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

Cosmic Microwave Background Radiation and Photography

  1. Apr 22, 2012 #1
    Hi firstly I am a photographer not a physicist,

    When taking photos there is always 'noise' in the image, especially at higher ISO. I presumed this is the same noise that we would hear on a radio, or see on a tv? Would it be possible to create a room that blocked out the background radiation allowing me to take a noiseless image?

    Thanks, P
     
  2. jcsd
  3. Apr 22, 2012 #2

    Vanadium 50

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    I don't think this has anything to do with the CMBR.
     
  4. Apr 22, 2012 #3

    phyzguy

    User Avatar
    Science Advisor

    I'm pretty sure that the noise in your photos is not due to the CMB, as your title implies. The CMB photons have far too little energy to change the state of the detector. Depending on your detector (what is it - photographic emulsion, CCD imager, CMOS imager?) the noise is probably thermal noise. You should be able to reduce it by cooling your detector. Astronomical supply houses sell CCD imagers with thermoelectric coolers which allow you to cool the detectors to reduce noise.
     
  5. Apr 22, 2012 #4
    Thanks for the reply, Wasn't sure what caused this noise, just a poorly informed presumption. Thanks for your help, have now been able to dig up alot of information on the subject. Not sure how difficult it would be to mod my camera (CMOS) with a cooler, could be a nice project though! So would I be right in thinking I would have cleaner images in colder environments? And would my images taken from a 'cold start' have less noise than an image taken if the sensor has fully warmed up?
     
  6. Apr 22, 2012 #5

    f95toli

    User Avatar
    Science Advisor
    Gold Member

    I doubt it would be practical to modify an existing camera to include a cooler, nor is it obvious that it would help (there are other -non-thermal- sources of noise).

    There are cameras/sensors for telescopes (meant for amateurs) that you can buy that do come with a peltier coolers .
     
  7. Apr 23, 2012 #6

    Drakkith

    User Avatar

    Staff: Mentor

    What are you taking pictures of? Are you taking very very short exposures at high ISO?
     
  8. Apr 23, 2012 #7
    Not really just general photography, mostly portraits and landscapes, nothing too scientific. Getting clean Images at ISO 50 or 100 has been difficult in some locations, getting lights with enough power and a decent CRI rating outdoors is a bit of a nightmare. I Would do star trails sometimes, they would also be victim to noise due to long shutter speed.

    Thanks again for all help, if an admin could change the title of this thread to 'Noise in Photography?' That would be great as some of this info could be useful to photographers/astronomers.
     
  9. Apr 23, 2012 #8

    Drakkith

    User Avatar

    Staff: Mentor

    I'd say post a thread in the Photography section of the General forum asking about noise.
     
  10. Apr 23, 2012 #9

    Chronos

    User Avatar
    Science Advisor
    Gold Member

    Even on an 'empty' radio or tv channel, the CMB only accounts for about 1% of the total 'noise'. The CMB peaks in the microwave frequency range, so it would have a negligible effect on photography.
     
  11. Apr 25, 2012 #10

    davenn

    User Avatar
    Science Advisor
    Gold Member

    regardless of if its film or digital photoghraphy, the primary presence of noise, is a low signal to noise ratio ... ie. not enough light to produce a clear image
    now in a digital camera situation you also have the significance of thermal noise of the detector,
    CCD or CMOS, electronics. This is more and more evident with longer exposures at lower light levels.

    cheers
    Dave
     
  12. Apr 25, 2012 #11
    I would have thought longer exposures reduce the noise, assuming the exposure is turned down proportionally, we have a longer sample and more time for the noise to average out to zero. Can't you lower the exposure compensation (not the aperture) until the noise becomes unnoticeable?

    In a similar way that summing 10 random Gaussian numbers and dividing by 10 gives a higher variance than 100 Gaussian random numbers divided by 100.
     
  13. Apr 25, 2012 #12

    davenn

    User Avatar
    Science Advisor
    Gold Member

    not on any digital camera I have had anything to do with. The longer the imaging chip is operating, the thermally hotter its getting. The only way to beat thermal noise is to take many shorter exposures and stack them.
    Peltier coolers as mentioned by another poster along with the many shorter exposures gets the least noise. very practical for astrophotography, not so for terresterial /portrait etc photography, where its always about lighting.

    actually there is at least one guy in the USA that is modifying specific model Canon cameras to include thermal cooling. I saw his posts on the "cloudynights.com" forum
    good results :)

    cheers
    Dave
     
  14. Apr 25, 2012 #13
    I'm afraid I'm a bit out of my depth here, but you most definitely experience more noise at longer exposures, its more noticeable in timelapses, which makes me believe that's it has more to do with 'thermal noise' that others have mentioned. Exposure compensation is normally controlled by the apeture (f-stop), some cameras may process the image differently to change exposure but that could be done in post more effeciently.
     
  15. Apr 25, 2012 #14

    Andy Resnick

    User Avatar
    Science Advisor
    Education Advisor
    2016 Award

    Most likely, the origin of noise in your digital images is from the electronics- amplifier noise, if you like. This will also vary with temperature- hotter electronics have a higher 'noise temperature'. "thermal noise", as the term is used in imaging, only becomes relevant for infrared imaging- the chip must be colder than the IR radiation it is imaging.

    I'm surprised you have objectionable noise at ISO 100 and 50 settings- the amplifier gain is very low there. Different cameras behave differently, but in general you should be able to operate up to ISO 1600 before the noise is bothersome.
     
  16. Apr 25, 2012 #15
    ISO 50-100 is absolutely fine, just a problem getting the lights required to certain locations. Surely IR makes up some of the light in my image? I suppose what I was asking was, what it would take to get a clean image at ISO 6400 or more.
     
  17. Apr 25, 2012 #16

    Andy Resnick

    User Avatar
    Science Advisor
    Education Advisor
    2016 Award

    Generally, camera manufacturers add a 'cut filter' to block wavelengths longer than say, 750 nm, from reaching the chip- silicon is sensitive out to about 1100 nm, and without the cut filter, lens aberrations will become intolerable since that light is outside of the design parameters. Plus the recorded image will not look much like you expect it to since you can't see those longer wavelengths. To be sure, there are specialized lenses that are well corrected over a very broad band, and some folks remove the cut filter to do IR and UV photography.

    ISO 6400 represents a fairly high amplifier gain, there's not much you can do other than apply sharpness-robbing filters to average out the noise. For extremely low light level imaging (single molecule stuff), I use a EMCCD camera, but that's not appropriate for 'normal' photography.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Cosmic Microwave Background Radiation and Photography
  1. Background radiation? (Replies: 24)

  2. Microwave Radiation? (Replies: 17)

Loading...