Go Back   IceInSpace > Equipment > Astrophotography and Imaging Equipment and Discussions

Reply
 
Thread Tools Rate Thread
  #1  
Old 11-12-2012, 04:43 PM
John K's Avatar
John K
Registered User

John K is offline
 
Join Date: Sep 2005
Location: Melbourne
Posts: 1,468
Latest processing planetary imaging techniques

Hi everyone,

One of the things that I have enjoyed over the years about the IIS community is how everyone has been able to share their knowledge in an open on-line forumn to assist everyone to move forward with better planetary imaging.

In the last few weeks I was visited by someone from O/S who came to AUS to view the eclipse and we got talking about the latest processing techniques on plantetary imaging and for example how guys like Damien Peach is still able to produce amazing photos from low altitude planetary objects.

Have not seen a huge amount of discussion about the latest methods used by everyone to push some boundaries in their planteray processing and thought it worthwhile to revisit this topic.

Other than now using Autostacker2 and starting some research on WinJupos de-rotation, I have to admit my planetary processing has remained the same for several years. Capture with Lucam Recorder -> Stack with Autostakker2 (use to be registax)- > Wavelets in Registax 6 -> Deconvolution in Astra Image - > RGB combining in Astra Image Touching up in photoshop.

Obviously WinJupos derotation is adding another dimension, but it appears that this is still evolving and there is limited info on the best methods there.

Thoughts?? Comments?? Methods??

John K.
Reply With Quote
  #2  
Old 11-12-2012, 10:20 PM
asimov's Avatar
asimov (John)
Planet photographer

asimov is offline
 
Join Date: May 2005
Location: Bundaberg
Posts: 8,819
Because I've been using the Imaging Source OSC cams (DBK) my processing routine has remained the same for years until relatively recently with the advent of Autostakkert & it's superior algorithm(s) on how my data was getting debayered. My biggest stumbling block for years, was being stuck with inferior algorithms (nearest neighbour) in Debayering capable programs (Registax & VirtualDub) to be precise.

Capture with an old version of IC Capture > Optimise/Debayer/Stack with Autostakkert > Waveletts etc. in Registax Ver. 6.1 > Post processing in Paint.net > Nothing else.

I no longer use any kind of deconvolution in programs like Astraimage & I can't be bothered learning how to use WinJupos because quite frankly, I'm not seeing much of a benefit compared to the results I'm already getting without using it. What HAS upped the anti for me is the 618 chip & like I said, the way the data gets Debayered.
Reply With Quote
  #3  
Old 12-12-2012, 08:19 PM
John K's Avatar
John K
Registered User

John K is offline
 
Join Date: Sep 2005
Location: Melbourne
Posts: 1,468
Thanks for the info Asi,

Seems that either everyone is keeping their processing close to their chest or that we in OZ are not following the WinJupos de-rotation trend that we are seeing in Cloudy Nights and other forumns.

I still beleive that good seeing still is the key! although a highly sensetive camera that can crank 120fps would also help.

But I also believe that guys like Damien Peach have a few processing tricks that they keep close to their chest as well!
Reply With Quote
  #4  
Old 12-12-2012, 08:57 PM
asimov's Avatar
asimov (John)
Planet photographer

asimov is offline
 
Join Date: May 2005
Location: Bundaberg
Posts: 8,819
I also think that seeing is the major ingredient John, although I really want to try IR imaging with a mono camera & capture colour data at the same time. I can't actually do that without forking out for another C11 + mount but what I think is feasible is to have both cams setup in a flip mirror & take 2 mins of IR & 1 min of colour data..I guess I could also do that via the mono cam & the filter wheel but it might be a bit too hectic to shoot through 4 filters in 3 mins.

I purchased an Atmospheric dispersion corrector only a few weeks ago but I'm not seeing much of an improvement in my images considering I'm getting 45 degrees of alt. from Jupiter this season from Bundaberg QLD.

Damian is certainly doing something that I'm not aware of lol.
Reply With Quote
  #5  
Old 19-12-2012, 06:13 PM
Shiraz's Avatar
Shiraz (Ray)
Registered User

Shiraz is offline
 
Join Date: Apr 2010
Location: ardrossan south australia
Posts: 4,918
Hi John

missed this one somehow - sorry.

First off, have to agree with Asi - seeing is the key to any form of success and will often determine the best processing path. I generally drive 130k to my imaging site and have to be able to predict when the seeing will be good. I need a breeze off the sea combined with 200hPa winds of less than about 30kts to have any chance of really good seeing. If the wind is not off the sea (or there is no wind) I get tearing or rolling local distortion - if the jetstream is too strong the fine detail goes and the overall contrast often drops drastically. Local distortion can sometimes be stacked out - especially with AS!2 - but nothing can restore lost fine detail.

