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

Binning question

Expand Messages
  • Eddy
    Hi all, Could somebody, in a few words, explain me why one would use binning? I see that some imagers are binning 2x2 the RGB channels but using 1x1 for the
    Message 1 of 22 , Oct 4, 2007
      Hi all,

      Could somebody, in a few words, explain me why one would use binning?

      I see that some imagers are binning 2x2 the RGB channels but using 1x1
      for the luminance. Others are using 1x1 for all the channels. To me,
      both ways look the same.

      I understand that binning 2x2 will reduce exposure times. But that is
      the only "advantage" I am seeing.
      When binning 2x2, the image scale will be different - if I understood
      this correctly - than with 1x1.
      How do you convert or stack these? Do you have to change the image
      size before stacking?

      Thanks for any info,
      Eddy
    • Paul K
      Hi Eddy, Binning has one advantage over not binning: it reduces the read-out noise of the camera. When taking filtered exposures of faint objects, read-out
      Message 2 of 22 , Oct 4, 2007
        Hi Eddy,

        Binning has one advantage over not binning: it reduces the read-out
        noise of the camera. When taking filtered exposures of faint
        objects, read-out noise can be the dominant noise in the image, so
        it may be advantageous to bin RGB frames, especially in LRGB imaging.

        But, if your RGB exposure is long enough to overcome read-out
        noise, or the object is bright enough to drown out this noise,
        there's no good reason to bin (maybe just one: the images are
        smaller and download faster <g>)

        If you bin 2x2, the image scale will be twice that of 1x1, i.e., one-
        half the resolution. To register and combine binned images with
        unbinned you can manually resize the image to 200%, or use software
        that can do this for you automatically (CCDStack, Registar, etc.)

        Regards,

        -Paul


        --- In ccd-newastro@yahoogroups.com, "Eddy" <EddyMuriel@...> wrote:
        >
        > Hi all,
        >
        > Could somebody, in a few words, explain me why one would use
        binning?
        >
        > I see that some imagers are binning 2x2 the RGB channels but using
        1x1
        > for the luminance. Others are using 1x1 for all the channels. To
        me,
        > both ways look the same.
        >
        > I understand that binning 2x2 will reduce exposure times. But that
        is
        > the only "advantage" I am seeing.
        > When binning 2x2, the image scale will be different - if I
        understood
        > this correctly - than with 1x1.
        > How do you convert or stack these? Do you have to change the image
        > size before stacking?
        >
        > Thanks for any info,
        > Eddy
        >
      • Mike Dodd
        ... you are correct; binning 2x2 creates a super pixel with four times the area of a single pixel, so it s four times more sensitive. Color filters typically
        Message 3 of 22 , Oct 4, 2007
          Eddy wrote:
          > Hi all,
          >
          > Could somebody, in a few words, explain me why one would use binning?
          >
          > I see that some imagers are binning 2x2 the RGB channels but using 1x1
          > for the luminance. Others are using 1x1 for all the channels. To me,
          > both ways look the same.
          >
          > I understand that binning 2x2 will reduce exposure times. But that is
          > the only "advantage" I am seeing.
          > When binning 2x2, the image scale will be different - if I understood
          > this correctly - than with 1x1.
          > How do you convert or stack these? Do you have to change the image
          > size before stacking?

          you are correct; binning 2x2 creates a "super pixel" with four times the
          area of a single pixel, so it's four times more sensitive. Color filters
          typically reduce the amount of light reaching the chip, so this
          increased sensitivity avoids having to take extremely long RGB
          exposures.the other benefit comets from using the color channels in
          conjunction with am unbinned luminance channel. the luminance channel
          provides the sharpness and detail, while the color channels don't have
          to be so sharp. In fact,many people blur the color channels before
          layering in Photoshop to form the LRGB image.

          You are correct about the image scale, but all that really affects is
          the resolution, which isn't important if you're going to blur the RGB
          channels anyway. the color frames are resized (doubled in each
          dimension) before combining to match the luminance frame size.

          --- Mike
          --

          Mike Dodd
          Montpelier, VA USA
          http://astronomy.mdodd.com
        • Eddy
          ... Thank you, Paul. More clear now. Did not know about the noise read, but sounds logic. Thanks, Eddy
          Message 4 of 22 , Oct 4, 2007
            --- In ccd-newastro@yahoogroups.com, "Paul K" <paul@...> wrote:
            >
            > Hi Eddy,
            >
            > Binning has one advantage over not binning: it reduces the read-out
            > noise of the camera. When taking filtered exposures of faint
            > objects, read-out noise can be the dominant noise in the image, so
            > it may be advantageous to bin RGB frames, especially in LRGB imaging.
            >
            > But, if your RGB exposure is long enough to overcome read-out
            > noise, or the object is bright enough to drown out this noise,
            > there's no good reason to bin (maybe just one: the images are
            > smaller and download faster <g>)
            >
            > If you bin 2x2, the image scale will be twice that of 1x1, i.e., one-
            > half the resolution. To register and combine binned images with
            > unbinned you can manually resize the image to 200%, or use software
            > that can do this for you automatically (CCDStack, Registar, etc.)
            >
            Thank you, Paul.
            More clear now. Did not know about the noise read, but sounds logic.

