1. Forum
  2. Usenet
  3. SCI.IMAGE.PROCESSING

  • image density, log exposure (-target density) and pixel level

    From imran.kanjoo@gmail.com@21:1/5 to All on Sat Jan 26 00:34:16 2019
    Hello, every body. I want to find the pixel level of the image. The image contains 6 gray scale squares with different intensity in each patch. I know the pixel level is given by

    pixel level = 255 * (10–density/1.01)(1/2.2)

    but I do not know how to find out the density. Finally I need to have a graph pixel level vs log exposure (-target density). Can any body also explain how can I find the log exposure (-target density). Thank you.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From dale@21:1/5 to imran.kanjoo@gmail.com on Sat Jan 26 16:19:41 2019
    On 1/26/2019 3:34 AM, imran.kanjoo@gmail.com wrote:
    Hello, every body. I want to find the pixel level of the image. The image contains 6 gray scale squares with different intensity in each patch. I know the pixel level is given by

    pixel level = 255 * (10–density/1.01)(1/2.2)

    but I do not know how to find out the density. Finally I need to have a graph pixel level vs log exposure (-target density). Can any body also explain how can I find the log exposure (-target density). Thank you.


    The pixel level you described (exposure) is an approximation of a gamma
    2.2 level system, closest to L* (Lightness), but not Lightness

    https://en.wikipedia.org/wiki/Lightness

    <https://en.wikipedia.org/wiki/Lightness>

    see gamma ...,

    https://en.wikipedia.org/wiki/Gamma_correction

    <https://en.wikipedia.org/wiki/Gamma_correction>

    Kinda depends on what you are working with ... calibration or
    characterization? sometimes calibration is "good enough"
    characterization ...

    to answer your second question it looks like you already know the
    exposure (pixel density discussed above) ... invert equation for density

    to answer you first question ... How do I find density? ... unless it is
    linear with exposure, or you have beforehand design knowledge of the
    system, you will have most likely have to measure



    --
    dale - https://www.dalekelly.org/
    Not a professional opinion unless specified.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From dale@21:1/5 to dale on Tue Feb 5 18:19:34 2019
    On 1/26/2019 4:19 PM, dale wrote:
    On 1/26/2019 3:34 AM, imran.kanjoo@gmail.com wrote:
    Hello, every body. I want to find the pixel level of the image. The
    image contains 6 gray scale squares with different intensity in each
    patch. I know the pixel level is given by

    pixel level = 255 * (10–density/1.01)(1/2.2)

    but I do not know how to find out the density. Finally I need to have
    a graph pixel level vs log exposure (-target density). Can any body
    also explain how can I find the log exposure (-target density). Thank
    you.


    The pixel level you described (exposure) is an approximation of a gamma
    2.2 level system, closest to L* (Lightness), but not Lightness

    https://en.wikipedia.org/wiki/Lightness

    <https://en.wikipedia.org/wiki/Lightness>

    see gamma ...,

    https://en.wikipedia.org/wiki/Gamma_correction

    <https://en.wikipedia.org/wiki/Gamma_correction>

    Kinda depends on what you are working with ... calibration or characterization? sometimes calibration is "good enough"
    characterization ...

    to answer your second question it looks like you already know the
    exposure (pixel density discussed above) ... invert equation for density

    to answer you first question ... How do I find density? ... unless it is linear with exposure, or you have beforehand design knowledge of the
    system, you will have most likely have to measure




    apologies gamma 1.8 instead of gamma 2.2 might be closer to L* with
    white point of 5000K

    --
    dale - https://www.dalekelly.org/
    Not a professional opinion unless specified.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)