Asking about white balance and our vision.


Status
Not open for further replies.

zoossh

Senior Member
Nov 29, 2005
8,725
0
36
Singapore
Got a question that bothers me, if anyone have read abt it or knows the answer, pls enlightened me.

The digital sensor captures things as they are, with manual inputs of white balance control from the user. On the other hand, our eyes automatically differentiate between light source and their wavelength, do autocorrection in our brain, thus allowing us to correct weird looking color cast that otherwise appear in our photos when the correct color balance is not input.

my question is that, if we see the fluorescent light and our mind corrects the green tinge, why is it that when a sensor capture a picture under fluorecent light without color balance correction input producing a picture with the green tinge, we can see the tinge instead and our mind does not autocorrect it?

is it becos the weird color casts that is reflected into our eyes is somehow at a different wavelength or with different properties when it is captured by the sensor and reproduced onto a print or a screen, which explains why the brain only corrects what the eyes sees directly and not what the eyes sees from the uncorrected print?
 

Got a question that bothers me, if anyone have read abt it or knows the answer, pls enlightened me.

The digital sensor captures things as they are, with manual inputs of white balance control from the user. On the other hand, our eyes automatically differentiate between light source and their wavelength, do autocorrection in our brain, thus allowing us to correct weird looking color cast that otherwise appear in our photos when the correct color balance is not input.

my question is that, if we see the fluorescent light and our mind corrects the green tinge, why is it that when a sensor capture a picture under fluorecent light without color balance correction input producing a picture with the green tinge, we can see the tinge instead and our mind does not autocorrect it?

is it becos the weird color casts that is reflected into our eyes is somehow at a different wavelength or with different properties when it is captured by the sensor and reproduced onto a print or a screen, which explains why the brain only corrects what the eyes sees directly and not what the eyes sees from the uncorrected print?
if im not wrong its not that our mind will auto correct the green tinge from flurescent, it is that our eyes are unable to see the green spectrum of light, the same as why we can't see UV rays.
 

if im not wrong its not that our mind will auto correct the green tinge from flurescent, it is that our eyes are unable to see the green spectrum of light, the same as why we can't see UV rays.

what about the other color casts? the fluorescent light is only an example. our eyes seem to correct even tungsten and candle light that can appear warmer in uncorrected photos.
 

if im not wrong its not that our mind will auto correct the green tinge from flurescent, it is that our eyes are unable to see the green spectrum of light, the same as why we can't see UV rays.

Hahaha, I wonder how we see green then? :)

Our eyes can perceive the green wavelengths, and are in fact more sensitive to them. CCD and CMOS sensors typically have 2 times more green pixels than red or blue for this reason, IIRC. So that the sensors pick up more green than red or blue, just like our eyes.
 

I think our eye will try it's best to correct any lighting cast, similar to AWB in cameras... It's just like our eyes can adjust to brightness of a dynamic range of 10 stops~ The eye is simply amazing~;)
 

Hahaha, I wonder how we see green then? :)

Our eyes can perceive the green wavelengths, and are in fact more sensitive to them. CCD and CMOS sensors typically have 2 times more green pixels than red or blue for this reason, IIRC. So that the sensors pick up more green than red or blue, just like our eyes.

what i mean is that when and how does the brain make adjustment to the vision of the colors? we know the brain can make adjustment to visualisation of tonal intensity and balances it off, how about colors?
 

I think our eye will try it's best to correct any lighting cast, similar to AWB in cameras... It's just like our eyes can adjust to brightness of a dynamic range of 10 stops~ The eye is simply amazing~;)

actually that is my question. if in a room that is filled with artificial light causing a weird color cast, let's say blue, our eyes may still see correct white balance. however if the weird color cast is presented as it is, as blue, on a printed photograph, we dun see white and we see blue, telling us that the color cast is not corrected. then why is there such a difference, why dun our eyes/brain also do the same correction as when we are in the room directly?
 

what i mean is that when and how does the brain make adjustment to the vision of the colors? we know the brain can make adjustment to visualisation of tonal intensity and balances it off, how about colors?

Ambient light.

Of course all this is postulation - but do you realise that when you're outdoors at night tungsten lights look ridiculously warm as well? But once you're indoors everything is fine?

Someone once mentioned in one of the photographic books I read that tweaks in Photoshop should be always done in an overboard manner first, instead of step by step. The reason given was that given enough time, the eye will also auto-correct the colour cast and you'd end up with horribly wrong WB and colours in the end as well.
 

Ambient light.

Of course all this is postulation - but do you realise that when you're outdoors at night tungsten lights look ridiculously warm as well? But once you're indoors everything is fine?

i know you are talking about is the mixture of light.

assuming that we can see fluorescent light as white, if you are indoor with fluorescent light overempowering a small tungsten light bulb, it will look fine. the same will go for a small tungsten light in bright day light outside. but in a pitch dark indoor and pitch dark outdoor, there won't be difference of the tungsten light in both situations. this is about our eyes.

i'm not sure if u r talking about digital capture though. i dun really shoot night nor dark indoors.... so not familiar with these lighting conditions other than outdoor ambient light which i do often in travelling.
 

Firstly, human vision is an amazing thing, only is it optically capable of amazing things (great low light performance for example, extreme exposure latitude), there is also a good wack of high performance post capture signal processing done.

As for 'white balance' correction. Well ambient light as some one said. When you are in said fluorescent lit room, the ENTIRE scene is lit by this 'non solar' colour light - so with out even thinking, some sort of signal processing happens in your brain to correct for it.

On the other hand, looking at a small sample of that scene later, you won't have the entire environment flooded in that 'non solar' light, so you brain doesn't correct for the colour cast on that little print in your hand or displayed on your PC screen.
You also have to take into account that our camera sensors, be it electronic or chemical have very different response properties that our eyes. CCD's for example are sensitive to Infrared. My digital camera can see the remote control off my TV clearly. I can't see a thing myself.


Might be interesting to attempt to find charts on the colour spectrums of various types of lamps. Fluorescents in particular come in also sorts of colours and spectrums as it's easy to adjust the phosphor coatings in the tubes to get the effect you want.
If you look through lighting catalogues, you will notice special tubes are made for commercial use where colour matching is going on - these tubes have more complex phosphor coatings than normal tubes to attempt to emulate more closely the colour of 'natural' sun light. (Or more importantly, try to have a wider spread of wavelengths being emitted - normal tubes tend to be harsh due to only emitting a very small number of different wavelengths - they are cheaper to make that way).


A former boss of mine had a vision defect that meant she was nearly blinded by normal 'cool white' commercial fluorescent lighting, but was perfectly comfortable with natural sunlight or incandesant lamps. We had to change the lights in her office so she could work comfortably.
 

not just ambient light, but the monitor is also adjusted to (usually) a daylight white balance... so any image projected from it, when compared to the other parts of the screen even in a dark room would look greenish if there is the fluorescent greenish effect captured...

actually, you can train your eyes to notice the greenishness of fluorescent... I know I can see it and tell the diff between greenish fluorescent and the non-greenish type of fluorescent light...
 

Well perhaps the eye is a incident and reflection light meter at the same time. It is able to do a custom white balance based on its memory of subjects which are naturally white.
 

Well perhaps the eye is a incident and reflection light meter at the same time. It is able to do a custom white balance based on its memory of subjects which are naturally white.

yes, i think it does spot metering, and possibly correct different locations separately which is not possible in digital sensors.
 

it probably has a lot to do with environment and if our brain can identify things it knows the colour. If you are in a white room that is lit by red light only, then our brain is not able to do proper WB, the white walls are still red.... either the red is too strong or there are not enough reference objects around to adjust WB. However if you use red-tinted shades the tint will disappear over time and when you take off the glasses every thing got a cyan cast....
its a process that sometimes goes fast sometimes slower and is very unpredictable....
 

not very scientific but er...

things look yellowish under tungsten bulbs cos they're lit by yellow light. but the object is still red or blue or whatever, so the eye can adjust.

a printed photo with red colour cast however is red because the pigments/inks used to produce it are red. the eye won't white balance that, since the object is red, rather than the light falling on it. otherwise red apples and green apples will all look grey already right?
 

Got a question that bothers me, if anyone have read abt it or knows the answer, pls enlightened me.

The digital sensor captures things as they are, with manual inputs of white balance control from the user. On the other hand, our eyes automatically differentiate between light source and their wavelength, do autocorrection in our brain, thus allowing us to correct weird looking color cast that otherwise appear in our photos when the correct color balance is not input.

my question is that, if we see the fluorescent light and our mind corrects the green tinge, why is it that when a sensor capture a picture under fluorecent light without color balance correction input producing a picture with the green tinge, we can see the tinge instead and our mind does not autocorrect it?

is it becos the weird color casts that is reflected into our eyes is somehow at a different wavelength or with different properties when it is captured by the sensor and reproduced onto a print or a screen, which explains why the brain only corrects what the eyes sees directly and not what the eyes sees from the uncorrected print?

Simple.. Everything is relative and you get a colour reference from the things you see around you. That's why it becomes very important that the WB you use to shoot the images is the same WB you use for viewing the images.

For example, the print of a shot under fluorescence light, if viewed under fluorescent light will be more green than a neutral object under the same light. If viewed in daylight, then it will be just the fluorescent cast.

A good way to understand this is during a mixed lighting situation, then your eye won't know which one to take reference from. eg, if daylight is coming in from window and the fluorescent light is still on, then you'll be able to see the green cast from the objects illuminated by the fluorescent light but not by the daylight.

That's why my fluorescent tubes are all the true colour type without the green cast. I can even use it to shoot using Cloudy WB and the colour will still be quite correct. They cost a bit more though.. ;p
 

Status
Not open for further replies.