Loading ...
Sorry, an error occurred while loading the content.

RGB data collection

Expand Messages
  • Joe Morris
    If you have a set of RGB filters that a G2V star calibration gave you the following weights,R= 1.2, G=1, B=1.8, which of the following would be the preferred
    Message 1 of 5 , Feb 26, 2009
    • 0 Attachment
      If you have a set of RGB filters that a G2V star calibration gave you
      the following weights,R= 1.2, G=1, B=1.8, which of the following would
      be the preferred senerio for data collection and stacking? Maybe it
      makes no difference which senerio is chosen although it would be easier
      to maintain a dark calibration library with senerio A or B.

      A. Take 10 ten minute images of each R,G,B and combine them using
      the 1.2:1: 1.8 weights

      B. Take 12 ten minute R images, 10 ten minute G images, and 18 ten B
      images and combine them using 1:1:1 weights

      C. Take 10 twelve minute R images, 10 ten minute G images, and 10
      eighteen G images and combine them using 1:1:1 weights

      Joe
    • San Diego Paul
      Wow that s a good Q and I think I ve tried all of the below options. My current MO is method #C -Paul in San Diego
      Message 2 of 5 , Feb 28, 2009
      • 0 Attachment
        Wow that's a good Q and I think I've tried all of the below options.
        My current MO is method #C

        -Paul in San Diego

        >
        > If you have a set of RGB filters that a G2V star calibration gave you
        > the following weights,R= 1.2, G=1, B=1.8, which of the following would
        > be the preferred senerio for data collection and stacking? Maybe it
        > makes no difference which senerio is chosen although it would be easier
        > to maintain a dark calibration library with senerio A or B.
        >
        > A. Take 10 ten minute images of each R,G,B and combine them using
        > the 1.2:1: 1.8 weights
        >
        > B. Take 12 ten minute R images, 10 ten minute G images, and 18 ten B
        > images and combine them using 1:1:1 weights
        >
        > C. Take 10 twelve minute R images, 10 ten minute G images, and 10
        > eighteen G images and combine them using 1:1:1 weights
        >
        > Joe
        >
      • Hilary Jones
        If you use dark frame scaling, all three options work well enough. But if you want to use darks that are exactly the same length as your exposures, then
        Message 3 of 5 , Feb 28, 2009
        • 0 Attachment
          If you use dark frame scaling, all three options work well enough.
          But if you want to use darks that are exactly the same length as your
          exposures, then options A and B certainly are easier.

          Options B and C tend to equalize the amount of shot noise in all
          three channels, so I would probably not use option A. (Actually
          truth be known, I use option A all the time because it's easier on my
          brain!)

          With option C, you equalize the read noise better than option B. My
          gut feeling is that read noise isn't an issue, but your mileage might
          vary.

          With option B, you have to be careful how you combine the images. If
          you average them, then the weights will be 1.2:1: 1.8, but if you sum
          them, then the weights will be 1:1:1.

          There's a caveat that you probably only have a certain amount of time
          available to do your exposures; so if the total time is 300 minutes,
          option C would actually use 9, 7.5, and 13.5 minute exposures, while
          option B might require 9, 8, and 13 exposures, depending on how you
          want to round the numbers. I haven't figured out if this has any
          real bearing on the question you asked, but I thought I would mention
          it anyhow.

          Hilary

          --- In ccd-newastro@yahoogroups.com, "Joe Morris" <joemorris@...>
          wrote:
          >
          > If you have a set of RGB filters that a G2V star calibration gave
          you
          > the following weights,R= 1.2, G=1, B=1.8, which of the following
          would
          > be the preferred senerio for data collection and stacking? Maybe
          it
          > makes no difference which senerio is chosen although it would be
          easier
          > to maintain a dark calibration library with senerio A or B.
          >
          > A. Take 10 ten minute images of each R,G,B and combine them
          using
          > the 1.2:1: 1.8 weights
          >
          > B. Take 12 ten minute R images, 10 ten minute G images, and 18 ten
          B
          > images and combine them using 1:1:1 weights
          >
          > C. Take 10 twelve minute R images, 10 ten minute G images, and 10
          > eighteen G images and combine them using 1:1:1 weights
          >
          > Joe
          >
        • Joe Morris
          Hi Hilary -- Yes, I was asking the question to find the most time efficient way to collect the data if in fact it makes no difference in the final quality of
          Message 4 of 5 , Mar 1 8:02 AM
          • 0 Attachment
            Hi Hilary -- Yes, I was asking the question to find the most time
            efficient way to collect the data if in fact it makes no difference in
            the final quality of the combined data. It seems to me that with "A",
            one is throwing out some data just to equalize the filter differences.
            "B" looks like the best in order to make life easier with the dark
            library. I usually use one of the statistical combining methods like
            SD Mask or sigma clip. Using method "B" with one of these, would the
            combine ratio be 1:1:1?

            Joe

            --- In ccd-newastro@yahoogroups.com, "Hilary Jones" <hilaryyahoo3@...>
            wrote:
            >
            > If you use dark frame scaling, all three options work well enough.
            > But if you want to use darks that are exactly the same length as your
            > exposures, then options A and B certainly are easier.
            >
            > Options B and C tend to equalize the amount of shot noise in all
            > three channels, so I would probably not use option A. (Actually
            > truth be known, I use option A all the time because it's easier on my
            > brain!)
            >
            > With option C, you equalize the read noise better than option B. My
            > gut feeling is that read noise isn't an issue, but your mileage might
            > vary.
            >
            > With option B, you have to be careful how you combine the images. If
            > you average them, then the weights will be 1.2:1: 1.8, but if you sum
            > them, then the weights will be 1:1:1.
            >
            > There's a caveat that you probably only have a certain amount of time
            > available to do your exposures; so if the total time is 300 minutes,
            > option C would actually use 9, 7.5, and 13.5 minute exposures, while
            > option B might require 9, 8, and 13 exposures, depending on how you
            > want to round the numbers. I haven't figured out if this has any
            > real bearing on the question you asked, but I thought I would mention
            > it anyhow.
            >
            > Hilary
            >
            > --- In ccd-newastro@yahoogroups.com, "Joe Morris" <joemorris@>
            > wrote:
            > >
            > > If you have a set of RGB filters that a G2V star calibration gave
            > you
            > > the following weights,R= 1.2, G=1, B=1.8, which of the following
            > would
            > > be the preferred senerio for data collection and stacking? Maybe
            > it
            > > makes no difference which senerio is chosen although it would be
            > easier
            > > to maintain a dark calibration library with senerio A or B.
            > >
            > > A. Take 10 ten minute images of each R,G,B and combine them
            > using
            > > the 1.2:1: 1.8 weights
            > >
            > > B. Take 12 ten minute R images, 10 ten minute G images, and 18 ten
            > B
            > > images and combine them using 1:1:1 weights
            > >
            > > C. Take 10 twelve minute R images, 10 ten minute G images, and 10
            > > eighteen G images and combine them using 1:1:1 weights
            > >
            > > Joe
            > >
            >
          • Hilary Jones
            SD Mask and sigma clipping are just fancy ways of doing averages, so the weights should be 1.2:1: 1.8. Hilary ... in ... with A , ... differences. ... the
            Message 5 of 5 , Mar 1 5:21 PM
            • 0 Attachment
              SD Mask and sigma clipping are just fancy ways of doing averages, so
              the weights should be 1.2:1: 1.8.

              Hilary

              --- In ccd-newastro@yahoogroups.com, "Joe Morris" <joemorris@...>
              wrote:
              >
              > Hi Hilary -- Yes, I was asking the question to find the most time
              > efficient way to collect the data if in fact it makes no difference
              in
              > the final quality of the combined data. It seems to me that
              with "A",
              > one is throwing out some data just to equalize the filter
              differences.
              > "B" looks like the best in order to make life easier with the dark
              > library. I usually use one of the statistical combining methods like
              > SD Mask or sigma clip. Using method "B" with one of these, would
              the
              > combine ratio be 1:1:1?
              >
              > Joe
              >
              > --- In ccd-newastro@yahoogroups.com, "Hilary Jones" <hilaryyahoo3@>
              > wrote:
              > >
              > > If you use dark frame scaling, all three options work well
              enough.
              > > But if you want to use darks that are exactly the same length as
              your
              > > exposures, then options A and B certainly are easier.
              > >
              > > Options B and C tend to equalize the amount of shot noise in all
              > > three channels, so I would probably not use option A. (Actually
              > > truth be known, I use option A all the time because it's easier
              on my
              > > brain!)
              > >
              > > With option C, you equalize the read noise better than option B.
              My
              > > gut feeling is that read noise isn't an issue, but your mileage
              might
              > > vary.
              > >
              > > With option B, you have to be careful how you combine the
              images. If
              > > you average them, then the weights will be 1.2:1: 1.8, but if you
              sum
              > > them, then the weights will be 1:1:1.
              > >
              > > There's a caveat that you probably only have a certain amount of
              time
              > > available to do your exposures; so if the total time is 300
              minutes,
              > > option C would actually use 9, 7.5, and 13.5 minute exposures,
              while
              > > option B might require 9, 8, and 13 exposures, depending on how
              you
              > > want to round the numbers. I haven't figured out if this has any
              > > real bearing on the question you asked, but I thought I would
              mention
              > > it anyhow.
              > >
              > > Hilary
              > >
              >
            Your message has been successfully submitted and would be delivered to recipients shortly.