            Thanks,
            Eddy
          • Eddy
            ... the ... filters ... Thank you Mike. But why blur these frames? I am afraid I don t quite understand this. Wouldn t the image look nicer when they are not
            Message 5 of 22 , Oct 4, 2007
              --- In ccd-newastro@yahoogroups.com, Mike Dodd <mike_dodd@...> wrote:
              >
              > you are correct; binning 2x2 creates a "super pixel" with four times
              the
              > area of a single pixel, so it's four times more sensitive. Color
              filters
              > typically reduce the amount of light reaching the chip, so this
              > increased sensitivity avoids having to take extremely long RGB
              > exposures.the other benefit comets from using the color channels in
              > conjunction with am unbinned luminance channel. the luminance channel
              > provides the sharpness and detail, while the color channels don't have
              > to be so sharp. In fact,many people blur the color channels before
              > layering in Photoshop to form the LRGB image.
              >
              > You are correct about the image scale, but all that really affects is
              > the resolution, which isn't important if you're going to blur the RGB
              > channels anyway. the color frames are resized (doubled in each
              > dimension) before combining to match the luminance frame size.
              >
              Thank you Mike. But why blur these frames? I am afraid I don't quite
              understand this. Wouldn't the image look nicer when they are not
              blurred or have a better resolution.
              I never did this, so it is maybe easier to understand when combining a
              LRGB.

              Eddy
            • Wodaski - Yahoo
              That s not quite correct. Let me explain. When you bin 2x2, you are not reducing the read noise; you are increasing the signal relative to the read noise by
              Message 6 of 22 , Oct 4, 2007
                That's not quite correct. Let me explain.

                When you bin 2x2, you are not reducing the read noise; you are increasing
                the signal relative to the read noise by combining signal from multiple
                pixels. So you get more signal, but the same read noise (one does of read
                noise per read).

                Also, the benefit from this depends on the length of the exposure. If your
                individual exposure times are reasonably optimal (that is, the read noise is
                1/15th or less of the total shot noise), then there is little to no benefit
                from binning. SO the brighter your skies, the less improvement you will see
                from this technique. The darker your skies, and the shorter your exposure
                times, the more likely you are to see a benefit from binning.

                Ron Wodaski

                -----Original Message-----
                From: ccd-newastro@yahoogroups.com [mailto:ccd-newastro@yahoogroups.com] On
                Behalf Of Paul K
                Sent: Thursday, October 04, 2007 6:00 AM
                To: ccd-newastro@yahoogroups.com
                Subject: [ccd-newastro] Re: Binning question

                Hi Eddy,

                Binning has one advantage over not binning: it reduces the read-out
                noise of the camera. When taking filtered exposures of faint
                objects, read-out noise can be the dominant noise in the image, so
                it may be advantageous to bin RGB frames, especially in LRGB imaging.

                But, if your RGB exposure is long enough to overcome read-out
                noise, or the object is bright enough to drown out this noise,
                there's no good reason to bin (maybe just one: the images are
                smaller and download faster <g>)

                If you bin 2x2, the image scale will be twice that of 1x1, i.e., one-
                half the resolution. To register and combine binned images with
                unbinned you can manually resize the image to 200%, or use software
                that can do this for you automatically (CCDStack, Registar, etc.)

                Regards,

                -Paul


                --- In ccd-newastro@yahoogroups.com, "Eddy" <EddyMuriel@...> wrote:
                >
                > Hi all,
                >
                > Could somebody, in a few words, explain me why one would use
                binning?
                >
                > I see that some imagers are binning 2x2 the RGB channels but using
                1x1
                > for the luminance. Others are using 1x1 for all the channels. To
                me,
                > both ways look the same.
                >
                > I understand that binning 2x2 will reduce exposure times. But that
                is
                > the only "advantage" I am seeing.
                > When binning 2x2, the image scale will be different - if I
                understood
                > this correctly - than with 1x1.
                > How do you convert or stack these? Do you have to change the image
                > size before stacking?
                >
                > Thanks for any info,
                > Eddy
                >




                Yahoo! Groups Links
              • Wodaski - Yahoo
                ... It s all about the noise. Blurring reduces color noise. Blurring and binning are just two techniques you can use to allow shorter exposure times. This will
                Message 7 of 22 , Oct 4, 2007
                  >> Why blur these frames?

                  It's all about the noise. Blurring reduces color noise.

                  Blurring and binning are just two techniques you can use to allow shorter
                  exposure times. This will keep the color in the bright areas, but hide the
                  dominance of noise in the dimmer areas.

