1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

I Perceived Color of objects: in White light from a TV

  1. Mar 2, 2016 #1
    Background: Normal white Sun light has a continuous spectrum in the whole of visible range. But, white light (rather what we perceive as white) coming out of a Monitor/TV is have only RGB in it and
    it looks white because of the Tricolor vision which excites all three types of cells in the eye.

    If we take a yellow object (say banana) to Sun light, it will look yellow because it reflects yellow light (wavelength) and this yellow wavelength light excites both Red and Green cells in the eye.

    Now, if we take the same Banana in front of a Monitor/TV which is displaying a white object, how should we perceive the color of the Banana?

    I was thinking that we should not be able to perceive the color because:
    The Monitor/TV puts out three different wavelengths (RGB) only, since the banana could reflect only yellow wavelength, it could not reflect the Red, Green or Blue wavelengths coming out of the Monitor/TV, that way it should be seen as dark.

    But, when I tried it, I could really see it as yellow. Could anyone explain this please?
     
  2. jcsd
  3. Mar 2, 2016 #2

    RUber

    User Avatar
    Homework Helper

    Perhaps the banana is more complicated than you first assumed.
    Your assumption is that the banana absorbs all wavelengths except for yellow. Reconsider this assumption based on new evidence.
     
  4. Mar 2, 2016 #3

    RUber

    User Avatar
    Homework Helper

    I found this interesting reference which might help.
    Note that as the banana ripens, the reflectance in the yellow, orange, and red spectra all significantly increase. The main takeaway here is that the reflectance at no point is simply a single peak in the yellow band.

    http://ucanr.edu/datastoreFiles/234-953.pdf

    banana.PNG
     
  5. Mar 2, 2016 #4
  6. Mar 2, 2016 #5
    Surprise: CONES in your eye actually record red, yellow, blues in combination and the brain combines them as 'yellow'. The object itself is not 'yellow'.

    About 64 percent of cones respond most strongly to red light, while about a third are set off the most by green light. Another 2 percent respond strongest to blue light. Rods detect brightness.

    When reflected light of different wavelengths from a banana hits the cones, it stimulates them to varying degrees. The resulting signal is passed along to the brain, which determines a color: yellow.

    Many birds and fish may have four types of cones, enabling them to see ultraviolet light, or light with wavelengths shorter than what the human eye can perceive.

    PS: good question!
     
  7. Mar 2, 2016 #6
    There are many situations in physics where our senses 'cannot be trusted'. That is, our evolution may not have equipped us to 'intuitively' see things as we imagine. A common example is Einstein passing everyday intuition and realizing that space and time are not fixed and immutable, but rather vary by observer. He correctly concluded only the speed of light in a vacuum is the real 'constant'.

    I can think of a few examples where you and I do not see the 'same color' yellow banana. One obvious situation is where we have inherited different numbers or proportions of color sensing cones. And having us look at a color chart would not enable us to 'pick a difference'. Also, people with color blindness, and I am not sure exactly what that is, may not be able discern,say, red from green. My cousin who is somewhat color blind sometimes see grays where I see 'colors'.

    Also, if you rapidly approach [or depart from ] a banana, approaching a significant percentage of light speed, you'll see a significantly different color than when you are stationary with respect to the banana. The speed of light is fixed, but you'd detect different wavelengths [colors]. Same situation if you were sitting here on earth and watched a banana on a rapidly moving spaceship. As the spaceship picks up speed, the color of the banana changes.
     
    Last edited: Mar 2, 2016
  8. Mar 2, 2016 #7

    tech99

    User Avatar
    Gold Member

    The Monitor/TV puts out three different wavelengths (RGB) only, since the banana could reflect only yellow wavelength, it could not reflect the Red, Green or Blue wavelengths coming out of the Monitor/TV, that way it should be seen as dark.

    But, when I tried it, I could really see it as yellow. Could anyone explain this please?[/QUOTE]
    The colours from the three phosphors are not a single wavelength but cover wide overlapping bands, so the banana does receive illumination.
     
  9. Mar 2, 2016 #8
    All true - the RGB monitor light sources are not monochromatic, neither is the banana spectral reflectance, nor is your retinal sensitivity. Also, the brain compensates for color casts to give you an 'expected color' anyway. If you look at a white shirt after wearing pink sunglasses for a while, you see it as white, although it might appear pink when you first put the glasses on.
     
  10. Mar 2, 2016 #9
    This is not true. Read the posts above, look ate the graph in post #3.
    Or read the paper
    http://ucce.ucdavis.edu/files/datastore/234-953.pdf
     
  11. Mar 2, 2016 #10

    davenn

    User Avatar
    Science Advisor
    Gold Member

    you are missing the point on how colour is created by the TV

    various RGB levels are not "transmitted" in different amounts from the screen and you eye/brain perceives a specific colour from that
    rather the specific colour is generated at/on the TV screen by the mixing of the appropriate 3 colours RGB and that colour is "transmitted" to your eye

    meaning ... a white area on the screen is just that, white, whether you are looking at it or not and that would be confirmed
    by using some other electronic colour sensor rather than your eye

    Dave
     
  12. Mar 2, 2016 #11

    davenn

    User Avatar
    Science Advisor
    Gold Member

  13. Mar 2, 2016 #12

    sophiecentaur

    User Avatar
    Science Advisor
    Gold Member

    There is a great temptation to confuse colour and wavelength. This should be avoided, as it will lead your understanding of colour vision right down the garden path.

    I have just looked at my Apple iMac monitor with my recently obtained, cheap and cheerful. diffraction grating spectroscope. Far from consisting of "three wavelengths". the light from a white screen consists of three very broad bands of 'Reds', 'Greens" and 'Blues' and not three single wavelengths at all. This is not at all surprising because it is hard to produce enough light from a monochromatic source. There is a definite gap between the bands and there are no wavelengths that could be described as yellow or cyan. Using appropriate amounts of each (broadband) phosphor, you can produce a simulation that's acceptable to most normally sighted people of any of the visible colours (with the exceptions of monochromatic sources ( consisting only of one spectral line).
    The answer to the banana question lies in the Analysis of the eye and its model that is used in a TV camera - not with the display technology.
    This link shows the responses of the three sets of colour receptors. Each sensor, although commonly thought of being selectively sensitive to reds, greens and blues, is actually sensitive to almost all visible wavelengths. A 'coloured' image or object will stimulate all three sensors to different degrees (tristimulus values) and it is the ratio of the respective amounts that gives the impression of 'colour'. An infinite number of combinations of incident wavelengths can produce the same perceived colour. The spectrum of light from a banana, as shown above, is very broadband and definitely not 'spectral' (monochromatic) yellow. You can obtain a very good match to this 'yellow' by mixing the outputs from the three RGB phosphors. Human colour vision has no way of doing better than this. It is not a spectrometer and has no need to be. It gets by quite well enough with just three analyses. If more was necessary then evolution would probably have given us more - thousands even, as with our sound detecting hardware (the cochlea) which has a 'proper' audio spectrum analyser built in.
     
  14. Mar 3, 2016 #13
    Thanks a lot everyone for the answers.. Found a lot of valuable information :)
    Unlike what I thought the 3 pixel phosphors emit a band spectrum.

    BTW I searched for spectrum of computer monitor and found the following links.. People have analyzed the spectrum using spectroscopes..
    http://www.chemistryland.com/CHM107Lab/Exp7/Spectroscope/Spectroscope.html
    http://home.freeuk.com/m.gavin/grism2.htm

    Anyone has an idea what type of Color Rendition Index (CRI) corresponding to the white light emitted by a Monitor/TV ?
     
  15. Mar 3, 2016 #14

    sophiecentaur

    User Avatar
    Science Advisor
    Gold Member

    It is not normal to use a TV monitor to illuminate objects and the CRI of a light source tells you how that light source will behave as an illuminant. I have not seen colour monitors assigned a CRI. A colour monitor has a 'white point', which is the point on the colour space of a picture that is taken to be where a perfectly white matt reflecting surface will sit on the CIE colour chart. It is a common operation to change the white point of a picture from where the original illuminant would sit. You can warm up a 'cold' scene by altering the colour balance and you can sometimes reduce the ghastly effects of colour cast that some fluorescent lamps can inflict on an otherwise pretty scene.
    There is a distinction between analysis and synthesis of colours and the two processes have to be treated differently. The finer points are beyond me and you are welcome to get stuck into the details of colourimetry. But watch out for the high levels of BS that non technical but often very artistically talented people use when describing how they deal with colour. A lot of the published material is about printing and paints, which is a very dark art, in my experience.
     
  16. Mar 3, 2016 #15

    Andy Resnick

    User Avatar
    Science Advisor
    Education Advisor
    2016 Award

    Playing is fun, isn't it? :)
     
  17. Mar 3, 2016 #16

    sophiecentaur

    User Avatar
    Science Advisor
    Gold Member

    The thing only cost me about $30 (including postage from the US). I have learned that LED lamps have pretty continuous spectra, which explains how they work much better than the CFLs with a similar perceived 'colour'. The CFL spectrum is like a dog's dinner with lines, odd bands and gaps all over the place. No wonder they play hell with colour matching.
    Every home should have one (spectroscope, I mean). Plus it doesn't need a danged battery!!!!!
     
  18. Mar 3, 2016 #17

    davenn

    User Avatar
    Science Advisor
    Gold Member


    and for that alone they should be banned
     
  19. Mar 3, 2016 #18

    sophiecentaur

    User Avatar
    Science Advisor
    Gold Member

    I think Natural Selection will cause their demise pretty soon. LEDs are getting so cheap and CFLs have pretty short lives, mostly. (Despite what we were told)
     
  20. Mar 3, 2016 #19

    davenn

    User Avatar
    Science Advisor
    Gold Member

    yeah, I have discovered a poor life span out of many of them
    plus they have the same toxic chemicals in them as full size fluoro's do :rolleyes::rolleyes:

    a total con perpetrated on society (another to add to the growing list haha)
     
  21. Mar 3, 2016 #20

    OmCheeto

    User Avatar
    Gold Member
    2016 Award

    My spectroscope came in yesterday. I did the same. :smile:

    I'm now developing plans on how best to illuminate a banana to get a clear spectrum, to verify that bananas are not in fact yellow, but just red & green spotted monsters. :biggrin: (See post #3)

    Then, I'll take an image of the banana in sunlight, as that is the "CRI" reference. (kind of, I think)
    Then, I'll transfer the image to my PC, and take an image of that image, and compare them spectrally, via RGB values.
    And then we will have the "CRI", within at least 2 orders magnitude accuracy, as that is usually how my experiments end up.

    ps. Those diffraction gratings are the bomb! I should have bought a dozen.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted



Similar Discussions: Perceived Color of objects: in White light from a TV
Loading...