Winjupos Problems

[Stub Mandrel]

Help!

Can anyone tell me where I'm  going wrong here?

I thought the dark edges of the planet were caused by me having the frame too large, but i made it smaller and the derotated image is still poor.

 

 

[Pete Presland]

How long is the video you are derotating? 

[Stub Mandrel]

5 hours ago, Pete Presland said:

How long is the video you are derotating? 

It isn't video, it's 12 stacks from 1-minute long videos.

[1CM69]

I'm very new to using WinJupos too so do not know that much about it's workings & I do have my own problem with it that I want to sort out but could it be as simple as you have not entered the coords for your observing location. I have also found that it seems to be particularly finicky with the frame aligning, I have been zooming in as far as possible making sure the whole image is still visible on the screen and aligning the frame that way.

Using the contrast & brightness options can help find a "hard" edge to align to amongst the fuzz, be careful though as too much adjustment of these options can shrink or expand the image and make aligning more difficult.

I initially followed this tutorial which got me off to a good start especially when using multi-datasets: http://www.thelondonastronomer.com/it-is-rocket-science/2018/6/7/a-quick-guide-to-planetary-imaging

The bares bones of my workflow is:

Start with the first recording made of the first dataset recorded; LRGB or RGB i.e L filter going through in order taken, (this means less toing & froing with the alignment frame), making sure to change the colour channel in the ADJ tab for the filter you are using before saving the IMS file.

Once you have done this for all you datasets, from the TOOLS menu choose the 'De-rotation of Images' option and start by adding all your L filters IMS files from the previous step, this creates a master L IMS file.

Do this now for all your R filters & so on....

Now in the TOOLS menu choose the 'De-rotation of RGB frames' and load the masters IMS files to their corresponding slots and hit compile.

 

Also, I assume that you are first stacking each filters recording in AS!3 or similar without sharpening, go for the highest quality you can get, I do not go below 75% quality in AS!3 & on a bad day I get lower than 1% of frames to stack but obviously AS!3 only allows 1 as the lowest entry.

Open the stacked image in Registax 6 and wavelet sharpen as much as possible without adding too much noise.

This has to be done for all recording, very laborious and time consuming and you'll end up with a massive folder structure containing the various files at various steps but you'll find your own system that works for you.

Hope this help somewhat.

 

Kirk

[Stub Mandrel]

Hi Kirk,

That's a pretty much my workflow.

What is confounding me is that I can get decent results with Mars and Jupiter, but Saturn has those terrible edges to the disc.

It SHOULD BE easy to align with Saturn as I have the Cassini division really well defined and the inner and outer edges of the rings as well as the planet itself to check against.

Is there a discussion forum for WINjupos?

[1CM69]

I haven’t come across a dedicated forum as yet, I’ve just scrapped together various bit lbs of info while trawling the web. 

I have also had issues with edges like you show, more so on Saturn as you point out but I’ve had it on Mars & Jupiter too, especially ghost like double edges. 

All I can suggest is to ask one of the pro imagers, Damian Peach for example, for their input. If they use WinJupos or maybe there’s other software. 

I fiddle around with the LD level which can alleviate the edge problem but this also causes a loss of definition. 

Kirk

[1CM69]

I have even considered if the whole process of using WinJupos with Saturn is worth the effort in the long run because unless you have top class equipment and totally perfect seeing to be able to make out the hexagon, Saturn only shows as various coloured bands going around the disc.

Majority of images even from the pros look like a freeze frame blur of a ball spinning very fast. There are no real individual cloud structure details to pull out like Jupiter has and certainly no surface features as Mars.

It might be just as easy or easier to stack filter captures then restack like filter stacks to make a master filter stack then assign each of these as layers to a colour channel in Photoshop or similar. Each layer can be manipulated to allow for change in planetary tilt.

I still have data from my most recent captures so I may try this and judge side by side.

Kirk

[Stub Mandrel]

Winjupos is what I call 'nearly' software - it's 'nearly brilliant'.

There is no meaningful documentation, it's very hard to adjust the settings and tehre's no guidance on how to use the very non-intuitive controls, especially for limb darkening.

Making maps is really difficult as it isn't explained that the images aren't merged, or stacked, just overlaid and the challenge is to find the best image for each bit of the map and blend them together, with several options that 'do something' without any explanation of what 'something' is.

It desperately, desperately, needs proper documentation and also its should be possible to enter things like LD values by typing them in.

Also some explanation of how some common faults (like crazy zig-zag patterns) arise and how to cure them.

 

[1CM69]

I totally agree, it’s almost like the developer got fed up part way through creating it. 

Like I said, I am very new to WJ and came to it by many people raving on about it. 

Like you said, it’s ‘nearly’ there but also lacking in vital areas. 

[dunc]

I have derotated Jupiter many times but never bothered with Saturn for the reasons noted above. My first observation is you haven't set your location ( as mentioned above).  WinJupos calulates the outline projection very accurately and you are not telling it you observed Saturn from Staffordshire. Therefore you can't line up your images and the outline well enough. I am sure your location also affects the de-rotation calulations as it works out were to move pixels tiny amounts. Accurate measurement is key. WinJupos was written for a scientific purpose, us planetary imagers using it for de-rotation of a set of consecutive images is a happy byproduct of whats needed to measure planetary features accurately.

As the Jupos.org website says...

"Most important software tool of the project is WinJUPOS written by Grischa Hahn. It is specially designed for amateur astronomers to record and analyse positions of features on Jupiter (and other planets), and to display them in suitable time-longitude or other coordinate systems. With its help you are able to process visual observations (Central Meridian transit timings, micrometer measurements) but also to measure electronic images (scanned photos, CCD or webcam images). Positions are written to a special JUPOS database, and can be queried, filtered and displayed graphically."

Hope that helps,

Duncan

[Stub Mandrel]

46 minutes ago, dunc said:

My first observation is you haven't set your location ( as mentioned above). 

How Odd!

I have it set for Jupiter and Saturn, I wonder why Saturn was unset.

I have put it in correctly, however I experimented and the actual value makes no visible difference to the wireframe for a body as far away as Saturn.

I'm sure it is significant for the moon, for example.

 

I do wish the program was documented, none of the tutorials goes past the basics.