why no true high resolution systems for "jetstream" seeing?

While it's true that webcams are doing wonders for certain types of seeing,
in many areas of the US during the Wintertime, the jetstream is constantly
overhead and even a webcam cannot undo this blurring. I keep reading about
these new adaptive optics scopes, like the new pro scope used for solar
imaging, which also incorporate speckle imaging and reconstruction- why
nothing for amateurs? Couldn't blurring of frames (caused by high
jetstreams) be deblurred or "reconstructed" so blurring is minimized. With
computers and the power they possess these days, I'm surprised there aren't
programs that can do this or maybe jetstream effects still can't be truly
nullified.

Thanks,
Frank

On Sat, 07 Jan 2006 16:16:24 GMT, "Frank Johnson"
wrote:

While it's true that webcams are doing wonders for certain types of seeing,
in many areas of the US during the Wintertime, the jetstream is constantly
overhead and even a webcam cannot undo this blurring. I keep reading about
these new adaptive optics scopes, like the new pro scope used for solar
imaging, which also incorporate speckle imaging and reconstruction- why
nothing for amateurs? Couldn't blurring of frames (caused by high
jetstreams) be deblurred or "reconstructed" so blurring is minimized. With
computers and the power they possess these days, I'm surprised there aren't
programs that can do this or maybe jetstream effects still can't be truly
nullified.

It simply isn't possible to do much along these lines with amateur
systems. Adaptive optics requires a reference object very close to the
target, and in almost all cases that object (if one exists at all) is
going to be dim. For this reference to be useful, you need to collect
enough photons to get out of the noise. Since detectors are already
nearly 100% efficient, the only way to get enough light is with a large
aperture. Adaptive optics starts becoming practical as apertures get in
the range of a meter, and larger is better. Obviously, this is well
outside what most would consider amateur optics.

The lack of available reference stars is another problem for amateurs.
Professional observatories can address this by making their own, using
lasers to excite atoms high in the atmosphere. But aside from cost,
shining such lasers at the sky requires special permits that pretty much
preclude their use by amateurs.

Finally, adaptive optics wouldn't provide much benefit to most amateurs
because it can only correct the field over a few arcseconds- much
smaller than typical observing or imaging fields. It is more a
scientific tool for examining very small zones at high resolution than
something useful for improving the aesthetics of an image or view.

Adaptive optics would really only be useful for most amateurs for
observing (or imaging) very small (or small areas) and very bright
objects- the Sun, Moon, and brighter planets. Right now, adaptive optics
are expensive enough to produce that there probably isn't enough market
for so limited a device to justify development. That might change in the
future, but don't expect to see anything useful for deep sky observing-
ever.

What may come around, assuming that low readout noise cameras become
generally available, is software that selects high quality short frames
and stacks them automatically. This is done already with planets, of
course, but doesn't work with dim objects. Unfortunately, as the size of
the field increases, the number of useable high resolution frames
decreases. It might take several hours to collect a few minutes worth of
data.

Deconvolution techniques can help to reconstruct distorted images, but
there is no certain way to determine just where any given photon came
from. Deconvolution is a sort of educated guess, but it frequently
produces invalid data- and there is no getting around that problem.

_________________________________________________

Chris L Peterson
Cloudbait Observatory
http://www.cloudbait.com

"Chris L Peterson" wrote in message
...
On Sat, 07 Jan 2006 16:16:24 GMT, "Frank Johnson"

What may come around, assuming that low readout noise cameras become
generally available, is software that selects high quality short frames
and stacks them automatically. This is done already with planets, of
course, but doesn't work with dim objects. Unfortunately, as the size of
the field increases, the number of useable high resolution frames
decreases. It might take several hours to collect a few minutes worth of
data.


The professionals are catching up with the amateurs in this area

http://news.bbc.co.uk/1/hi/sci/tech/4456988.stm

Robin

On Sat, 7 Jan 2006 18:08:37 -0000, "Robin Leadbeater"
wrote:

The professionals are catching up with the amateurs in this area

http://news.bbc.co.uk/1/hi/sci/tech/4456988.stm

Thanks for the link. "Lucky imaging"; great name! Sounds like a good
amateur project, looking for previously unknown doubles. Perfect for
this technique because the sources are bright, and the field is small.

_________________________________________________

Chris L Peterson
Cloudbait Observatory
http://www.cloudbait.com

"Chris L Peterson" wrote in message
...
On Sat, 07 Jan 2006 16:16:24 GMT, "Frank Johnson"
wrote:

While it's true that webcams are doing wonders for certain types of
seeing,
in many areas of the US during the Wintertime, the jetstream is
constantly
overhead and even a webcam cannot undo this blurring. I keep reading
about
these new adaptive optics scopes, like the new pro scope used for solar
imaging, which also incorporate speckle imaging and reconstruction- why
nothing for amateurs? Couldn't blurring of frames (caused by high
jetstreams) be deblurred or "reconstructed" so blurring is minimized.
With
computers and the power they possess these days, I'm surprised there
aren't
programs that can do this or maybe jetstream effects still can't be truly
nullified.

It simply isn't possible to do much along these lines with amateur
systems. Adaptive optics requires a reference object very close to the
target, and in almost all cases that object (if one exists at all) is
going to be dim. For this reference to be useful, you need to collect
enough photons to get out of the noise. Since detectors are already
nearly 100% efficient, the only way to get enough light is with a large
aperture. Adaptive optics starts becoming practical as apertures get in
the range of a meter, and larger is better. Obviously, this is well
outside what most would consider amateur optics.

The lack of available reference stars is another problem for amateurs.
Professional observatories can address this by making their own, using
lasers to excite atoms high in the atmosphere. But aside from cost,
shining such lasers at the sky requires special permits that pretty much
preclude their use by amateurs.

Finally, adaptive optics wouldn't provide much benefit to most amateurs
because it can only correct the field over a few arcseconds- much
smaller than typical observing or imaging fields. It is more a
scientific tool for examining very small zones at high resolution than
something useful for improving the aesthetics of an image or view.

Adaptive optics would really only be useful for most amateurs for
observing (or imaging) very small (or small areas) and very bright
objects- the Sun, Moon, and brighter planets. Right now, adaptive optics
are expensive enough to produce that there probably isn't enough market
for so limited a device to justify development. That might change in the
future, but don't expect to see anything useful for deep sky observing-
ever.

What may come around, assuming that low readout noise cameras become
generally available, is software that selects high quality short frames
and stacks them automatically. This is done already with planets, of
course, but doesn't work with dim objects. Unfortunately, as the size of
the field increases, the number of useable high resolution frames
decreases. It might take several hours to collect a few minutes worth of
data.

Deconvolution techniques can help to reconstruct distorted images, but
there is no certain way to determine just where any given photon came
from. Deconvolution is a sort of educated guess, but it frequently
produces invalid data- and there is no getting around that problem.


Hi, yes, I already pretty much understand the current techniques involving
adaptive optics. I myself have experimented on Jupiter by using one of its
moons as a PSF. Then I tried to apply max ent to Jupiter only to end up
with a very noisy and artifact prone result. As you said, there simply
might have not been enough signal to work with.

There *is* a phenomenon I've noticed with my setup on nights when seeing is
plagued with an overhead jetream: because my tracking isn't perfect, I make
small drive corrections during capture to keep a chosen planet centered in
the field. Sometimes when I've adjusted the tracking, there will be moments
while the drive corrector is engaged where the planet I'm imaging suddenly
becomes clear. When I let go of the corrector, the blurs caused by the
overhead jet return but for a few moments there were clear frames. One
might suspect a sort of vibration as the reason, but I have noticed this
with several different mount/ scope combinations and I think it has
something to do with maybe "catching up with" or "matching" jetstream winds
for a moment (a second) while the planet is being moved back to the center
of field.

I still search for a CCD camera with very high sensitivity, rapid frame
captures and very low noise. The webcam is ok, but falls short in the noise
and sensitivity department. Commercial CCD cams, although much less noisy,
can't capture nearly as many frames per time as a webcam and are much more
expensive. I recently read here about meteor CCD cameras possibly offering
the best signal to noise ratios and sensitivity, but it's tough finding
anyone currently using those.

Frank


_________________________________________________

Chris L Peterson
Cloudbait Observatory
http://www.cloudbait.com

If you're willing to pay the hefty price there are plenty of cameras on
the market.

Andrea T.

Frank Johnson wrote:
"Chris L Peterson" wrote in message
...

On Sat, 07 Jan 2006 16:16:24 GMT, "Frank Johnson"
wrote:

While it's true that webcams are doing wonders for certain types of
seeing,
in many areas of the US during the Wintertime, the jetstream is
constantly
overhead and even a webcam cannot undo this blurring.

The problem here is the scale length of the jet stream turbulence is
shorter and it is moving past your aperture a lot faster. In principle
if you could freeze the seeing you would eventually get some lucky
frames but the problem is that a webcam isn't fast enough to do it.

these new adaptive optics scopes, like the new pro scope used for solar
imaging, which also incorporate speckle imaging and reconstruction- why
nothing for amateurs? Couldn't blurring of frames (caused by high
jetstreams) be deblurred or "reconstructed" so blurring is minimized.

Speckle is only realistic on apertures of 0.5m and above and requires
narrow bandpass filters to do it. It should be within the abilities of
amateurs to do now on bright stars but I don't know of anyone doing it.

programs that can do this or maybe jetstream effects still can't be truly
nullified.

It is better to get good data in the first place.

Deconvolution techniques can help to reconstruct distorted images, but
there is no certain way to determine just where any given photon came
from. Deconvolution is a sort of educated guess, but it frequently
produces invalid data- and there is no getting around that problem.

It is probably better to say that deconvolution attempts to produces a
representation of what the sky brightness looked like before it was
convolved with your instrumental response function. It is valid in the
sense that what it produces when blurred would look like your
observational data to within the measurement noise. But there are many
such images that meet these requirements and choosing a suitably
represenative one is fraught with difficulty.

Serious problems will arise if you ask the wrong question of your data.

Even in a perfect world these deconvolution codes produce images with
artefacts that are different and unfamiliar and that can be a problem in
eg medical diagnosis.

Hi, yes, I already pretty much understand the current techniques involving
adaptive optics. I myself have experimented on Jupiter by using one of its
moons as a PSF. Then I tried to apply max ent to Jupiter only to end up
with a very noisy and artifact prone result. As you said, there simply
might have not been enough signal to work with.

Signal to noise was probably OK. The thing that would most likely have
caused trouble is that the Galilean moons are not unresolved objects in
any reasonable sized scope. You set a the program problem that had no
reasonable answer.

You need the instrumental psf determined from a bright field star (and
it is rare to have one) or modelled from theory. Any errors in the psf
determination are amplified by any of the deconvolution methods.

Deconvolving with a broader psf than the true instrumental response will
amplify the high frequency noise and also cause excessive ringing on
sharp edges.

Regards,
Martin Brown

On Sat, 07 Jan 2006 16:46:24 GMT, Chris L Peterson wrote:

It simply isn't possible to do much along these lines with amateur
systems. Adaptive optics requires a reference object very close to the
target, and in almost all cases that object (if one exists at all) is
going to be dim. For this reference to be useful, you need to collect
enough photons to get out of the noise. Since detectors are already
nearly 100% efficient, the only way to get enough light is with a large
aperture. Adaptive optics starts becoming practical as apertures get in
the range of a meter, and larger is better. Obviously, this is well
outside what most would consider amateur optics.

Hi Chris,

Most of your original reply is accurate and useful, but your comment on the
aperture size follows a common misconception. Large apertures do not
collect more photons for an adaptive optics system; the only photons that
count are those in a subaperture about the diameter of r0, typically 20 cm.
Larger aperture telescopes require more subapertures. The only way to get
more photons in the subaperture are 1) wait for good seeing, when you can
get by with larger subapertures; 2) go to longer IR wavelengths, even to 10
microns, where r0 is always larger than visible light; 3) use a laser guide
star. (Longer integration periods in good seeing also help collect more
photons.) Solution 1) is not useful for a world-class observatory, since
every night is too valuable to waste; solution 2) is used by some
observatories, but then the resolution lambda/D is not very good; solution
3) is pretty expensive, but its getting better, and is probably the
ultimate solution.

The amatuer solution is to look at only very bright, very small targets.
There are very few of these, limiting the market.

I recall a few years ago, that some observers were able to detect craters
on Mercury, on a side that was not imaged by the space probes, by using a
million short exposure images, and looking for the higher resolution "lucky
shots". So, there is opportunity for useful work in this area.

Don Bruns

Chris L Peterson wrote:
It simply isn't possible to do much along these lines with amateur
systems. Adaptive optics requires a reference object very close to the
target, and in almost all cases that object (if one exists at all) is
going to be dim. For this reference to be useful, you need to collect
enough photons to get out of the noise. Since detectors are already
nearly 100% efficient, the only way to get enough light is with a large
aperture. Adaptive optics starts becoming practical as apertures get in
the range of a meter, and larger is better. Obviously, this is well
outside what most would consider amateur optics.

Wouldn't you need good coherence at one-meter scales for that? Isn't
the Fried parameter shorter than that most of the time?

--
Brian Tung
The Astronomy Corner at http://astro.isi.edu/
Unofficial C5+ Home Page at http://astro.isi.edu/c5plus/
The PleiadAtlas Home Page at http://astro.isi.edu/pleiadatlas/
My Own Personal FAQ (SAA) at http://astro.isi.edu/reference/faq.html

On Sat, 7 Jan 2006 19:21:20 -0800 (PST), (Brian Tung)
wrote:

Wouldn't you need good coherence at one-meter scales for that? Isn't
the Fried parameter shorter than that most of the time?

True, as also noted by Don. So you really just need a bright reference-
no problem if you have a nice dye laser pointing back up along the
optical path (and permission from the FAA to use it). Optically, there
are advantages to using a large primary mirror to generate the guide
star, although not everyone is doing it that way.

_________________________________________________

Chris L Peterson
Cloudbait Observatory
http://www.cloudbait.com