tool in Lightroom or ACR

[Irakly Shanidze]

i also wonder if cognitive activity correlates with intensity of IR radiation of foreheads... in this case M8 may become an indispensible tool for con artists: potential victims can be selected by picking individuals with the darkest foreheads from a series of photos taken by a seemingly harmless street photographer.

[sdai]

This can't be a IR problem ... if you know how the infrared thermometer works, you should know the forehead surface temperature is pretty close to the tympanic temperature. So if Ruben's forehead is causing IR problems, then you should see highlight blown out around his ears too.

 

Anyone who has played enough with Nikon's compressed NEF knows this is a common highlight compression problem.

[BerndReini]

I don't think this is an IR problem. I always use IR filters and I have shots like this. It happens all the time. All the subjects have very high, hairless foreheads, and I would guess that this is caused by an overhead light behind them. This light is reflected at a direct angle directly into the lens and causes a specular highlight that is extremely overexposed. I am a cinematographer and have worked with a bald high profile celebrity, who prohibits the use of a backlight for this reason. The next time this show up in a picture, walk around the subject and change the angle of the light, and I bet the reflection will be gone. And this even though the subjects brain activity is still active ... hopefully.

[eronald]

Standard specular. Blows out most digitals. A back might be ok, or a true 16 bit Leica file, if they made one.

 

In practice you fix this by retouching, or preferably by makeup (powder) during photograpy.

 

Edmund

 

 

 

 

I don't think this is an IR problem. I always use IR filters and I have shots like this. It happens all the time. All the subjects have very high, hairless foreheads, and I would guess that this is caused by an overhead light behind them. This light is reflected at a direct angle directly into the lens and causes a specular highlight that is extremely overexposed. I am a cinematographer and have worked with a bald high profile celebrity, who prohibits the use of a backlight for this reason. The next time this show up in a picture, walk around the subject and change the angle of the light, and I bet the reflection will be gone. And this even though the subjects brain activity is still active ... hopefully.

[jaapv]

Advertisement (gone after registration)

Slide film does this as well.... Life used to be a lot easier on photographers way back when gentlemen and Chicago gangsters wore fedora hats.

[Vivek Iyer]

I thought I found something new about the M8 (http://www.leica-camera-user.com/digital-forum/19983-ultraviolet-sensitivity-m8.html)..

 

You seem to have captured the inner glow of these folks (Kirlian? )!

 

Far out!

[rosuna]

The first picture is my portrait. Looking at it you can understand why I have such success among women.

 

I have the forehead problem in many pictures.

 

I was thinking on it, looking for the reasons.

 

It can be a small area of extreme highlights out of the range of the camera, but I don't think so. The IR theory is a possibility, but I think that the 8-bit compression could have a role here. The tonal range in the highlights is heavily compressed. I can see some detail in the small area of the forehead. It is not fully burnt. The tonal variation has disappeared, that's all.

 

I feel very uncomfortable with the 8-bits compression. A non-linear compression is a great idea, but I would like to have some degrees of freedom in the tonal range. Leica says this compression doesn't affect the photographs in practice, but one can find situations in which it has some visible effects. Whether the A/D converter works in a 16-bit space, why losing too much information during the storing process? I would suggest a 10-bit non-linear encoding for providing the files with a safer margin. The question is, would it increase the size of the files too much? I don't think so. The 16 bits files of the DMR weight 20MB, the 8 bits files of the M8 weight 10MB. I am thinking in somethong in between, maybe 12 or 14MB. It is acceptable to me if I can keep additional tonal detail. Leica is suggesting the use of additional JPG files for a B&W preview, but a wider encoding is a much important thing.

[tom0511]

I too have experienced the "forehead problem" in many images (also when using ir-cut) and IMO its probably a question of the tonal/highlight transition of the M8.

I agree that 16bit might improve it. There is a difference in the highlight-behaviour of the d2x when using uncompressed raw vs compressed raw - I allways use uncompressed.

A softer shoulder of the tone curve in the highlights would be appreicated from my side.

 

The first picture is my portrait. Looking at it you can understand why I have such success among women.

 

I have the forehead problem in many pictures.

 

I was thinking on it, looking for the reasons.

 

It can be a small area of extreme highlights out of the range of the camera, but I don't think so. The IR theory is a possibility, but I think that the 8-bit compression could have a role here. The tonal range in the highlights is heavily compressed. I can see some detail in the small area of the forehead. It is not fully burnt. The tonal variation has disappeared, that's all.

 

I feel very uncomfortable with the 8-bits compression. A non-linear compression is a great idea, but I would like to have some degrees of freedom in the tonal range. Leica says this compression doesn't affect the photographs in practice, but one can find situations in which it has some visible effects. Whether the A/D converter works in a 16-bit space, why losing too much information during the storing process? I would suggest a 10-bit non-linear encoding for providing the files with a safer margin. The question is, would it increase the size of the files too much? I don't think so. The 16 bits files of the DMR weight 20MB, the 8 bits files of the M8 weight 10MB. I am thinking in somethong in between, maybe 12 or 14MB. It is acceptable to me if I can keep additional tonal detail. Leica is suggesting the use of additional JPG files for a B&W preview, but a wider encoding is a much important thing.

[wparsonsgisnet]

I have the same problem in several portraits. I agree with Edmund that powder before the session would have solved the problem.

 

The highlights on the face are blown out and need some work.

 

M8, Version.1.09 (original), iso 160, 1/250, f 5.6?, 486 flter in place, either 50 or 35 'lux asph,

processed with Edmund's 1.09 Linear LoSat.

 

I have reduced the exposure to -0.50 in C1.

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!

[rosuna]

You can do some tricks for reducing the contrast of the scene, but it is not a "solution" because you detect the problem after the picture is taken. I see the histogram and I cannot observe clippings, but the patch is there when I see the photograph in my computer. It happens with other types of pictures, but it is less obvious because you don't know for sure if the burnt highlights are caused by a limit in the dynamic range or by compression problems. When the problem appears in portraits with soft lights you realize something strange occurs.

[j. borger]

These blown highlights are also a negative side-effect of some of the latest Leica lensdesigns, which have too much macro-contrast, especially for B&W ........

With summer knocking on the door .. it's time to pick some lower contrast lenses

When it comes to pullin detail out of highlights ..... i have much more succes using C1 instead of ACR/ Lightroom!

[stunsworth]

These blown highlights may not be due to the 8 bit non-linear processing.

 

If an 8 bit value is blown - i.e. has a value of 255, doesn't that mean that the original 16 bit value was also blown? Isn't the 16-8 bit processing compressing the dynamic range, not reducing it by throwing away all the highest or lowest values? I could understand posterisation if the compression was severe, but not the introduction of blown highlights.

 

As usual I may be talking rubbish.

 

I have had the same effect on my Canon 5D.

[pfogle]

These blown highlights may not be due to the 8 bit non-linear processing.

 

If an 8 bit value is blown - i.e. has a value of 255, doesn't that mean that the original 16 bit value was also blown? Isn't the 16-8 bit processing compressing the dynamic range, not reducing it by throwing away all the highest or lowest values? I could understand posterisation if the compression was severe, but not the introduction of blown highlights.

 

As usual I may be talking rubbish.

 

I have had the same effect on my Canon 5D.

from what I gather (please, any experts, chip in here!) the compression down to 8 bits is done using a gamma curve to move more 'levels' into the shadow areas - in a purely linear response curve, the top one stop of exposure will have fully half the available levels, while the shadows are under-represented.

 

This is why there are people who advocate 'expose to the right' with the histogram - it puts the shadows further up the scale to make more levels available for lightening in post production.

 

However, if Leica applied a curve (which I believe to be the case) to push more levels into the shadows, that would make less of the 256 (total available with 8 bits) available for the highlights, resulting in less 'headroom' (no pun intended!) for post-production exposure correction - so I'd be expecting more blown out highlights with the Leica.

 

The fact is, most people don't report worse highlight response, so I really don't know... but it would be interesting to compare results from the M8 and another 12 bit camera and see what happens...

 

ps you can't see IR from the cortex - the skull gets in the way!

[rosuna]

An interesting comparison would be between M8 and DMR.

[Machaon]

Hi!!!

 

Leica should take care of its customers by hiring a dermatologist in order to solve some skinny issues.

The three men are rather bald. An excess of sebum can give a shining skin when exposed to light and it can sometimes explain hair loss. It is common in males because of hormons. But, it can happen in women who also producing male hormons in variable amount and may have greasy skins : just powder the skin or take a medication...

 

Nothing to do with IR!!! Just ask the doctors... some fora (plural for forum) can be dangerous to M8 owners’ mental health …

[johnll]

Can you remember if the camera histogram was up against the right edge? It might help to underexpose a tad, relative to what you have been doing. The highlights are surely near-specular reflections off the forehead/scalp.

Edit: 1 tad = 1/3 to 2/3 stops (:

[Advertisement]

Hallo,

Look here: Interesting Leica gear

[Guest guy_mancuso]

Pretty much just specular highlights, seen it on every camera. What happens is the angle of the light hitting that part of the forehead is directly reflecting back into the lens. Think of it as a mirror reflecting the light into the camera . Basically the same thing. Makeup , grow more hair helps. LOL

 

But just changing the light or the subject will help. I have been shooting portraits with the IR filters in place and things look pretty normal. But even with a softbox I will get this above the eyes on the forehead if the angle is wrong.

[wparsonsgisnet]

For the example I posted, I was using the Metz 43, with a diffuser snout, bounced off the ceiling (white). I got a number of these blown reflections at this shoot.

 

Again, I did not have this problem with the D2.

 

If it's the file compression, let's fix it, please.

 

Guy, since you have a DMR, have you had this problem with that back?

 

And, yes, I'll suggest to my subject that she grow some forehead hair.