#1  
Old 29-01-2019, 11:13 AM
Startrek (Martin)
Registered User

Startrek is offline
 
Join Date: Dec 2017
Location: Sydney and South Coast NSW
Posts: 5,990
Dithering in BYEOS

I've been imaging with my DSLR and 6" newt for 8 months now without dithering and just relying on darks and flats etc for noise reduction


Should I go down that path or stick with using dark frames etc...



Ive heard the term "Dither or Die" mentioned on some forums when it comes to DSLR astrophotography


Any advice or comments regarding dithering most appreciated



Cheers
Reply With Quote
  #2  
Old 29-01-2019, 11:25 AM
JA
.....

JA is offline
 
Join Date: Oct 2016
Location: Melbourne, Australia
Posts: 2,945
Hello Martin,

Take a look at this:


The whole presentation is great, but the dithering discussion starts at 5:48, which is hopefully where the link I've added will start, otherwise move it to there or watch the lot.

Best
JA
Reply With Quote
  #3  
Old 29-01-2019, 02:03 PM
Startrek (Martin)
Registered User

Startrek is offline
 
Join Date: Dec 2017
Location: Sydney and South Coast NSW
Posts: 5,990
Hi JA
Thanks for replying
Yes Ive watched the Tony Hallas video a few times but apparently he was incorrect on a number of points as mentioned by Ivo Jager ( Creator of Startools ) on his Startools forum in 2014


Quote / un quotes



1/ "Colour Mottle" is not a physical phenonenon or an attribute of a CCD chip. Colour Mottle is introduced into the image by the use of Adobe Camera Raw and its debayering algorithm and noise reduction filters


2/ Dithering is only effective in reducing "fixed Pattern noise" . Tony's colour mottle is simply the result of transformed ( By Debayering ) random shot noise , the colour mottle is never in the same place , so dithering isnt required to fix that
3/ Tony solution to use dithering is just a roundabout way of fixing up his initial error of debayering his frames using Adobe camera raw
4/ Tonys pre processing starts messing with his data in Adobe camera raw which is already debayered and colour balanced, which increases noise,causing his colour mottle
5/ Then he applies chromatic aberration making the data useless for deconvolution

6/ He doing everything on a per frame basis which is worse than if you stacked the image resulting in a higher SNR and smarter noise reduction
7/ He mentions his data is 16 bit but in fact it is 14 bit because it hasn't been stacked yet
8/ At this stage he basically shooting in Jpeg



Ivo Jager final message on this video is Tony is doing a lot of budding astronomers a great disservice here !


JA
I'm still a bit lost with whether to dither or not , basically I prefer to listen and take notice of IIS members ( maybe yourself ) who have had first hand experience with dithering using their DSLRs in the telescope and comparing before and after results in various conditions ( Seeing conditions, Higher outside air temperatures which is the enemy of DSLR's , various degees of guiding error etc..... )


Welcome any further advice and comments on dithering using DSLR's



Cheers
Reply With Quote
  #4  
Old 29-01-2019, 04:56 PM
The_bluester's Avatar
The_bluester (Paul)
Registered User

The_bluester is offline
 
Join Date: Feb 2011
Location: Kilmore, Australia
Posts: 3,342
If you have hot pixels that either produce maximum pixel values or close to it, dark frames will result in "Measles" on your images as the dark frame subtraction will result in more or less zero values in those pixels, regardless of what actual signal might really be there.

I fiddled for some while with a 350D letting it do in camera darks (Lazy me at the time) and I only ever noticed the issue when I started using my C925 and guiding as well instead of doing widefield images. When dark frames were subtracted either in camera or using properly shot darks turned into a master dark and guiding was introduced the hot pixels would fall in the same place on the image target sub after sub after sub, so the dark frames resulted in any real information in those pixels being lost to the subtraction of the hot pixels.

If you introduce dithering, the hot pixels will be subtracted from different pixels within the target in the light frames so then integration can work better, the zero or low values of the subtracted hot pixels should end up rejected and the true signal will be more dominant.
Reply With Quote
  #5  
Old 30-01-2019, 08:15 PM
Startrek (Martin)
Registered User

Startrek is offline
 
Join Date: Dec 2017
Location: Sydney and South Coast NSW
Posts: 5,990
Thanks for replying


Ive seen some forums where the Dither settings on BYEOS are -



Dither Aggressiveness 2


Settle Dither at 0.25


Calm down period 25Dither every X image 1


Some people have said dithering can add between 45 sceonds and 2 minutes between subs ??


Is that correct ??


Seems like a time consuming process compared to taking darks at the end of your imaging session while your asleep in bed


Welcome some feedback on the above


Thanks in advance
Reply With Quote
  #6  
Old 30-01-2019, 08:47 PM
The_bluester's Avatar
The_bluester (Paul)
Registered User

The_bluester is offline
 
Join Date: Feb 2011
Location: Kilmore, Australia
Posts: 3,342
I never used dithering via BYE but now using the software APT (Astro Photography Tool) and PHD2 for guiding I reckon it results in 15 to 20 seconds added between subs while the dither is done and guiding settles again. APT may treat it differently to BYE, it monitors the guide star and when that is falling within a specified number of pixels of the target it assumes the dither is complete and moves on.

That sounds like BYE simply waits for a defined time to assume the dither is complete.

Darks you should not need to shoot every night if you have a good dark library built up at different iso settings and temperatures.
Reply With Quote
  #7  
Old 31-01-2019, 07:51 AM
Startrek (Martin)
Registered User

Startrek is offline
 
Join Date: Dec 2017
Location: Sydney and South Coast NSW
Posts: 5,990
Paul
I think BYEOS works in a similar fashion where a Dither is sent as a random number of pixels in X and Y set under Aggressiveness
Then there is a Settle parameter which specifies how far off the star is from the lock position before the Dither is considered successful
Then a Calm down period which is a set number of seconds between dithering to allow PHD2 to stabilise

The weather on the NSW mid to south east coast during December and January has been ordinary this summer for AP , cloudy , very very hot , thunderstorms and very few clear nights ( only had 3 or 4 nights imaging in just over 2 months ) typical summer

Will see what February brings and hopefully can try some Dithering on BYEOS and PHD2

One question , in PHD2 ( Brain- Global tab ) should I set dithering to random or spiral ??? Any other settings in PHD2 I should be aware of ???

Thanks
Cheers
Reply With Quote
  #8  
Old 31-01-2019, 04:12 PM
The_bluester's Avatar
The_bluester (Paul)
Registered User

The_bluester is offline
 
Join Date: Feb 2011
Location: Kilmore, Australia
Posts: 3,342
I am at the start of this process so I am not too sure about PHD settings, I did use random dithering rather than spiral. I suspect that one will come down to personal preference.


The dithering distance and scale I left alone, but if you find that you still get spots from hot pixel subtraction it might be worth increasing the dither distance.
Reply With Quote
  #9  
Old 31-01-2019, 05:15 PM
Startrek (Martin)
Registered User

Startrek is offline
 
Join Date: Dec 2017
Location: Sydney and South Coast NSW
Posts: 5,990
I found some information on a old PHD google forum this afternoon which explains a bit about dithering
Random dithering was the first process to be developed and is used extensively but Spiral was developed later to assist with mounts that have DEC backlash issues
The Spiral Dither doesn’t not reverse back on itself like the Random Dither which helps with Dec backlash issues in guiding.The movement starts from the centre (guide star )and moves out and around rather than the zig zag movement of Random
I will try both methods and see how the mount and guiding recovers after a Dither

Thanks
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 07:47 AM.

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