3 ms·

In the case of digital astronomy using various opto-electronic devices to take the light and convert to charge (CMOS or CCD sensor or whatever else they use), I

by xt00 7y ago

In the case of digital astronomy using various opto-electronic devices to take the light and convert to charge (CMOS or CCD sensor or whatever else they use), I wonder how possible it is to essentially sum together lots of pictures one after another where occasionally you would reject photos that had a satellite passing through them? Like is that how they deal with it now? In the case of film there is no "dark noise" that you have to deal with, so you just leave the shutter open for hours. But with digital if you leave the charge well open for even seconds, the dark noise starts to accumulate in the pixel. Not totally sure though how long exposures work on digital cameras -- if its actually a sum of shorter exposures or what..

joshvm 7y ago
> In the case of digital astronomy using various opto-electronic devices to take the light and convert to charge (CMOS or CCD sensor or whatever else they use), I wonder how possible it is to essentially sum together lots of pictures one after another where occasionally you would reject photos that had a satellite passing through them? An exposure that a satellite passes through is potentially ruined. On the plus side, although there are lots of satellites, the odds of one of them passing through your target is relatively slim. A lot of the time astronomers are interested in a single star in the entire field of view. This has happened to colleagues of mine however, so it definitely occurs. Other things which can ruin exposures are cosmic ray hits which you have no control over. You could do a median sum to get rid of the offending noise. Problem is you often don't know until after you've got the data back; it's quite rare for astronomers to do their own observations and you get allocated a fixed amount of telescope time. So you might get given 1 hour of time on your target, but if you messed up the exposure calculations or a satellite happened to pass through, there's very little you can do about it. > Not totally sure though how long exposures work on digital cameras -- if its actually a sum of shorter exposures or what.. The detectors used in astronomy are extremely low noise and can be used for fairly long exposures (10s of minutes) without stacking. That said, you can still use stacking to get better signal to noise. It's much more commonly performed with amateur observations though, due to noisier detectors. You also need to take dark, bias and flat frames to compensate for detector noise. https://www.eso.org/~ohainaut/ccd/ https://www.eso.org/~ohainaut/ccd/
cbanek 7y ago
> I wonder how possible it is to essentially sum together lots of pictures one after another where occasionally you would reject photos that had a satellite passing through them? Like is that how they deal with it now? This is what LSST is going to do, sum the images to look deeper than just one image. There are a few commercial software packages that also do this co-addition (my friend has a large but still amateur telescope and that software does this). The exposures aren't terribly long, so the idea is to add a bunch of them together, since the CCD wells can become saturated, and even "leak" between them. This is why the camera is cooled down by extra cryogenics.

by xt00 7y ago

joshvm 7y ago

> In the case of digital astronomy using various opto-electronic devices to take the light and convert to charge (CMOS or CCD sensor or whatever else they use), I wonder how possible it is to essentially sum together lots of pictures one after another where occasionally you would reject photos that had a satellite passing through them? An exposure that a satellite passes through is potentially ruined. On the plus side, although there are lots of satellites, the odds of one of them passing through your target is relatively slim. A lot of the time astronomers are interested in a single star in the entire field of view. This has happened to colleagues of mine however, so it definitely occurs. Other things which can ruin exposures are cosmic ray hits which you have no control over. You could do a median sum to get rid of the offending noise. Problem is you often don't know until after you've got the data back; it's quite rare for astronomers to do their own observations and you get allocated a fixed amount of telescope time. So you might get given 1 hour of time on your target, but if you messed up the exposure calculations or a satellite happened to pass through, there's very little you can do about it. > Not totally sure though how long exposures work on digital cameras -- if its actually a sum of shorter exposures or what.. The detectors used in astronomy are extremely low noise and can be used for fairly long exposures (10s of minutes) without stacking. That said, you can still use stacking to get better signal to noise. It's much more commonly performed with amateur observations though, due to noisier detectors. You also need to take dark, bias and flat frames to compensate for detector noise. https://www.eso.org/~ohainaut/ccd/ https://www.eso.org/~ohainaut/ccd/

cbanek 7y ago

> I wonder how possible it is to essentially sum together lots of pictures one after another where occasionally you would reject photos that had a satellite passing through them? Like is that how they deal with it now? This is what LSST is going to do, sum the images to look deeper than just one image. There are a few commercial software packages that also do this co-addition (my friend has a large but still amateur telescope and that software does this). The exposures aren't terribly long, so the idea is to add a bunch of them together, since the CCD wells can become saturated, and even "leak" between them. This is why the camera is cooled down by extra cryogenics.