                  If you are willing to take sufficiently long exposures, you don't need to
                  use such tricks. So it's a trade-off: you can take shorter individual
                  exposures and less total exposure time, and then use some tricks to hide the
                  resulting noise. Or you can take lots of time to expose, and have better
                  color from the get-go.

                  This might seem trivial, but it takes hour and hours of exposure time to get
                  beyond the need for tricks! Not that many people are willing to take that
                  route because of the time required to do it.

                  Ron W

                  -----Original Message-----
                  From: ccd-newastro@yahoogroups.com [mailto:ccd-newastro@yahoogroups.com] On
                  Behalf Of Eddy
                  Sent: Thursday, October 04, 2007 9:44 AM
                  To: ccd-newastro@yahoogroups.com
                  Subject: [ccd-newastro] Re: Binning question

                  --- In ccd-newastro@yahoogroups.com, Mike Dodd <mike_dodd@...> wrote:
                  >
                  > you are correct; binning 2x2 creates a "super pixel" with four times
                  the
                  > area of a single pixel, so it's four times more sensitive. Color
                  filters
                  > typically reduce the amount of light reaching the chip, so this
                  > increased sensitivity avoids having to take extremely long RGB
                  > exposures.the other benefit comets from using the color channels in
                  > conjunction with am unbinned luminance channel. the luminance channel
                  > provides the sharpness and detail, while the color channels don't have
                  > to be so sharp. In fact,many people blur the color channels before
                  > layering in Photoshop to form the LRGB image.
                  >
                  > You are correct about the image scale, but all that really affects is
                  > the resolution, which isn't important if you're going to blur the RGB
                  > channels anyway. the color frames are resized (doubled in each
                  > dimension) before combining to match the luminance frame size.
                  >
                  Thank you Mike. But why blur these frames? I am afraid I don't quite
                  understand this. Wouldn't the image look nicer when they are not
                  blurred or have a better resolution.
                  I never did this, so it is maybe easier to understand when combining a
                  LRGB.

                  Eddy



                  Yahoo! Groups Links
                • Mike Dodd
                  ... with an LRGB image, the color channels are combined with the luminance channel using a color blending mode. In this case, the luminance channel provides
                  Message 8 of 22 , Oct 4, 2007
                    Eddy wrote:
                    >>
                    > Thank you Mike. But why blur these frames? I am afraid I don't quite
                    > understand this. Wouldn't the image look nicer when they are not
                    > blurred or have a better resolution.
                    > I never did this, so it is maybe easier to understand when combining a
                    > LRGB.


                    with an LRGB image, the color channels are combined with the luminance
                    channel using a "color" blending mode. In this case, the luminance
                    channel provides the details, and the color channels can be blurred. Try
                    it for yourself to see the color combine in action. It always brings a
                    smile to my face to see the fine color details appear after I select the
                    color blending mode, especially knowing that the color channels are
                    intentionally blurred.

                    --- Mike
                    --

                    Mike Dodd
                    Montpelier, VA USA
                    http://astronomy.mdodd.com
                  • Eddy
                    Ok, Thank you, Ron. I am understanding it now and I guess I have to try the technique so see for myself. I do have time (I am retired), so I could skip the
                    Message 9 of 22 , Oct 4, 2007
                      Ok, Thank you, Ron. I am understanding it now and I guess I have to
                      try the technique so see for myself.

                      I do have time (I am retired), so I could skip the binning technique,
                      but the main problem is that the weather is not always cooperative...
                      I guess this is also a reason for binning. Unless somebody could wait
                      a couple of months before having a nice picture.

                      Thank you,
                      Eddy
                    • Paul K
                      Sorry to disagree with you, Ron :) Binning does indeed reduce read noise: each binned super pixel is read only once, so the amount of read noise is halved for
                      Message 10 of 22 , Oct 4, 2007
                        Sorry to disagree with you, Ron :)

                        Binning does indeed reduce read noise: each binned super pixel is
                        read only once, so the amount of read noise is halved for the same
                        area of the sky when binned 2x2. When binned 1x1, four pixels are
                        read, resulting in 2x the amount of read noise (sqrt(4)).

                        The amount of signal does not increase when binning: the same number
                        of photons are received by the 2x2 super pixel as the total received
                        by the four underlying pixels.

                        Regards,

                        -Paul

                        --- In ccd-newastro@yahoogroups.com, "Wodaski - Yahoo" <yahoo@...>
                        wrote:
                        >
                        > That's not quite correct. Let me explain.
                        >
                        > When you bin 2x2, you are not reducing the read noise; you are
                        increasing
                        > the signal relative to the read noise by combining signal from
                        multiple
                        > pixels. So you get more signal, but the same read noise (one does
                        of read
                        > noise per read).
                        >
                        > Also, the benefit from this depends on the length of the exposure.
                        If your
                        > individual exposure times are reasonably optimal (that is, the
                        read noise is
                        > 1/15th or less of the total shot noise), then there is little to
                        no benefit
                        > from binning. SO the brighter your skies, the less improvement you
                        will see
                        > from this technique. The darker your skies, and the shorter your
                        exposure
                        > times, the more likely you are to see a benefit from binning.
                        >
                        > Ron Wodaski
                        >
                        > -----Original Message-----
                        > From: ccd-newastro@yahoogroups.com [mailto:ccd-
                        newastro@yahoogroups.com] On
                        > Behalf Of Paul K
                        > Sent: Thursday, October 04, 2007 6:00 AM
                        > To: ccd-newastro@yahoogroups.com
                        > Subject: [ccd-newastro] Re: Binning question
                        >
                        > Hi Eddy,
                        >
                        > Binning has one advantage over not binning: it reduces the read-
                        out
                        > noise of the camera. When taking filtered exposures of faint
                        > objects, read-out noise can be the dominant noise in the image, so
                        > it may be advantageous to bin RGB frames, especially in LRGB
                        imaging.
                        >
                        > But, if your RGB exposure is long enough to overcome read-out
                        > noise, or the object is bright enough to drown out this noise,
                        > there's no good reason to bin (maybe just one: the images are
                        > smaller and download faster <g>)
                        >
                        > If you bin 2x2, the image scale will be twice that of 1x1, i.e.,
                        one-
                        > half the resolution. To register and combine binned images with
                        > unbinned you can manually resize the image to 200%, or use
                        software
                        > that can do this for you automatically (CCDStack, Registar, etc.)
                        >
                        > Regards,
                        >
                        > -Paul
                        >
                        >
                        > --- In ccd-newastro@yahoogroups.com, "Eddy" <EddyMuriel@> wrote:
                        > >
                        > > Hi all,
                        > >
                        > > Could somebody, in a few words, explain me why one would use
                        > binning?
                        > >
                        > > I see that some imagers are binning 2x2 the RGB channels but
                        using
                        > 1x1
                        > > for the luminance. Others are using 1x1 for all the channels. To
                        > me,
                        > > both ways look the same.
                        > >
                        > > I understand that binning 2x2 will reduce exposure times. But
                        that
                        > is
                        > > the only "advantage" I am seeing.
                        > > When binning 2x2, the image scale will be different - if I
                        > understood
                        > > this correctly - than with 1x1.
                        > > How do you convert or stack these? Do you have to change the
                        image
                        > > size before stacking?
                        > >
                        > > Thanks for any info,
                        > > Eddy
                        > >
                        >
                        >
                        >
                        >
                        > Yahoo! Groups Links
                        >
                      • Wodaski - Yahoo
                        We are not really disagreeing. I didn t say there was more signal. I said there was more signal _relative to the read noise_. My point was that the read noise
                        Message 11 of 22 , Oct 4, 2007
                          We are not really disagreeing. I didn't say there was more signal. I said
                          there was more signal _relative to the read noise_.

                          My point was that the read noise is constant per read: you read once, you
                          get one dose of read noise. In that sense, the read noise does not change.
                          This is what I wanted to emphasize: that read noise is a constant, one dose
                          per read. Binning simply applies that noise to more than one pixel,
                          increasing S/N.

                          In the end, what is important with binning is that the _ratio_ of signal to
                          read noise is different.

                          Note that it is important to remember that this is only important if read
                          noise is a significant component of total noise. If you have long enough
                          sub-exposures, then by definition read noise is a small part of the total
                          noise and binning will make very little difference in the signal to noise
                          ratio. (Shot noise changes with the signal; shot noise is the square root of
                          the signal.)

                          Ron Wodaski

                          -----Original Message-----
                          From: ccd-newastro@yahoogroups.com [mailto:ccd-newastro@yahoogroups.com] On
                          Behalf Of Paul K
                          Sent: Thursday, October 04, 2007 10:08 AM
                          To: ccd-newastro@yahoogroups.com
                          Subject: [ccd-newastro] Re: Binning question

                          Sorry to disagree with you, Ron :)

                          Binning does indeed reduce read noise: each binned super pixel is
                          read only once, so the amount of read noise is halved for the same
                          area of the sky when binned 2x2. When binned 1x1, four pixels are
                          read, resulting in 2x the amount of read noise (sqrt(4)).

                          The amount of signal does not increase when binning: the same number
                          of photons are received by the 2x2 super pixel as the total received
                          by the four underlying pixels.

                          Regards,

                          -Paul

                          --- In ccd-newastro@yahoogroups.com, "Wodaski - Yahoo" <yahoo@...>
                          wrote:
                          >
                          > That's not quite correct. Let me explain.
                          >
                          > When you bin 2x2, you are not reducing the read noise; you are
                          increasing
                          > the signal relative to the read noise by combining signal from
                          multiple
                          > pixels. So you get more signal, but the same read noise (one does
                          of read
                          > noise per read).
                          >
                          > Also, the benefit from this depends on the length of the exposure.
                          If your
                          > individual exposure times are reasonably optimal (that is, the
                          read noise is
                          > 1/15th or less of the total shot noise), then there is little to
                          no benefit
                          > from binning. SO the brighter your skies, the less improvement you
                          will see
                          > from this technique. The darker your skies, and the shorter your
                          exposure
                          > times, the more likely you are to see a benefit from binning.
                          >
                          > Ron Wodaski
                          >
                          > -----Original Message-----
                          > From: ccd-newastro@yahoogroups.com [mailto:ccd-
                          newastro@yahoogroups.com] On
                          > Behalf Of Paul K
                          > Sent: Thursday, October 04, 2007 6:00 AM
                          > To: ccd-newastro@yahoogroups.com
                          > Subject: [ccd-newastro] Re: Binning question
                          >
                          > Hi Eddy,
                          >
                          > Binning has one advantage over not binning: it reduces the read-
                          out
                          > noise of the camera. When taking filtered exposures of faint
                          > objects, read-out noise can be the dominant noise in the image, so
                          > it may be advantageous to bin RGB frames, especially in LRGB
                          imaging.
                          >
                          > But, if your RGB exposure is long enough to overcome read-out
                          > noise, or the object is bright enough to drown out this noise,
                          > there's no good reason to bin (maybe just one: the images are
                          > smaller and download faster <g>)
                          >
                          > If you bin 2x2, the image scale will be twice that of 1x1, i.e.,
                          one-
                          > half the resolution. To register and combine binned images with
                          > unbinned you can manually resize the image to 200%, or use
                          software
                          > that can do this for you automatically (CCDStack, Registar, etc.)
                          >
                          > Regards,
                          >
                          > -Paul
                          >
                          >
                          > --- In ccd-newastro@yahoogroups.com, "Eddy" <EddyMuriel@> wrote:
                          > >
                          > > Hi all,
                          > >
                          > > Could somebody, in a few words, explain me why one would use
                          > binning?
                          > >
                          > > I see that some imagers are binning 2x2 the RGB channels but
                          using
                          > 1x1
                          > > for the luminance. Others are using 1x1 for all the channels. To
                          > me,
                          > > both ways look the same.
                          > >
                          > > I understand that binning 2x2 will reduce exposure times. But
                          that
                          > is
                          > > the only "advantage" I am seeing.
                          > > When binning 2x2, the image scale will be different - if I
                          > understood
                          > > this correctly - than with 1x1.
                          > > How do you convert or stack these? Do you have to change the
                          image
                          > > size before stacking?
                          > >
                          > > Thanks for any info,
                          > > Eddy
                          > >
                          >
                          >
                          >
                          >
                          > Yahoo! Groups Links
                          >




                          Yahoo! Groups Links
                        • Paul K
                          Ah, yes. I see that we are in agreement :) Read noise is generated at a constant level per each pixel, no matter whether it s binned or not. In the binned
                          Message 12 of 22 , Oct 4, 2007
                            Ah, yes. I see that we are in agreement :)

                            Read noise is generated at a constant level per each pixel, no
                            matter whether it's binned or not. In the binned case, there are a
                            lot fewer pixels being read, so the signal-to-noise ratio goes up
                            significantly. But, as you point out, this is only important in the
                            case where read noise is dominant (which happens more frequently
                            with filtered exposures).

                            If read noise is a problem, the solution is to either bin, or to
                            stack more of, or to take longer unbinned exposures to increase S/N.

                            Regards,

                            -Paul

                            --- In ccd-newastro@yahoogroups.com, "Wodaski - Yahoo" <yahoo@...>
                            wrote:
                            >
                            > We are not really disagreeing. I didn't say there was more signal.
                            I said
                            > there was more signal _relative to the read noise_.
                            >
                            > My point was that the read noise is constant per read: you read
                            once, you
                            > get one dose of read noise. In that sense, the read noise does not
                            change.
                            > This is what I wanted to emphasize: that read noise is a constant,
                            one dose
                            > per read. Binning simply applies that noise to more than one pixel,
                            > increasing S/N.
                            >
                            > In the end, what is important with binning is that the _ratio_ of
                            signal to
                            > read noise is different.
                            >
                            > Note that it is important to remember that this is only important
                            if read
                            > noise is a significant component of total noise. If you have long
                            enough
                            > sub-exposures, then by definition read noise is a small part of
                            the total
                            > noise and binning will make very little difference in the signal
                            to noise
                            > ratio. (Shot noise changes with the signal; shot noise is the
                            square root of
                            > the signal.)
                            >
                            > Ron Wodaski
                            >
                            > -----Original Message-----
                            > From: ccd-newastro@yahoogroups.com [mailto:ccd-
                            newastro@yahoogroups.com] On
                            > Behalf Of Paul K
                            > Sent: Thursday, October 04, 2007 10:08 AM
                            > To: ccd-newastro@yahoogroups.com
                            > Subject: [ccd-newastro] Re: Binning question
                            >
                            > Sorry to disagree with you, Ron :)
                            >
                            > Binning does indeed reduce read noise: each binned super pixel is
                            > read only once, so the amount of read noise is halved for the same
                            > area of the sky when binned 2x2. When binned 1x1, four pixels are
                            > read, resulting in 2x the amount of read noise (sqrt(4)).
                            >
                            > The amount of signal does not increase when binning: the same
                            number
                            > of photons are received by the 2x2 super pixel as the total
                            received
                            > by the four underlying pixels.
                            >
                            > Regards,
                            >
                            > -Paul
                            >
                            > --- In ccd-newastro@yahoogroups.com, "Wodaski - Yahoo" <yahoo@>
                            > wrote:
                            > >
                            > > That's not quite correct. Let me explain.
                            > >
                            > > When you bin 2x2, you are not reducing the read noise; you are
                            > increasing
                            > > the signal relative to the read noise by combining signal from
                            > multiple
                            > > pixels. So you get more signal, but the same read noise (one
                            does
                            > of read
                            > > noise per read).
                            > >
                            > > Also, the benefit from this depends on the length of the
                            exposure.
                            > If your
                            > > individual exposure times are reasonably optimal (that is, the
                            > read noise is
                            > > 1/15th or less of the total shot noise), then there is little to
                            > no benefit
                            > > from binning. SO the brighter your skies, the less improvement
                            you
                            > will see
                            > > from this technique. The darker your skies, and the shorter your
                            > exposure
                            > > times, the more likely you are to see a benefit from binning.
                            > >
                            > > Ron Wodaski
                            > >
                            > > -----Original Message-----
                            > > From: ccd-newastro@yahoogroups.com [mailto:ccd-
                            > newastro@yahoogroups.com] On
                            > > Behalf Of Paul K
                            > > Sent: Thursday, October 04, 2007 6:00 AM
                            > > To: ccd-newastro@yahoogroups.com
                            > > Subject: [ccd-newastro] Re: Binning question
                            > >
                            > > Hi Eddy,
                            > >
                            > > Binning has one advantage over not binning: it reduces the read-
                            > out
                            > > noise of the camera. When taking filtered exposures of faint
                            > > objects, read-out noise can be the dominant noise in the image,
                            so
                            > > it may be advantageous to bin RGB frames, especially in LRGB
                            > imaging.
                            > >
                            > > But, if your RGB exposure is long enough to overcome read-out
                            > > noise, or the object is bright enough to drown out this noise,
                            > > there's no good reason to bin (maybe just one: the images are
                            > > smaller and download faster <g>)
                            > >
                            > > If you bin 2x2, the image scale will be twice that of 1x1, i.e.,
                            > one-
                            > > half the resolution. To register and combine binned images with
                            > > unbinned you can manually resize the image to 200%, or use
                            > software
                            > > that can do this for you automatically (CCDStack, Registar, etc.)
                            > >
                            > > Regards,
                            > >
                            > > -Paul
                            > >
                            > >
                            > > --- In ccd-newastro@yahoogroups.com, "Eddy" <EddyMuriel@> wrote:
                            > > >
                            > > > Hi all,
                            > > >
                            > > > Could somebody, in a few words, explain me why one would use
                            > > binning?
                            > > >
                            > > > I see that some imagers are binning 2x2 the RGB channels but
                            > using
                            > > 1x1
                            > > > for the luminance. Others are using 1x1 for all the channels.
                            To
                            > > me,
                            > > > both ways look the same.
                            > > >
                            > > > I understand that binning 2x2 will reduce exposure times. But
                            > that
                            > > is
                            > > > the only "advantage" I am seeing.
                            > > > When binning 2x2, the image scale will be different - if I
                            > > understood
                            > > > this correctly - than with 1x1.
                            > > > How do you convert or stack these? Do you have to change the
                            > image
                            > > > size before stacking?
                            > > >
                            > > > Thanks for any info,
                            > > > Eddy
                            > > >
                            > >
                            > >
                            > >
                            > >
                            > > Yahoo! Groups Links
                            > >
                            >
                            >
                            >
                            >
                            > Yahoo! Groups Links
                            >
                          • Mike Dodd
                            ... I recently tried taking the color frames unbinned; the total exposure was _very_long, and the color subs were quite noisy. I ve since decided that unbinned
                            Message 13 of 22 , Oct 4, 2007
                              Eddy wrote:
                              > I do have time (I am retired), so I could skip the binning technique,
                              > but the main problem is that the weather is not always cooperative...

                              I recently tried taking the color frames unbinned; the total exposure
                              was _very_long, and the color subs were quite noisy. I've since
                              decided that unbinned color isn't a good choice for me.

                              --- Mike

                              --

                              Mike Dodd
                              Montpelier, VA USA
                              http://astronomy.mdodd.com
                            • Randy Nulman
                              Hi All, Also note that even with TV transmissions, the color is blurred with the monochrome data providing the detail. I don t know how, or if, this has
                              Message 14 of 22 , Oct 4, 2007
                                Hi All,

                                Also note that even with TV transmissions, the color is "blurred"
                                with the monochrome data providing the detail. I don't know how, or
                                if, this has changed with HD signals...but this was the norm.

                                Randy Nulman
                                http://www.nulman.darkhorizons.org


                                --- In ccd-newastro@yahoogroups.com, Mike Dodd <mike_dodd@...> wrote:
                                >
                                > Eddy wrote:
                                > > I do have time (I am retired), so I could skip the binning
                                technique,
                                > > but the main problem is that the weather is not always
                                cooperative...
                                >
                                > I recently tried taking the color frames unbinned; the total
                                exposure
                                > was _very_long, and the color subs were quite noisy. I've since
                                > decided that unbinned color isn't a good choice for me.
                                >
                                > --- Mike
                                >
                                > --
                                >
                                > Mike Dodd
                                > Montpelier, VA USA
                                > http://astronomy.mdodd.com
                                >
                              • wt5l_jon
                                Hi Eddy, Most of the advantages of binning have been mentioned. One of the disadvantages is the need to capture and keep a separate set of calibration frames
                                Message 15 of 22 , Oct 4, 2007
                                  Hi Eddy,

                                  Most of the advantages of binning have been mentioned. One of the
                                  disadvantages is the need to capture and keep a separate set of
                                  calibration frames (darks) for each binning that you use. Usually
                                  this is not considered a problem since these can be captured during
                                  poor imaging conditions and then reused. Just another level of
                                  complexity...

                                  Jon


                                  --- In ccd-newastro@yahoogroups.com, "Eddy" <EddyMuriel@...> wrote:
                                  >
                                  > Hi all,
                                  >
                                  > Could somebody, in a few words, explain me why one would use binning?
                                  >
                                  > I see that some imagers are binning 2x2 the RGB channels but using 1x1
                                  > for the luminance. Others are using 1x1 for all the channels. To me,
                                  > both ways look the same.
                                  >
                                  > I understand that binning 2x2 will reduce exposure times. But that is
                                  > the only "advantage" I am seeing.
                                  > When binning 2x2, the image scale will be different - if I understood
                                  > this correctly - than with 1x1.
                                  > How do you convert or stack these? Do you have to change the image
                                  > size before stacking?
                                  >
                                  > Thanks for any info,
                                  > Eddy
                                  >
                                • sc02492
                                  Eddy, I m coming in on this discussion fashionably late . Here is a little write up on binning that I put together on my website last year:
                                  Message 16 of 22 , Oct 4, 2007
                                    Eddy, I'm coming in on this discussion fashionably late <g>. Here is
                                    a little write up on binning that I put together on my website last year:

                                    http://www.starrywonders.com/binning.html

                                    If you don't like math (not really as complicated as it looks), then
                                    just scroll down to the "bottom line" summary.


                                    Steve

                                    Steve Cannistra
                                    http://www.starrywonders.com

                                    --- In ccd-newastro@yahoogroups.com, "Eddy" <EddyMuriel@...> wrote:
                                    >
                                    > Hi all,
                                    >
                                    > Could somebody, in a few words, explain me why one would use binning?
                                    >
                                    > I see that some imagers are binning 2x2 the RGB channels but using 1x1
                                    > for the luminance. Others are using 1x1 for all the channels. To me,
                                    > both ways look the same.
                                    >
                                    > I understand that binning 2x2 will reduce exposure times. But that is
                                    > the only "advantage" I am seeing.
                                    > When binning 2x2, the image scale will be different - if I understood
                                    > this correctly - than with 1x1.
                                    > How do you convert or stack these? Do you have to change the image
                                    > size before stacking?
                                    >
                                    > Thanks for any info,
                                    > Eddy
                                    >
                                  • Neil Fleming
                                    I tried binning a few times, and never liked the result. Too big a compromise on the overall result. I know that many swear by this approach, though. ...
                                    Message 17 of 22 , Oct 4, 2007
                                      I tried binning a few times, and never liked the
                                      result. Too big a compromise on the overall result.
                                      I know that many swear by this approach, though.

                                      ...Neil

                                      --- Eddy <EddyMuriel@...> wrote:

                                      > Ok, Thank you, Ron. I am understanding it now and I
                                      > guess I have to
                                      > try the technique so see for myself.
                                      >
                                      > I do have time (I am retired), so I could skip the
                                      > binning technique,
                                      > but the main problem is that the weather is not
                                      > always cooperative...
                                      > I guess this is also a reason for binning. Unless
                                      > somebody could wait
                                      > a couple of months before having a nice picture.
                                      >
                                      > Thank you,
                                      > Eddy
                                      >
                                      >


                                      www.flemingastrophotography.com
                                      Direct from Boston - brilliant diamonds in pea soup
                                    • Eddy
                                      Thank you all for your replies. Steve, nice write up. Thanks! Jon mentioned you need a new set of darks, but if I understood everything up to now, you don t
                                      Message 18 of 22 , Oct 5, 2007
                                        Thank you all for your replies.

                                        Steve, nice write up. Thanks!

                                        Jon mentioned you need a new set of darks, but if I understood
                                        everything up to now, you don't need to take new flats when binning,
                                        right? The image train remains the same and binning doesn't change
                                        anything on this, correct?

                                        Eddy
                                      • wt5l_jon
                                        Eddy, Correct, you only need the dark and bias (if you use them) frames for each binning you use. You can take one set of unbinned flats to serve all binned
                                        Message 19 of 22 , Oct 5, 2007
                                          Eddy,

                                          Correct, you only need the dark and bias (if you use them) frames for
                                          each binning you use. You can take one set of unbinned flats to serve
                                          all binned or unbinned images.

                                          Jon

                                          --- In ccd-newastro@yahoogroups.com, "Eddy" <EddyMuriel@...> wrote:
                                          >
                                          > Thank you all for your replies.
                                          >
                                          > Steve, nice write up. Thanks!
                                          >
                                          > Jon mentioned you need a new set of darks, but if I understood
                                          > everything up to now, you don't need to take new flats when binning,
                                          > right? The image train remains the same and binning doesn't change
                                          > anything on this, correct?
                                          >
                                          > Eddy
                                          >
                                        • Eddy
                                          ... Thank you for confirming this, Jon. Eddy
                                          Message 20 of 22 , Oct 5, 2007
                                            --- In ccd-newastro@yahoogroups.com, "wt5l_jon" <jonolson@...> wrote:
                                            >
                                            > Eddy,
                                            >
                                            > Correct, you only need the dark and bias (if you use them) frames for
                                            > each binning you use. You can take one set of unbinned flats to serve
                                            > all binned or unbinned images.
                                            >
                                            Thank you for confirming this, Jon.

                                            Eddy
                                          • Donald Waid
                                            Eddy, My understanding is you can use the unbinned flats to reduce binned frames but you have to rescale the flats to the binned size first. If I am
                                            Message 21 of 22 , Oct 5, 2007
                                              Eddy,

                                              My understanding is you can use the unbinned flats to reduce binned frames
                                              but you have to "rescale" the flats to the binned size first. If I am
                                              incorrect on this, someone please let me know.

                                              Don Waid

                                              ----- Original Message -----
                                              From: "Eddy" <EddyMuriel@...>
                                              To: <ccd-newastro@yahoogroups.com>
                                              Sent: Friday, October 05, 2007 4:13 AM
                                              Subject: [ccd-newastro] Re: Binning question


                                              > Thank you all for your replies.
                                              >
                                              > Steve, nice write up. Thanks!
                                              >
                                              > Jon mentioned you need a new set of darks, but if I understood
                                              > everything up to now, you don't need to take new flats when binning,
                                              > right? The image train remains the same and binning doesn't change
                                              > anything on this, correct?
                                              >
                                              > Eddy
                                              >
                                              >
                                              >
                                              > Yahoo! Groups Links
                                              >
                                              >
                                              >
                                            • Wodaski - Yahoo
                                              I ve always had good success by re-sizing my 1x1 flats for binned images. You usually need to change the bin setting in the FITS header when you do this,
                                              Message 22 of 22 , Oct 5, 2007
                                                I've always had good success by re-sizing my 1x1 flats for binned images.

                                                You usually need to change the bin setting in the FITS header when you do
                                                this, unless the software you are using (e.g., MaxIm DL) will do it for you.
                                                (Otherwise, the flat will not be recognized for what it now is by the
                                                automated features of the calibration process in said software.)

                                                Ron W


                                                -----Original Message-----
                                                From: ccd-newastro@yahoogroups.com [mailto:ccd-newastro@yahoogroups.com] On
                                                Behalf Of Eddy
                                                Sent: Friday, October 05, 2007 2:14 AM
                                                To: ccd-newastro@yahoogroups.com
                                                Subject: [ccd-newastro] Re: Binning question

                                                Thank you all for your replies.

                                                Steve, nice write up. Thanks!

                                                Jon mentioned you need a new set of darks, but if I understood
                                                everything up to now, you don't need to take new flats when binning,
                                                right? The image train remains the same and binning doesn't change
                                                anything on this, correct?

                                                Eddy



                                                Yahoo! Groups Links
                                              Your message has been successfully submitted and would be delivered to recipients shortly.