Equipment is a 300mm f5 Newt with 3x Barlow or 5x Powermate. Astronomik filters and possibly a dispersion corrector for use below about 40deg. Biggest payback from the corrector is for Saturn where it allows a broadband lum image to be obtained at high framerate. This can help a lot in some seeing. Also good for tidying up the blue channel with Jupiter. General observation would be that the corrector really only helps a lot when the seeing is good - which it almost never is at low altitude.

I use either Firecapture or ICCapture for the DMK21618 camera. Normally capture for 60 seconds per channel at 60hz for Jupiter and 100 sec 30/15Hz for Saturn, but occasionally will vary scale if seeing suggests that shorter exposures might help. Firecapture will not reliably run at 60Hz (drops about 10-20% of frames on my i7) so use ICCap when running at 60. Jupiter is better at 30Hz if seeing will allow - lower frame rate reduces read noise .


processing is:
1. castrator to centre up images
2. stack in AS!2 normally - maybe 40-50 points on Jupiter and perhaps 25 on Saturn. If the seeing works against AS!2, use R6 with an intermediate stack to generate a reference image from a single stack point and then use multiple stack points on the reference - this gets around the distortion that can occur with R6. also, R6 runs best if a prefilter is used to smooth noise - I use a scale of 4.
3. sharpen with AS!2 plus additional sharpening in R6 with linked wavelets used as sparingly as possible. The noise limiting capability of R6 is worth having. An alternative if enough frames have been stackable is to use multiple scales of VanCittert deconvolution in IRIS with median and Gaussian filters to control noise. Have found it important to use the same sharpening in all colour channels to keep final colours looking reasonable - different smoothing may be applied though.
4. colour combine RGB or LRGB in WINJUPOS using derotation of the stacked frames. this really helps tidy up the colours and has been a major advance in processing RGB. If there is still some residual mismatch between the colour channels, I will split the stacked image in IRIS and recombine with manual adjustment to remove the slight misalignment. If there is any residual noise on the planet edge, will use annular select in GIMP with feathered edge and reduce the saturation or blur around the edge. I have tried both combining multiple colour frames and video derotation in WINJUPOS , but with limited success to date.
5. tidy up the image in GIMP. Colour balance using the "grey point" add in to set a white point and then compare with a standard image (eg from Cassini, Pic du Midi or Trevor) to get the final balance and saturation as close as possible to "right". I try to use gamma of 1 throughout, unless the seeing has been so poor that contrast is almost gone - then will sometimes use modified curves to compensate. also sometimes use a very light adaptive smoothing which only removes noise in regions where there is little contrast - needs to be used very sparingly if at all.

would be very interested in any discussion. regards ray

Last edited by Shiraz; 20-12-2012 at 09:12 PM.
Reply With Quote
  #6  
Old 20-12-2012, 09:15 AM
Paul Haese's Avatar
Paul Haese
Registered User

Paul Haese is offline
 
Join Date: Jan 2009
Location: Adelaide
Posts: 9,937
I'll disagree with John (Asi) here and say outright WinJupos is the next evolution in processing. I have seen so many great images processed with it of late that I am now convinced it adds another several orders of magnitude to the final image. Damian has been using this program for two years now and I have had some recent discussions with him about its use. He has fine tuned it to produce the images he gets when Jupiter is only 50 degrees above the horizon. Chris go's images of late are a quantum leap upward too. Good seeing really helps when you use this program. I am still trying to work out where it all works but like I said convinced this is the next step.

I no longer use Registax and prefer AS!2. Still do deconvolution but in minor amounts for fine detail. My Saturns in the last two years have been with this techhnique. Mars last apparition too. I no longer image from Adelaide hills and instead prefer to image from Clayton where the seeing is much better.

I have an ADC on loan from a mate but yet to use it. Need to give this a try. Damian tells me it is very good at low elevations, but needs to be tuned exceptionally well. Bird has said the same thing. The ADC is probably best used when doing LRGB rather than RGB. The L like DSO imaging becomes the detail layer but it will be blurred if uncorrected for dispersion.

Next step for me other than processing is size of scope. I see resolving power as a necessity. Still working out which scope diameter and type of scope I want.
Reply With Quote
  #7  
Old 20-12-2012, 10:58 AM
asimov's Avatar
asimov (John)
Planet photographer

asimov is offline
 
Join Date: May 2005
Location: Bundaberg
Posts: 8,819
Well I never said WinJupos doesn't work Paul, I just said I'm not seeing any great leaps & bounds in Damians (or anyone elses) images derived from it. Elevation of target, not sure what that has to do with it (Winjupos) - That's where the ADC comes in. I would agree about the better the seeing, the better the noticable quality with WinJupos - Precisely why I haven't bothered. I've not even had a 7/10 or better this season.

I'll continue to do my thing & the gurus can do theirs I guess
Reply With Quote
  #8  
Old 21-12-2012, 09:16 AM
John K's Avatar
John K
Registered User

John K is offline
 
Join Date: Sep 2005
Location: Melbourne
Posts: 1,468
Good discussion guys, keep it going!

I personally think it’s important to push the boundaries at every level including software, equipment, processing etc. I cannot wait for more sensitive higher frame rate cameras (would love for someone to develop something that can do 200 fps!) and fasters PC’s that allow more data captures and better software to process everything. The ADC evolution will also be interesting to watch.

Don’t think there is a right and wrong answer. All new things can work and can also fail. It all depends on a range of things such as location, conditions, instruments etc etc. The important thing is for us as an astronomical community to continuously share the learnings and keep moving forward and I think in particular Aussie amateurs do this very well with forums such as IIS.

Really like the comments about Jupos De-rotation and it would be good for someone to post a really good tutorial on this. Combined with an ADC unit and a higher frame rate camera (I am still using a Skynix 2.0) it all sounds quite positive.

Also, I agree with the comments that seeing is still king, but the question becomes how to get reasonably good images even at low elevations or average seeing. In my location, dreaded Melbourne, I may only get 2-3 days of exceptional seeing per year if that (at least when I am imaging) so it’s a pot luck scenario.

Merry Xmas to all.

John K.
Reply With Quote
  #9  
Old 21-12-2012, 09:24 AM
iceman's Avatar
iceman (Mike)
Sir Post a Lot!

iceman is offline
 
Join Date: Sep 2004
Location: Gosford, NSW, Australia
Posts: 36,760
I'm so out of touch with planetary imaging and processing techniques now. I haven't taken a planetary image in like 3 years

One day I hope to get back into it. It'll be a whole new learning curve for me.
Reply With Quote
  #10  
Old 21-12-2012, 07:05 PM
Shiraz's Avatar
Shiraz (Ray)
Registered User

Shiraz is offline
 
Join Date: Apr 2010
Location: ardrossan south australia
Posts: 4,918
"I cannot wait for more sensitive higher frame rate cameras"

http://learn.hamamatsu.com/articles/emccds.html

now that QE is approaching 100% and thermal noise is small, there is only one niche left to make major advances in CCD performance - read noise. Interesting to see if these devices reach the low end market.

If you haven't come across it, Piotr Malinski did a good summary of how to use WINJUPOS http://www.rkblog.rk.edu.pl/w/p/plan...tion-winjupos/

regards Ray

Last edited by Shiraz; 22-12-2012 at 04:21 PM.
Reply With Quote
  #11  
Old 23-12-2012, 11:23 AM
Shiraz's Avatar
Shiraz (Ray)
Registered User

Shiraz is offline
 
Join Date: Apr 2010
Location: ardrossan south australia
Posts: 4,918
Hi John. Further info re the new EMCCD cameras

the Andor EMCCD cameras are cooled to -95C

Edmond have a relatively low spec EMCCD camera in stock for $8500.

the gain mechanism slowly wears out and pixels also die in significant numbers as you use an EMCCD.

Unless you use photon counting, the additional noise from the amplification mechanism reduces the SNR and the gain mechanism also limits the effective well depth - if you have a gain of 100x, the effective well depth (in photon input terms) reduces to 1/100 its no-gain level.

So it's not all good news and there might be a bit of a wait for the new tech cameras to reach "hobby" levels of price and reliability - sigh. However, they should allow us to apply lucky imaging techniques to some DSOs as well as the dimmer solar system objects - that will be a great new area of entertainment for amateurs.

regards Ray

Last edited by Shiraz; 23-12-2012 at 03:26 PM.
Reply With Quote
  #12  
Old 24-12-2012, 12:41 PM
alpal's Avatar
alpal
Registered User

alpal is offline
 
Join Date: Jan 2012
Location: Melbourne
Posts: 3,602
After about 10 emails I finally got the prices for EMCCD's.

They cost about $32,000.

They have only a super small sensor size.

The cameras are very heavy & would not be easy to mount.

Still - they are the next obvious step in lucky imaging at high frame rates.
I expect we'll start seeing some results in about 10 years
because it will take that long before amateurs can afford them.
Reply With Quote
Reply

Bookmarks

Thread Tools
Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +10. The time is now 10:58 PM.

Powered by vBulletin Version 3.8.7 | Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Advertisement
Testar
Advertisement
Bintel
Advertisement