Helpful Posts Helpful Posts:  0
Results 1 to 6 of 6

Thread: RGB Cavities

  1. #1

    RGB Cavities

    So, i was reading the first tutorial/lesson, which is about light/color cavities.

    "To capture color images, a filter has to be placed over each cavity that permits only particular colors of light. Virtually all current digital cameras can only capture one of three primary colors in each cavity, and so they discard roughly 2/3 of the incoming light. As a result, the camera has to approximate the other two primary colors in order to have full color at every pixel. "

    For which i understood, there are two green filters, one blue filter, and one red filter in each of the four photosites that compound every pixel. So why does the camera has to discard, and approximate "the other two primary colors in order to have full color at every pixel". ?

    I love Cambdrige in Colour.
    Bruno.

  2. #2
    GiacomoD's Avatar
    Join Date
    Jan 2013
    Location
    Florence, Italy
    Posts
    168
    Real Name
    Giacomo

    Re: RGB Cavities

    Hi Bruno, welcome on CiC, I love it too!

    Well, the subject is not the most easy to be treated. Anyway, for what I understood, each single cavity of the sensor actually record only one of the three primary colors, discarding the incoming light in the two others. That's why in the tutorial you find that it discards 2/3 of the incoming light. For some technical reasons that are beyond my knowledge, cavities recording the green color are - numerically - double than those recording red and blue.
    When opening a RAW file in any raw converter, the full color of each pixel is reconstructed on the basis of the values of the surrounding pixels, giving a normal RGB image.

    Maybe other members will be able to provide more in depth explanations

    Cheers

    Giacomo

  3. #3
    Moderator GrumpyDiver's Avatar
    Join Date
    Mar 2012
    Location
    Ottawa, Canada
    Posts
    12,357
    Real Name
    Manfred Mueller

    Re: RGB Cavities

    Bruno – a photodetector is sensitive to all frequencies of light (and in fact sensitive to both near ultraviolet and infrared as well). If the photodetector did not have the coloured filters, you would have a monochrome camera. This is exactly how the Leica Monochrom camera works.

    If you want to record colour data, the light has to be filtered to break it down into the three constituent wavelengths that our eyes see as Red, Green and Blue (RGB); technically, this is implemented using a Bayer filter array. Coloured filters are fitted over the photodiodes to ensure that each sensor counts the photons that only correspond to those specific light frequencies and discard all others. This is why only 1/3 of the total light falling on an individual photodiode is recorded.

    Giacomo –the reason for two greens is that this closely matches the colour sensitivity of the human eye; which is much more sensitive to green than either red or blue.

  4. #4
    GiacomoD's Avatar
    Join Date
    Jan 2013
    Location
    Florence, Italy
    Posts
    168
    Real Name
    Giacomo

    Re: RGB Cavities

    Quote Originally Posted by GrumpyDiver View Post
    Giacomo –the reason for two greens is that this closely matches the colour sensitivity of the human eye; which is much more sensitive to green than either red or blue.
    Manfred,
    this is interesting and surprising to me at the same time, since it seems to be contradicting what I learnt many years ago when studying Remote Sensing and Satellite Images Interpretation: in Remote Sensing it's commonly accepted, for environmental studies, the use of false color RGB images, where the Near Infrared channel (the most important for detection of healthy vegetation) is arbitrarily represented with the red channel. This was justified with the fact that the human eye is more sentitive to slight variations in this color than in green and blue.
    I'll have to investigate more on this

    Thanks
    Giacomo

  5. #5
    Moderator GrumpyDiver's Avatar
    Join Date
    Mar 2012
    Location
    Ottawa, Canada
    Posts
    12,357
    Real Name
    Manfred Mueller

    Re: RGB Cavities

    Quote Originally Posted by GiacomoD View Post
    Manfred,
    this is interesting and surprising to me at the same time, since it seems to be contradicting what I learnt many years ago when studying Remote Sensing and Satellite Images Interpretation: in Remote Sensing it's commonly accepted, for environmental studies, the use of false color RGB images, where the Near Infrared channel (the most important for detection of healthy vegetation) is arbitrarily represented with the red channel. This was justified with the fact that the human eye is more sentitive to slight variations in this color than in green and blue.
    I'll have to investigate more on this

    Thanks
    Giacomo
    I suspect that this might be more of a historical artifact than anything else.

    I remember shooting Kodak Infrared Ektachrome colour reversal film many years ago, and the IR component was represented in the red tones. From a technical standpoint that made sense, as the IR frequencies would be closest to the red colour frequencies and any sensitization / filters in the film emulsion would have been easiest to implement in the red layer.

  6. #6

    Join Date
    Feb 2013
    Location
    Turkey
    Posts
    27
    Real Name
    Altay

    Re: RGB Cavities

    Human eye is more sensitive to green at every luminance levels, so we can see greens more easily at every lighting conditions. In dim conditions, this is a little bit light green.

    You may google for "photopic luminosity function" for daylight sensitivities of human eye and "scotopic luminosity function" for dim and dark light sensitivities of human eye.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •