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

Jupiter - newbie with questions

  • Please log in to reply
17 replies to this topic

#1 Gavin Bray

Gavin Bray

    Messenger

  • -----
  • topic starter
  • Posts: 430
  • Joined: 31 Dec 2006
  • Loc: Gold Coast, Australia

Posted 24 July 2019 - 03:57 AM

Hi,

 

I've just started attempting planetary imaging and, although I've read as much as I can, I'm still a bit confused.

 

I'm using a 14" SCT @F20 with an ASI224MC. I have a Starlight Instruments motorized focuser controlled through TheSkyX. Jupiter was near zenith when I captured it but seeing was probably average at best. I used the TheSkyX @Focus2 and I think the focus is ok. I think my collimation is ok. I think I understand how to use FireCapture and AS!3. Note that I'm not definite about any of this.

 

I used Registax for the RGB balance and wavelet sharpening/de-noising. However, I don't really have a good understanding of this and just played with the sliders and values until it looked ok.

 

Is there some way of knowing whether the result is the best I can expect given the seeing conditions?

 

What does a good A!3 quality graph look like?

 

I suspect my lack of processing skills, rather than the seeing, is hampering the result and I'm sure I have a lot to learn.

 

Any guidance is most welcome, especially on the processing side.

 

Thanks.

Attached Thumbnails

  • small-jupiter.png
  • as3.png

  • RedLionNJ and Foc like this

#2 Tulloch

Tulloch

    Viking 1

  • *****
  • Posts: 586
  • Joined: 02 Mar 2019
  • Loc: Melbourne, Australia

Posted 24 July 2019 - 04:19 AM

Hi there, where in Australia are you?

 

Down here in Melbourne the jet stream had been over 70m/s recently, that could easily cause poor seeing. It’s dropped now, just in time for the clouds to roll in frown.gif.

 

just a thought, Andrew



#3 Gavin Bray

Gavin Bray

    Messenger

  • -----
  • topic starter
  • Posts: 430
  • Joined: 31 Dec 2006
  • Loc: Gold Coast, Australia

Posted 24 July 2019 - 04:44 AM

I'm on the Gold Coast. According to this the jet stream has been up over 60 m/s.

 

https://www.meteoblu...stralia_8348930

 

If my results are limited by the seeing, that would be great in some senses. However, I suspect a lot of it has to do with my inexperience, especially with the processing.



#4 Tulloch

Tulloch

    Viking 1

  • *****
  • Posts: 586
  • Joined: 02 Mar 2019
  • Loc: Melbourne, Australia

Posted 24 July 2019 - 04:59 AM

I'm on the Gold Coast. According to this the jet stream has been up over 60 m/s.

 

https://www.meteoblu...stralia_8348930

 

If my results are limited by the seeing, that would be great in some senses. However, I suspect a lot of it has to do with my inexperience, especially with the processing.

Well, I tried some imaging down here last week because the clouds finally cleared after 2 weeks of rain, but with a 60 m/s jetstream. I'm no expert in processing and have significantly inferior equipment to you, but after an hour or two of processing the data I took, it all ended up in the "Recycle Bin" and I vowed never to try imaging with jetstreams that high again.

 

Mind you, I'm not saying your processing skills are perfect, but sometimes you just can't polish t**ds into diamonds ...

 

Andrew



#5 Gavin Bray

Gavin Bray

    Messenger

  • -----
  • topic starter
  • Posts: 430
  • Joined: 31 Dec 2006
  • Loc: Gold Coast, Australia

Posted 24 July 2019 - 06:21 AM

True. :-)

I'll keep trying. Hopefully the seeing will improve. I'll also see if I can do more with Registax.

What sort of jet stream speeds do you consider good?



#6 happylimpet

happylimpet

    Skylab

  • *****
  • Posts: 4042
  • Joined: 29 Sep 2013
  • Loc: Southampton, UK

Posted 24 July 2019 - 06:44 AM

Looks good.

 

Use laplace 6 as a default - 4 needs very sharp , very high SNR images to work.

 

You might as well do a range of stack %ages, I do 5,12,30 and 75 as defaults, then you can see where the best compromise between signal/noise and sharpness lies. 25 probably is about right though.

 

I set the wavelets in registax to dyadic, turn 1 right up and then apply however much noise reduction is needed, turn up 2 a fair bit too, and thats about it.


  • RedLionNJ and Billytk like this

#7 Tulloch

Tulloch

    Viking 1

  • *****
  • Posts: 586
  • Joined: 02 Mar 2019
  • Loc: Melbourne, Australia

Posted 24 July 2019 - 06:55 AM

True. :-)

I'll keep trying. Hopefully the seeing will improve. I'll also see if I can do more with Registax.

What sort of jet stream speeds do you consider good?

I don't have a good handle on it yet (which is why I went out when it was 60), but meteoblue reckons anything above 20 is bad and it goes red above 30. That's probably as good an indication as any ...

 

Andrew



#8 yock1960

yock1960

    Aurora

  • *****
  • Posts: 4808
  • Joined: 22 Jun 2008
  • Loc: (Crossroad of clouds) Ohio, USA

Posted 24 July 2019 - 08:15 AM

I would say in general, that your quality graph is not particularly good...but it does look quite a bit like what we've been seeing in the Northern hemisphere this year! grin.gif

 

There is some variation in graph profiles and it can be 'fooled' in various cases, but the farther to the right that it crosses the 50% quality line, the better. Once in a great while, I'll get some where it stays above 50% quite a long ways past the 50% frame mark...those typically produce good images. As with many things though YMMV!

 

Steve 



#9 Sunspot

Sunspot

    Cosmos

  • *****
  • Posts: 8993
  • Joined: 15 Mar 2005
  • Loc: Surprise, AZ

Posted 24 July 2019 - 10:03 AM

I never check the "Sharpen Blend" box. Does anyone else use that?


  • yock1960 and RedLionNJ like this

#10 Tulloch

Tulloch

    Viking 1

  • *****
  • Posts: 586
  • Joined: 02 Mar 2019
  • Loc: Melbourne, Australia

Posted 24 July 2019 - 04:17 PM

I never check the "Sharpen Blend" box. Does anyone else use that?

I used to, but don't any more. All it does is creates 2 stacked images, one slightly sharpened so you can see if it's worth sending the other unsharpened one to Registax. While it might give your some idea, I send everything to Registax so I don't bother with creating the sharpened one any more.

 

As others have said, based on your AS! image you appear to be on the right track. Set Laplace delta to 6, play around with the stacking levels (I usually choose 25% and 50% and pick the best looking one after sharpening), I also use 3x drizzle as it makes the image size 3x larger so that you can sharpen on a larger image before downsizing it for the final image which removes/hides some of the artifacts.

 

Registax is where you can turn your fuzzy, blurry blobs into crisp clear images, but I've always found it hard to find the right settings. Too high, and everything become noisy and looks wrong, too high a de-noise figure and it all turns mushy. Unfortunately, what works for one scope doesn't usually work for another, and seeing on the night can also change the "best" settings. Registax is where I spent the most time trying to polish my data, with mixed success.

 

If you haven't already done so, Steve (aka WarmWeatherGuy) has produced some great videos on the subject, have a look at his one on processing for example settings. I still find things in them that help me improve.

 

http://planetaryimagingtutorials.com/

 

Good luck!

 

Andrew


  • Billytk likes this

#11 RedLionNJ

RedLionNJ

    Soyuz

  • *****
  • Posts: 3739
  • Joined: 29 Dec 2009
  • Loc: Red Lion, NJ, USA

Posted 24 July 2019 - 05:06 PM

I'm not totally familiar with the focusing terminology you used, but I get the impression perhaps you used a computerized focusing method?  If so, that's never going to yield optimum focus for Jupiter and possibly not for Saturn, either. The edges aren't sharp enough for this to work effectively.  Plus, fine focus on the planet is best assessed by how much detail is visible on the preview screen. A small tweak there can make a MASSIVE difference in the outcome.

 

Other than that, the result looks fairly typical for this year, although most of the southern hemisphere images seem to have more 'pzazz' and vibrancy.  Your result is (as is mentioned above, already) fairly typical for a northern hemisphere imager this season.

 

One other thing worth assessing (and not already mentioned) is the collimation. Take an image of a mid-brightness star (in what you consider to be best focus) but don't over-expose it. Is the Airy Disk "clean", small and symmetrical? If not, perhaps a small collimation tweak is in order. This has to be done on a nearby star, not on Jupiter or even one of its satellites.

 

As far as processing - don't see a whole lot wrong, there. Everybody has their own method for sharpening and color/gamma tweaking in Registax.  As Mark mentioned, a noise robust of 6 is a good value for AutoStakkert's quality assessment. And definitely try Nick's suggestion of multiple percentages - one may genuinely jump out as far superior to the others (but in this case, the smooth graph makes that unlikely).  And nobody really uses the 'Sharpened' option in practice. Let all the sharpening happen in Registax.

 

Great start!

 

Grant


  • Billytk likes this

#12 happylimpet

happylimpet

    Skylab

  • *****
  • Posts: 4042
  • Joined: 29 Sep 2013
  • Loc: Southampton, UK

Posted 24 July 2019 - 06:57 PM

The 'sharpened' stack can be an excellent quick look guide to which are the best stacks. And if Im honest, I have used them, especially when starting out. They can be really rather good.



#13 Gavin Bray

Gavin Bray

    Messenger

  • -----
  • topic starter
  • Posts: 430
  • Joined: 31 Dec 2006
  • Loc: Gold Coast, Australia

Posted 24 July 2019 - 07:02 PM

Thanks everyone for your suggestions.

 

I'll switch the Laplace setting to 6 and will turn off the sharpened checkbox. I wasn't using the sharpened checkbox anyway. I have tried different stack percentages but 25% seems to be about right at the moment.

 

I found Steve's videos very useful but I wish there was something more on Registax wavelet processing. It seems to be a black art. :-)

 

I used TheSkyX's @Focus2 to set the focus on a 5th magnitude (or thereabouts) star near Jupiter. I think the focus is ok. I also think the collimation is reasonably ok.

 

I'm going to keep trying, waiting for better seeing and playing more with Registax.



#14 roelb

roelb

    Apollo

  • *****
  • Posts: 1340
  • Joined: 21 Dec 2013
  • Loc: Belgium, Antwerp

Posted 24 July 2019 - 07:13 PM

Don't focus on a nearby star but on the planet itself.
  • happylimpet likes this

#15 Gavin Bray

Gavin Bray

    Messenger

  • -----
  • topic starter
  • Posts: 430
  • Joined: 31 Dec 2006
  • Loc: Gold Coast, Australia

Posted 24 July 2019 - 07:17 PM

How do you do that?



#16 Kokatha man

Kokatha man

    Voyager 1

  • *****
  • Posts: 13012
  • Joined: 13 Sep 2009
  • Loc: "cooker-ta man" downunda...

Posted 24 July 2019 - 08:20 PM

I used TheSkyX's @Focus2 to set the focus on a 5th magnitude (or thereabouts) star near Jupiter. I think the focus is ok. I also think the collimation is reasonably ok.

 

You really need to get experience focusing on planets to advance your imaging outcomes I have to say!

 

After initially getting a sharp edge to the planet & adjusting the gain such that any detail is apparent an "increase" in the gamma value creates greater contrast to allow finer judgement of the clarity/sharpness of any visible features on the disk...for the actual capture this should be returned to the default vale btw.

 

Jupiter has numerous "indicators" to gauge optimum focus: the festoons or thread-like filaments that project into the EZ, getting these gossamer-like features as sharp & well-defined as possible is a good approach & at the same time in the EZ there might appear small "splotches" of dark matter (often associated with the festoons) as additional guides.

 

Another indicator are the small, circular-like white storm spots, these become more apparent as you get closer to optimum focus, especially concentrating upon the smaller ones if they are visible further N or S of the Eq.

 

The darker NEB & SEB, definition alters from a bland darker colouration out of focus to a much clearer set of features with clear definition in the edges of these features where the borders are curled, curved or fluted - & lighter & darker details reveal themselves within these dark bands...

 

Saturn is relatively easy in comparison: the grayscale used for basic computer screen calibration displays 256 (8-bit) tonal images from black through to white where it should be quite easy to detect the differences between the blackest & all the other shades; all that is really needed is to adjust the focus on Saturn until the Cassini Division appears "blackest."

 

In average seeing or above, this is quite self-evident: once you have achieved this you will also notice (if the seeing is good enough) that you are able to define numerous bands on the disk of Saturn in larger apertures...

 

I'll leave Uranus & Neptune for now lol.gif but there is other information that you might find useful on our website also: accurate collimation is also an essential for good image outcomes btw. (see link in signature)


  • eros312, Billytk, roelb and 4 others like this

#17 Gavin Bray

Gavin Bray

    Messenger

  • -----
  • topic starter
  • Posts: 430
  • Joined: 31 Dec 2006
  • Loc: Gold Coast, Australia

Posted 24 July 2019 - 08:45 PM

Thank you for the detailed explanation. I will try that next time. There's lots to learn obviously.

 

To summarize, I need to:

 

1. Look out for better seeing conditions.

2. Make sure the collimation is spot on.

3. Focus on the planet as you describe.

4. FireCapture - I think I'm ok here.

5. A!3 - adjust the Laplace setting as suggested above.

6. Registax - keep playing with the wavelet settings.

 

Does that sound about right?

 

I see you have some imaging processing tutorials. I'll go through those too.


  • Billytk and gfstallin like this

#18 Kokatha man

Kokatha man

    Voyager 1

  • *****
  • Posts: 13012
  • Joined: 13 Sep 2009
  • Loc: "cooker-ta man" downunda...

Posted 24 July 2019 - 10:05 PM

"Yes" - when you increase gamma to assist focus you will obviously need to alter gain (above that which you wish to capture at) to make the image bright enough to discern features best: some folks like to focus on the colour image & others like to switch off the colour & use a mono image for focus...

 

In AS!3 "6" is a general "value that fits all" unless either your feed is extremely good (although here it is less of an issue) or extremely bad, where you might increase it to perhaps "7".




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