Why red is always problematic to the camera sensor?
can anybody explain to me why the modern DSLR cameras always have problems with preserving details of red colored subjects?
I shoot a lot of live shows, with some of the scenes lit entirely with deep red light (monochromatic) and what I noticed is that for the camera (either D4 or D800) it is always more difficult to find precise focus and the amount of details preserved is always lower than in the images lit with some other colorful, but monochromatic lights. I believe blue and green and actually the easiest for the sensor and the camera to work with, but red and orange seem to be the most problematic.
Your replies will be appreciated.
#1. "RE: Why red is always problematic to the camera sensor?" | In response to Reply # 0elec164 Nikonian since 15th Jan 2009Fri 05-Oct-12 06:08 PM | edited Fri 05-Oct-12 06:11 PM by elec164
This would be tough question to answer for the average person. On a Yahoo group I belong to there was a discusion that sort of touched upon this topic.
If I remember correctly, it has to do with the spectral properties of silicon. As it turns out, silicon makes a good spectral filter. In fact it's so good it's the basis upon the Foveon X3 design. If you looked at the Wiki site I linked to, you'll notice that the blue spectrum gets absorbed rather quickly, green penetrates a bit deeper and red spectrum penetrates the deepest.
Also if you look at this Kodak PDF and scroll down to page 15, you'll see a graph that also suggest that silicon is less absorptive for red spectrum.
I infer from this information that less red photons are absorbed by the silicon and more are available to be detected by the sensing layer than with blue or green.
At least that is my amateur reasoning and understanding. Hopefully there is a Nikonian member that is more knowledgeable that can either confirm or correct me.
Visit my Nikonians gallery.
#2. "RE: Why red is always problematic to the camera sensor?" | In response to Reply # 0Noel Holland Charter MemberTue 09-Oct-12 11:47 AM
This is a very interesting question and I consider it deserves more attention than it has so far. So in order to encourage some thoughts among others I going to throw my own ideas out. I'm not saying that they are the reasons but more the areas where I think the explanation might lie. Hopefully someone else can take this post further and provide a more definitive resolution.
For starters, normally the big bugbear in photography is purple. That's because it is a blend of blue and red but no green. The extremes of the spectrum but nothing in the middle. I find it interesting that you are having trouble with red light and specifically red filtered artificial lighting.
From what you are describing there are two issues. The first is exposure range and the second is focus.
1) Relative exposure judgments between the cameras spectral response to red light and the human response to red light
We normally consider that humans see in RGB but this is actually incorrect. We actually see in LMS denoting Long, Medium and Short light waves. None precisely match Red, Green or Blue but the Red is the largest difference. Our cones are more closely attuned to a sort of green/orange shade. That is why when you view light viewed through RGB filters the red one often seems the darker shade even though the light behind the filters is of the same intensity.
As a result, what we see is not what the camera sees, and our ability to judge the required dynamic range for subjects under red light is often quite different as it would be for blue or green light sources.
Note also that typically the bayer filter on cameras has twice as many green filters as red or blue. This helps explain why you find green lit subjects to be better then red subjects. The blue light dynamic range may also be helped by the fact that the difference between Short cones (Blue) and Medium cones (Green) is much wider than than between Medium and Long cones spectral sensitivity resulting in a greater ability for us to differentiate and comprehend the dynamic range.
2) Focus tracking
DSLR do not use the main sensor for setting focus but a seperate sensor in the base of the camera. The main mirror is actually partially transparent and there is a tiny mirror at the back of the main mirror which flips the light downwards towards the autofocus system. I don't know the spectral of the focus sensor array but it might be that the focus sensor is responding to light you can't see - here may be a high amount of IR light in the red lights which is throwing the focus sensors off. Or it might be that the af sensor is has lower response to red light than to other frequencies.
Visit my Nikonians gallery.