Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Color images of red/green lasers

  1. Apr 29, 2015 #1
    Hi all,

    I'm new to the forums.

    I recently tried to take an image of two laser dots on a white wall, using my laptop's USB camera. I saved them to PNGs.

    I have a number of questions...
    1. There was an artifact with the green laser, where a smaller second dot appeared for some reason. What caused the artifact?

    2. The lasers are AFAIK monochromatic, however the red or green in the images clearly are not monochromatic. Why is this?

    3. Why are the central, brightest regions of the laser dots colored as white? My hunch is that it has something to do with the Bayer interpolation algorithm that interprets very bright red/green (and probably, blue) spots as white, without checking whether the other colors are bright too.

    Cheers,
    J.
     

    Attached Files:

  2. jcsd
  3. Apr 29, 2015 #2

    DrClaude

    User Avatar

    Staff: Mentor

    Welcome to PF!

    I would guess internal reflection. What type of camera did you use? If you center the green laser, do you see multiple additional dots?


    I think these two are related. The light is probably so bright that it not only saturates the red or green sensors, but is also detected by the other sensors (imperfect filtering).
     
  4. Apr 29, 2015 #3

    meBigGuy

    User Avatar
    Gold Member

    the camera has filters for the CCD (http://en.wikipedia.org/wiki/Bayer_filter). If the red and green do not just happen to exactly match the filters, or if there is filter overlap, the image will appear as "not monochromatic". That is, it will show contents in multiple sensors.

    Note that there are also more green sensors than blue or red. Maybe the extra dot is caused by the post processing algorithms. You can try to analyze what happens as you change angles and positions. (it is most likely a reflection, as DrClaude suggested, it could be in the lens array)
     
  5. Apr 29, 2015 #4
    Hi all,

    Thanks for the replies.

    Regarding the little green dot artifact, it is still there if I try taking it from a different position; see second attached photo. The camera is an in-built camera on the laptop (driver is that of a USB camera).

    So, if I understand your explanations correctly, an imperfect filter is a filter that does not entirely reduce the spectral transmittance of other wavelengths intended to be filtered, to zero. Hence for the red or green laser, some light is detected by the green/blue or red/blue pixels, and the interpolation algorithm interpolates this to be white light.

    Maybe these are silly questions, but: clearly any real-life scene is a juxtaposition of different color wavelengths + intensities. How can any digital, 2D color image represent different intensities of monochromatic light? How does the human eye represent different intensities of monochromatic light? We clearly have a concept in our brain of a bright or dim color.

    If you wanted to take a picture of a monochromatic laser dot, and wanted it to display the dot as:
    1. Monochromatic
    2. Without any white-sections

    How would you do it?

    Thanks for all your input!

    Cheers,
    J.
     

    Attached Files:

  6. Apr 29, 2015 #5

    russ_watters

    User Avatar

    Staff: Mentor

    Reduce the exposure/brightness so the center of the dot isn't blown-out.
     
  7. Apr 29, 2015 #6

    meBigGuy

    User Avatar
    Gold Member

    Advanced industrial cameras have sensors for each color. For example, take three CCDs each with a different color filter. On some systems multiple exposures are made on a single sensor, each with a different filter (like mars rover, satellites, etc.)

    All filters have "bandwidth" and multiple filters will overlap. The odds of a filter's center frequency exactly matching your laser is small unless you specify it to be such.

    If you want a monochromatic picture, use a monochromatic sensor. Take a B/W picture and edit it to be green.

    You are also seeing dispersion for various reasons.

    As for the dot, you need to examine how its position moves as the camera angle changes.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook