Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

The future of Astrophotography?

  • Please log in to reply
28 replies to this topic

#1 Startex

Startex

    Vostok 1

  • -----
  • topic starter
  • Posts: 130
  • Joined: 23 Dec 2018
  • Loc: Texas

Posted 16 October 2019 - 03:18 AM

So I was thinking how much tech has changed in the last 20 years and I got to thinking what AP technology will be like in another 20-30-50 years.

 

Could we have Laser guide star and adaptive optics in our tool bags in the future?

 

https://en.wikipedia...aser_guide_star       

 

Any other ideas?


Edited by Startex, 16 October 2019 - 03:19 AM.


#2 Astrojensen

Astrojensen

    Voyager 1

  • *****
  • Posts: 12316
  • Joined: 05 Oct 2008
  • Loc: Bornholm, Denmark

Posted 16 October 2019 - 03:41 AM

I don't think we will ever see amateur operated laser guidance systems, as the lasers need to be REALLY powerful and amateurs are generally living in or near big cities with airports, etc. It's very unlikely that governments will want to see such powerful lasers operated near airports... Adaptive optics for amateurs thus need to work on a different principle. 

 

What we WILL see is further developement of sensors, possibly completely new technologies. Maybe even some that can work as adaptive optics, to some degree. Sensitivity will no doubt continue to improve and this will allow lucky imaging of ever fainter objects. Time will tell. Software will also improve, of course, as will mounts. Even more advanced optical systems with huge, diffraction limited fields will be introduced, to take advantage of the new, gigantic sensors that will come. We may need quantum computing to be able to handle the extremely massive amounts of data.  

 

"Telescope farms" in faraway countries with good weather will become ever more popular and prices will drop. Maybe there will even be an amateur space telescope that you can rent time on?

 

The most exciting things will be the ones we can't even imagine. 

 

 

Clear skies!
Thomas, Denmark


Edited by Astrojensen, 16 October 2019 - 03:41 AM.

  • dswtan, Augustus, Flatha and 1 other like this

#3 sg6

sg6

    Fly Me to the Moon

  • *****
  • Posts: 6108
  • Joined: 14 Feb 2010
  • Loc: Norfolk, UK.

Posted 16 October 2019 - 03:44 AM

I may well be wrong but I have an idea that adaptive optics is applicable to large scopes and mirrors, each individual mirror being in effect bigger then the general sizes we use. Also guess that you need multiple mirrors to accomplish - in effect you cannot adapt a single unit.

 

Is the principal that you alter 1 mirror in a set of 7 to get a better result. 7 example arranged as a central hexagonal and 6 copies around it.

 

Whether or not modification to smaller mirrors can be feasible (assumes I am correct and no guarantee) would be the question.

 

Also I would offer that the advance is in software more then anything. Other then the step from film to digital, but what would be a replacement for digital? So computing power becomes key. A laser "star" likely needs a seperate camera and computer to analyse the "star", compute the perturbations and make changes to each mirror.



#4 gionk

gionk

    Sputnik

  • -----
  • Posts: 45
  • Joined: 03 Mar 2019

Posted 16 October 2019 - 03:58 AM

I guess from a digital standpoint, given the necessary speed of the scope and brightness of the objects, DSO lucky imaging might evolve too. Capturing thousands of images with high sensitivity / low noise cameras and speeds faster than f4, I guess it's already possible to freeze seeing and then compare frames to evaluate stability, even in fragments within the image. Rating those fragments / sectors and stacking based on dropout and weighting, one could achieve much higher resolution. This is all done already for planetary imaging, but for DSO the limiting factor have been sensors and missing digital software optimized for DSO. However, I can imagine this will change in the near future.



#5 OleCuss

OleCuss

    Mercury-Atlas

  • *****
  • Posts: 2525
  • Joined: 22 Nov 2010

Posted 16 October 2019 - 04:38 AM

I agree that there will likely be further sensor technology improvement.  The thing is, however, that when we are already getting low-noise CMOS cameras with a peak QE in the range of 80% I'm not sure it will be as dramatically better as I might hope.

 

But we're already about to get a tech which will likely allow a pixel to detect saturation and re-set itself to zero while counting the number of times it has re-set.  This could allow a dramatic increase in dynamic range.  Maybe something with 1.5 micron pixels but putting out true 16-bit data?  You could have a short focal length system without under-sampling but great QE, low noise, and huge dynamic range!  I believe the tech has already been demonstrated but I'm not sure how close it is to the market (might already be available somewhere).

 

 

We might also get large, lightweight panels which can substitute for standard reflective and refractive optics.  Too early to tell for sure but I remember seeing something a few years ago with some promise in this regard although it was not even close to being practical for our use at that time.  It's be amazing to have a 2-meter panel in an observatory!  I'm not counting on it ever happening, though.  Even if it becomes technically possible and even if the panel is cheap, mounting it would likely still be difficult and expensive just due to the size and the need to prevent flexing.

 

 

The most likely change will also be a change in what amateur astronomy is.  The Vaonis Stellina, the Unistellar eVscope, and the Hiuni may eventually make Observational AP accessible to a whole new group of people who will have become a type of amateur astronomers.

 

There may also be increasing attempts to utilize/link amateur gear to perform scientific observations when needed by the pros.



#6 james7ca

james7ca

    Fly Me to the Moon

  • *****
  • Posts: 7494
  • Joined: 21 May 2011
  • Loc: San Diego, CA

Posted 16 October 2019 - 05:51 AM

One thing to note, today's CCD and CMOS cameras are NOT fully digital. They are analog-to-digital devices, not digital from start to finish. However, there is work being done to bring fully digital sensors to the consumer market, which could eventually work their way over to amateur astrophotography. The so-called JOT sensor is such a device (mostly, I think, although we may not know for sure until they actually ship a product).

 

I also agree with Astrojensen, we're not likely to see any amateur-based adaptive optics based upon lasers.

 

What we will see, however, is the likely end of the premium mount (at least for the vast majority of users). What I mean is that there will be no real need for accurate guiding, since we'll all be using lucky imaging techniques with photon-counting sensors that have practically no read noise or other signal artifacts. When we have essentially unlimited storage and bandwidth and computing power then ten thousand times 100ms will be just as good as a single one thousand second sub (or better, since you could align and grade each 100ms sub to create the final image and as you combine those individual subs you could output any bit depth that you want). In fact, the initial JOT sensor was a 1-bit device (it either counted a photon hit or not and it just accumulates the count over time).


  • happylimpet, Astrola72, entilza and 1 other like this

#7 spokeshave

spokeshave

    Vanguard

  • *****
  • Posts: 2149
  • Joined: 08 Apr 2015

Posted 16 October 2019 - 06:48 AM

My sense is that most of the advancements in the near future will be evolutionary, not revolutionary. I think robotic mounts will become more prevalent and less expensive, potentially obviating the need for guiding altogether. There are very encouraging advancements in lens/mirror additive manufacturing, with the potential for relatively inexpensive manufacture of higher-order optics. That has the potential to lead to flatter, better-corrected fields in astrographs at lower cost. 

 

Sensors can't get a whole lot better than the best sensors available now. QE's are pretty high and read noise is pretty low. There is still room for improvement, but the improvements will be marginal. I think the biggest improvements will be in features such as "infinite" well depth, faster reads, better binning (for CMOS sensors), etc. One thing I would really like to see is for someone to adapt the Foveon sensor technology to be more suited to astrophotography. Such a sensor with modern low noise architecture would be a huge boon - it would allow the collection of LRGB simultaneously with a single sensor.

 

I see a lot of advances in image processing particularly in the areas of artificial intelligence and machine learning. These can be applied to noise reduction and signal extraction and have great potential.

 

Tim


  • Olhado and psandelle like this

#8 TOMDEY

TOMDEY

    Aurora

  • *****
  • Posts: 4940
  • Joined: 10 Feb 2014
  • Loc: Springwater, NY

Posted 16 October 2019 - 07:28 AM

My sense is that most of the advancements in the near future will be evolutionary, not revolutionary. I think robotic mounts will become more prevalent and less expensive, potentially obviating the need for guiding altogether. There are very encouraging advancements in lens/mirror additive manufacturing, with the potential for relatively inexpensive manufacture of higher-order optics. That has the potential to lead to flatter, better-corrected fields in astrographs at lower cost. 

 

Sensors can't get a whole lot better than the best sensors available now. QE's are pretty high and read noise is pretty low. There is still room for improvement, but the improvements will be marginal. I think the biggest improvements will be in features such as "infinite" well depth, faster reads, better binning (for CMOS sensors), etc. One thing I would really like to see is for someone to adapt the Foveon sensor technology to be more suited to astrophotography. Such a sensor with modern low noise architecture would be a huge boon - it would allow the collection of LRGB simultaneously with a single sensor.

 

I see a lot of advances in image processing particularly in the areas of artificial intelligence and machine learning. These can be applied to noise reduction and signal extraction and have great potential.

 

Tim

I'd say your perspective is the most realistic, Tim. Some of the other stuff... flat telescopes, remote observing farms, big adaptive, wide field, cheap do everything scopes etc. etc. are more on the side of wishful thinking. And the attitude that unnamed experts will develop and successfully market all this stuff and just gift it all to us amateurs. Mostly pipe dreams. But it's fun to imagine. If you really want innovation... the old-fashioned path to actualization is to... dive in and participate, become one of the experts! No free lunch.   Tom



#9 gionk

gionk

    Sputnik

  • -----
  • Posts: 45
  • Joined: 03 Mar 2019

Posted 16 October 2019 - 07:42 AM

Don't forget the increase in artificial satellites orbiting earth and we might end up throwing 50% of our subs into the trash :-)))


  • Fernando134 and Dynan like this

#10 kingjamez

kingjamez

    Vanguard

  • *****
  • Posts: 2160
  • Joined: 03 Oct 2006
  • Loc: Fairfax, VA

Posted 16 October 2019 - 07:51 AM

Don't forget the increase in artificial satellites orbiting earth and we might end up throwing 50% of our subs into the trash :-)))

Nah... our AI processing software will just remove them.
  • sayitfast and OldManSky like this

#11 dhaval

dhaval

    Vendor

  • *****
  • Vendors
  • Posts: 1749
  • Joined: 21 Jul 2008
  • Loc: Round Rock, TX

Posted 16 October 2019 - 08:19 AM

I imagine the most bang for buck will be in software - automating a lot of the mundane and even AI/ML may be something that we start to see in image processing. At least, I am hoping we do.

 

CS! 



#12 Astrojensen

Astrojensen

    Voyager 1

  • *****
  • Posts: 12316
  • Joined: 05 Oct 2008
  • Loc: Bornholm, Denmark

Posted 16 October 2019 - 08:37 AM

 Some of the other stuff... flat telescopes, remote observing farms, big adaptive, wide field, cheap do everything scopes etc. etc. are more on the side of wishful thinking.

But there is already quite a few of these in operation, some operating as many as a dozen amateur telescopes. It's already reality for many, not wishful thinking. And with ever increasing light pollution and poor weather conditions, due to global warming, I think it's not unlikely that they will become ever more popular with imagers. Especially since imaging is also getting increasingly popular, it seems. A friend of mine already buys a lot of time online on big telescopes, rather than observe with his own. He said that if he had spend all the money, he spend on his 10" reflector on online time instead, he could have logged many hundreds of hours with a 20" scope under much darker skies than from home with his 10". If you live under poor skies, it makes a lot of sense economically. 

 

https://www.itelescope.net/

 

https://www.deepskychile.com/en/

 

https://www.spaceobs.com/en

 

https://gloria-project.eu/en/

 

http://sierrastars.com/

 

https://www.virtualtelescope.eu/

 

https://www.insighto.../home-page.html

 

 

Clear skies!
Thomas, Denmark


  • dswtan likes this

#13 Pauls72

Pauls72

    Mercury-Atlas

  • *****
  • Posts: 2586
  • Joined: 28 Oct 2007
  • Loc: LaPorte, IN USA

Posted 16 October 2019 - 08:43 AM

On the professional level for about 15 years now they have been working with MKID sensors (Microwave Kinetic Inductance Detector). These sensors can detect the color (frequency) and intensity (charge level) of each photon that strikes each sensor cell with no bayer matrix or any other filters. They work on a super wide range of frequencies from the far-infrared to x-ray.  The down side of these sensors is they need to be super cooled to around 1° Kelvin  and the sensor cells are rather large 120um.

 

https://en.wikipedia...ctance_detector

http://www.astro.cal..._aug2016_v1.pdf


  • kingjamez, 2ghouls and gionk like this

#14 charotarguy

charotarguy

    Apollo

  • *****
  • Posts: 1272
  • Joined: 10 Sep 2014

Posted 16 October 2019 - 09:06 AM

Driverless cars, so we can get our much needed sleep while driving to a dark site.


Edited by charotarguy, 16 October 2019 - 09:07 AM.

  • Stelios, Olhado, psandelle and 3 others like this

#15 gionk

gionk

    Sputnik

  • -----
  • Posts: 45
  • Joined: 03 Mar 2019

Posted 16 October 2019 - 09:29 AM

On the professional level for about 15 years now they have been working with MKID sensors (Microwave Kinetic Inductance Detector). These sensors can detect the color (frequency) and intensity (charge level) of each photon that strikes each sensor cell with no bayer matrix or any other filters. They work on a super wide range of frequencies from the far-infrared to x-ray.  The down side of these sensors is they need to be super cooled to around 1° Kelvin  and the sensor cells are rather large 120um.

 

https://en.wikipedia...ctance_detector

http://www.astro.cal..._aug2016_v1.pdf

Well, cooling to 1K will be no issue with our portable pocket fusion reactors... :-)) Okay, I need to stop now...


  • psandelle and Fernando134 like this

#16 dpaigen

dpaigen

    Explorer 1

  • -----
  • Posts: 52
  • Joined: 30 Jul 2019

Posted 16 October 2019 - 10:00 AM

It isn't too hard to imagine 20 years out (but really hard to get it right) but 50 years is just an exercise in fantasy.  Nevertheless...

 

We are already seeing technology trends today towards thinner, flatter lenses; small run and individualized manufacturing; sensor improvements in nearly every aspect; increased computing power at the scope and machine learning models.

 

20 years from now I can picture a high end AP rig consisting of a 500mm f/2.8 refractor massing less than 5 kg.  90% of the mount is 3d printed with 24 bit encoders and pointing accuracy.  The imaging sensor has 90% QE, essentially zero dark noise, unlimited dynamic range, and can measure the frequency of arriving photons.  I'm not sure DSOs put out enough photons for lucky imaging, but guide scope cams are sensitive enough to do adaptive optics from a bright guide star.  ML figures out the contrasts and shapes and delivers images with effective resolution of 0.1 arc-second.

 

(BTW, adaptive optics do not require a multi mirror primary nor do they deform or adjust the primary.  AO puts a flat, thin mirror in the light path that can be deformed by 1000's of actuators afixed to the back)

 

I also think 20 years will see increased urbanization resulting in more dark sky areas and more robot telescope farms.  50 years might even see amateurs renting time on space based robot telescope farms.

 

ON THE OTHER HAND, the economics of scale won't change.  The rig I imagine will likely cost twice as much in constant currency as an advanced rig does today.


  • psandelle likes this

#17 TOMDEY

TOMDEY

    Aurora

  • *****
  • Posts: 4940
  • Joined: 10 Feb 2014
  • Loc: Springwater, NY

Posted 16 October 2019 - 10:33 AM

On the professional level for about 15 years now they have been working with MKID sensors (Microwave Kinetic Inductance Detector). These sensors can detect the color (frequency) and intensity (charge level) of each photon that strikes each sensor cell with no bayer matrix or any other filters. They work on a super wide range of frequencies from the far-infrared to x-ray.  The down side of these sensors is they need to be super cooled to around 1° Kelvin  and the sensor cells are rather large 120um.

 

https://en.wikipedia...ctance_detector

http://www.astro.cal..._aug2016_v1.pdf

That's just ~Popular Science~ mumbo jumbo... will never become useful. Related... to this day, cars still use inflated rubber tires --- because nothing else actually works! And a hundred years from now... people will still be dragging their 60mm Christmas Refractors out into the back yard. But, I know... it's still fun to imagine and believe how wonderful things will be down the pike, with just a tad more development.   Tom


Edited by TOMDEY, 16 October 2019 - 10:33 AM.

  • Fernando134 likes this

#18 Pauls72

Pauls72

    Mercury-Atlas

  • *****
  • Posts: 2586
  • Joined: 28 Oct 2007
  • Loc: LaPorte, IN USA

Posted 16 October 2019 - 03:41 PM

That's just ~Popular Science~ mumbo jumbo... will never become useful. Related... to this day, cars still use inflated rubber tires --- because nothing else actually works! And a hundred years from now... people will still be dragging their 60mm Christmas Refractors out into the back yard. But, I know... it's still fun to imagine and believe how wonderful things will be down the pike, with just a tad more development.   Tom

One of the clubs I belong to had one of the engineers from Fermilab's who is working on MKID do a presentation on it. It is not mumbo jumbo. It is real science in development and it keeps getting better all the time. Will we see it on the amateur level, probably not. But we will see more of it on the professional level and on telescopes in space.

 

You need to read this article on new tire technology too.

https://www.cnet.com...m-future-tires/


  • kingjamez likes this

#19 TOMDEY

TOMDEY

    Aurora

  • *****
  • Posts: 4940
  • Joined: 10 Feb 2014
  • Loc: Springwater, NY

Posted 16 October 2019 - 05:03 PM

One of the clubs I belong to had one of the engineers from Fermilab's who is working on MKID do a presentation on it. It is not mumbo jumbo. It is real science in development and it keeps getting better all the time. Will we see it on the amateur level, probably not. But we will see more of it on the professional level and on telescopes in space.

 

You need to read this article on new tire technology too.

https://www.cnet.com...m-future-tires/

I'm jaded... "in development" has a way of never making it into utilitarian production --- most often... ever... as in never. But, I do concede... still worth trying!    Tom, aka doubting Thomas



#20 kingjamez

kingjamez

    Vanguard

  • *****
  • Posts: 2160
  • Joined: 03 Oct 2006
  • Loc: Fairfax, VA

Posted 16 October 2019 - 05:11 PM

I'm jaded... "in development" has a way of never making it into utilitarian production --- most often... ever... as in never. But, I do concede... still worth trying!    Tom, aka doubting Thomas

You know that CCD cameras started in exactly the same manner right?



#21 TOMDEY

TOMDEY

    Aurora

  • *****
  • Posts: 4940
  • Joined: 10 Feb 2014
  • Loc: Springwater, NY

Posted 16 October 2019 - 05:40 PM

You know that CCD cameras started in exactly the same manner right?

Ummm... no! CCD cameras were always a sure thing! Just pack a lot of reliable-technology radiometers next to each other and turn it on. The rest was just details. The thing that kills this universal photon qunatum dingus is --- that little 1oK requirement! It's the difference between strolling up a cool, mild grade and scaling frigid Mt. Everest! Both can be done... one is easy... the other takes heroic effort, and the moment you get there... you beat fuzzy feet outa there!

 

Reminds me of this old cig commercial; I quit smoking 45 years ago!  Tom

Attached Thumbnails

  • 151 camel cigarette walk a mile.jpg


#22 kingjamez

kingjamez

    Vanguard

  • *****
  • Posts: 2160
  • Joined: 03 Oct 2006
  • Loc: Fairfax, VA

Posted 16 October 2019 - 05:52 PM

You are right. Technology never improves or finds solutions to hard problems. Yep, never gonna happen.

#23 OldManSky

OldManSky

    Vanguard

  • *****
  • Posts: 2094
  • Joined: 03 Jan 2019
  • Loc: Valley Center, CA USA

Posted 16 October 2019 - 06:05 PM

From a slightly different perspective, maybe a look *back* may give some clues to the way forward...

 

I was heavily into imaging from the late 90's to 2008.  Had all the expensive toys, was getting images published, etc.  Then the big recession hit, I wound up having to sell all my stuff, and unfortunately got completely out of astronomy for 10 years.  Got back to a point where I could get back in just this past December.

 

So what changed in those 10 (nearly 11) years?

 

Guiding.  In 2008, StarlightXpress had their first "mini" guide cameras (1.25" eyepiece size), but they were CCD, they were noisy, and they were expensive.  The ST-4 was still in wide use.  Only the high-end SBIG cameras had a second chip integrated in the camera body for built-in "off-axis" guiding.  Guiding software was MaximDL or...not much else.  10/11 years later, PHD2 is free and widely available, CMOS eyepiece-sized guide cameras are cheap and plentiful, mini guide-scopes are readily available and inexpensive.  The idea of guiding a 1000mm FL scope with a little 50mm f/3.6 guidescope in 2008 was ludicrous...now it's common place.  Guiding is SO much simpler/easier/cheaper now!

 

CMOS cameras.  No cooled CMOS cameras existed in 2008.  The 6MP, APS-C sized CCD camera I used at the time (StarlightXpress) was over $3k in 2008 dollars.  The 16MP Kodak chip wasn't available.  The vast majority of people shot with chips similar in size to my current 183MM chip, but with vastly less resolution and at a much higher cost.  Since then, CCDs had a better run (with bigger, higher-resolution chips coming out), but then CMOS came along and turned all of that upside down -- in cost, resolution, read noise, etc.  

 

Mounts.  Oh, yes, there were inexpensive GOTO mounts in 2008.  They were total crap for the most part (especially for imaging).  I know, I owned a couple of them!  When I got back in and saw the iOptron CEM25P mount for $900, with its payload and features and guaranteed < 10" PE, I could hardly believe it.  Other mounts have evolved spectacularly/had new models come out as well.  And what is perhaps the most impressive evolution...an AP mount doesn't cost much more than it did 10/11 years ago, it's as if AP completely ignored inflation!  Amazing!

 

SGP/NINA/Voyager didn't exist then.  'Nuff said. 

 

There are others, but I think that makes the point.  There have been a few "revolutionary" things occur, but most of the changes have been better gear at lower prices, better software, and better use of already existing technology.  I suspect the next 10-20 years will likely be quite similar :)


  • psandelle, zxx and TOMDEY like this

#24 TOMDEY

TOMDEY

    Aurora

  • *****
  • Posts: 4940
  • Joined: 10 Feb 2014
  • Loc: Springwater, NY

Posted 17 October 2019 - 07:03 AM

You are right. Technology never improves or finds solutions to hard problems. Yep, never gonna happen.

It's the learned skill to choose one's battles wisely. As in knowing the subtle difference between optimism and pipe-dream. Else wind up just another Don Quixote, taking on windmills... and puzzled why we don't seem able to make any progress... ever! Many of these wishful initiatives are what Richard Feynman called ~Cargo-Cult Technologies~ That is, looks like progress, smells like progress... but never quite gets to the point of taking off... always needs just a bit more tweaking... because fatally-flawed. A commercial camera running at 1oK is most certainly in that ill-fated league.    Tom

Attached Thumbnails

  • 155 don quixote tilting windmills.jpg


#25 klaussius

klaussius

    Explorer 1

  • -----
  • Posts: 90
  • Joined: 02 Apr 2019
  • Loc: Buenos Aires

Posted 17 October 2019 - 08:46 AM

Computerized finder and polar scopes, with built-in plate solving and AR HUD.

 

Best thing about it, is that it's doable.




CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics