17 replies to this topic

[BigNickelJohnny]

The other night I tried photographing the pinwheel galaxy with my Sony a7R3 and 70-200mm Sony lens with a SkyGuiderPro.

 

I processed the lights only with Siril on a Mac and got this linked image as a final result. This is really my first time using Siril so I have a lot to learn.  Any ideas why I am getting this circular gradient in the stretched image?  It is not apparent in the raw lights.  Is this a processing issue?n Would the result be better if I took biases flats and darks?

 

Looking for some direction.

 

Thanks in advance.

 

John

[gsaramet]

I suppose this is a crop and the center of the image corresponds to the center of the circles?

 

The main thing is... you really want to do the stuff by the book. Bias, darks, flats. Until you do that, you can't eliminate causes. 

 

So, take those calibration frames now (even if they won't be perfect) and redo the stack. If the problem goes away, that's your answer. If it changes, you have an extra data point for troubleshooting. 

[Zambiadarkskies]

I suspect it is some sort of in camera processing - vignetting correction or similar.  

[rollomonk]

The advice above is good. Something else…. how long were your exposures? The shorter they are, the more stretching is required which can magnify issues not corrected by calibration frames…. so in addition to taking calibration frames (especially flats), perhaps you need to increase exposure time?

[rj144]

The advice above is good. Something else…. how long were your exposures? The shorter they are, the more stretching is required which can magnify issues not corrected by calibration frames…. so in addition to taking calibration frames (especially flats), perhaps you need to increase exposure time?

It's not sub length that matters.  It's integration time.

Edited by rj144, 11 November 2024 - 12:41 PM.

[Normmalin]

I've seen this before when I didn't treat the data as floating point but did the processing as integer.  

[BigNickelJohnny]

Thanks everyone for the suggestions thus far.  I will definitely try to capture the flats and biases now and redo the processing. Fingers crossed.

 

During this session I captured 16, 60 sec exposures for the lights and 11 darks.

 

Another thing I am not certain about... after the images are captured, should I be doing ANY processing of those images in LR before going to Siril??  I ask because I think that the original raw frames were underexposed (still learning here) so I bumped the exposure a bit in LR before exporting as TIF files.

 

As for floating point vs integer data... where is this done??? In Siril?

 

Thanks again for the quick responses.

[gsaramet]

Thanks everyone for the suggestions thus far.  I will definitely try to capture the flats and biases now and redo the processing. Fingers crossed.

 

During this session I captured 16, 60 sec exposures for the lights and 11 darks.

 

Another thing I am not certain about... after the images are captured, should I be doing ANY processing of those images in LR before going to Siril??  I ask because I think that the original raw frames were underexposed (still learning here) so I bumped the exposure a bit in LR before exporting as TIF files.

 

As for floating point vs integer data... where is this done??? In Siril?

 

Thanks again for the quick responses.

That's a big huge no-no. Let your raw data alone. There's nothing good you can do to your raw files in any terrestrial imaging software and a lot of bad things that are almost guaranteed to happen. 

[Normmalin]

Thanks everyone for the suggestions thus far.  I will definitely try to capture the flats and biases now and redo the processing. Fingers crossed.

 

During this session I captured 16, 60 sec exposures for the lights and 11 darks.

 

Another thing I am not certain about... after the images are captured, should I be doing ANY processing of those images in LR before going to Siril??  I ask because I think that the original raw frames were underexposed (still learning here) so I bumped the exposure a bit in LR before exporting as TIF files.

 

As for floating point vs integer data... where is this done??? In Siril?

 

Thanks again for the quick responses.

 

I think you are doing it in Lightroom.  It exports 16-bit TIF files by default.  Don't process the raw files at all before processing in Siril and I bet this problem goes away.

[BQ Octantis]

I give it about 87% probability that the banding is caused by the de-vignetting tool in Lightroom. Terrestrial vignette models are not designed to be stretched with AP stretches. The cause is either insufficient bit depth (int8 or int16) or too coarse of a mathematical model (likely both).

 

You should be de-vignetting the final float32 stack out of Siril—and it's far better to de-vignette with flats to deconvolve the vignette (a division problem) from the light pollution gradient (a subtraction problem). And as already stated, you're best off feeding the raws into Siril for calibrating, stacking, and stretching in float32 and doing Lightroom processing afterward.

 

Of course, that path introduces the problem of the missing CCM…

 

BQ

Edited by BQ Octantis, 11 November 2024 - 02:23 PM.

[BigNickelJohnny]

So I went ahead and took flats and bias frames (40 each) and reprocessed all images using raw (ARW) files only and the banding is still there, albeit not as noticeable as before, but still very obvious.

 

I guess I will have to look at the camera and see if there is a setting or two that needs tweaking.

 

Thanks for all the guidance. I'll keep plugging away and learning. Clear skies.

 

 

John

[Noltimier]

Hi John.

 

There was a long thread on a very similar issue happening with Nikon Cameras.

Here is a link:

 

https://www.cloudyni...ncentric-rings/

 

It turns out that the colored rings in the Nikon cameras are caused by a side effect of Nikon's lossy data compression.

Something similar maybe happening with your Sony...

 

Hope this helps you sort it out.

Regards,

 

Mark

 

[BigNickelJohnny]

Hi John.

 

There was a long thread on a very similar issue happening with Nikon Cameras.

Here is a link:

 

https://www.cloudyni...ncentric-rings/

 

It turns out that the colored rings in the Nikon cameras are caused by a side effect of Nikon's lossy data compression.

Something similar maybe happening with your Sony...

 

Hope this helps you sort it out.

Regards,

 

Mark

Thanks Mark... I was just actually reviewing my camera settings and noticed that my RAW file setting is set to Compressed. I wonder if I change this to Uncompressed if the problem goes away!?  Will have to try this next time I have clear skies.

 

Thanks again for chiming in. Appreciate it.

 

John

[gsaramet]

Have you tried restacking with calibration frames and no interference from Lightroom or other terrestrial software?

[sharkmelley]

I processed the lights only with Siril on a Mac and got this linked image as a final result. 

What you have there is a very obvious case of the Sony coloured concentric polygons caused by a lens correction applied to the raw data by the camera:

 

https://www.markshel...d_polygons.html

 

Best advice is to switch off all in-camera lens corrections.  This usually improves things but may not remove them entirely.  The only permanent solution is to use a lens that the camera firmware does not recognise or use the camera on a telescope.

[BigNickelJohnny]

What you have there is a very obvious case of the Sony coloured concentric polygons caused by a lens correction applied to the raw data by the camera:

 

https://www.markshel...d_polygons.html

 

Best advice is to switch off all in-camera lens corrections.  This usually improves things but may not remove them entirely.  The only permanent solution is to use a lens that the camera firmware does not recognise or use the camera on a telescope.

Thank you so much!! I think I will try tonight with my Celestron NexStar 6SE instead.

 

Clear skies!

[BigNickelJohnny]

Have you tried restacking with calibration frames and no interference from Lightroom or other terrestrial software?

Yes, I tried that, with no success.

[sharkmelley]

Thank you so much!! I think I will try tonight with my Celestron NexStar 6SE instead.

The other thing I noticed in your image are short black bars either side of bright stars.  This is caused by Sony's lossy compression.  Use uncompressed or (if available) lossless compression.