PixInsight Forum

PixInsight => Image Processing Challenges => Topic started by: agengard on 2019 May 08 00:57:47

Title: Strange background noise
Post by: agengard on 2019 May 08 00:57:47
Hi all!

I have a small problem and I can’t figure out the cause. Appreciate any help or suggestions.

During three consecutive nights in April, I shot M13, The Great Globular Cluster in Hercules, using my EdgeHD 8” f/7 and ASI294MC. I used a Baader UV/IR-cut filter, 60 seconds exposures at Unity Gain and -20 degrees. I used my usual Superbias, matching MasterDark and MasterFlat from the first night (using my LED-panel). I did not Dither.

I used Blink to sort out less good subs and subs with sat trails, and the Subframe Selector Process to weigh the better ones after FWHM, SNR and Eccentricity. I then used the best sub for registration and integration. It happened to be a sub from the second night.

April 21 I got 168x60sec exposures, April 22 - 240x60sec and April 23 - 219x60sec.

Using the best sub, mentioned above, as reference during stacking subs from any ONE of the nights, results in a photo with a reasonably smooth background and good looking stars. However, any combination of TWO or THREE nights results in a photo with lines in the background in a 45-degree angle. Still the stars look good.

The settings I use are mostly PixInsight Default, and they are also the settings I always use, most times with good result from a novice’s point of view.

In the jpg below one can see the difference between my integrations.

Here are links to XISF-versions. They are just integrated and cropped, no further development.

One night integrated:

Three nights integrated:

Is there an obvious reason to these lines? Is it a no-dither-multi-session-issue? I have stacked multi-sessions before without this phenomenon…

Below is a link to the same corner of my photo of M51 from the week before. Shot with the same equipment except for the filter. My development process was the same, however I used dithering, longer subs, LPS-filter and a few hours longer integration time. It consists of 233x180sec and 89x120sec.

Title: Re: Strange background noise
Post by: Greg Schwimer on 2019 May 08 07:49:34
This is likely due to not dithering or inadequate dithering. I think it is called walking noise. If you blink through the images in the sequence they were taken do you see evidence of drift in the direction of the noise?
Title: Re: Strange background noise
Post by: wvanreeven on 2019 May 08 07:52:13
If dithering was always done in the same direction then that could still account for the noise as well. In general dithering should be done in a general direction every time of course but perhaps the software used for guiding doesn't do that (possibly due to a bug).

Title: Re: Strange background noise
Post by: agengard on 2019 May 09 02:16:20
Greg and Wouter... Thanks a lot!

I did a web-search for “walking noise” and found several threads with images that show identical issues. My problem is clearly due to not dithering at all and a just little less than perfect PA.

When blinking the subs with the shortest time available, I notice the noise seems to creep in the direction of my lines. So I’m convinced this is the reason behind the problem.

Checking my one-night-integrations, I can see lines appear also there, but very faint. Of course, doubling or tripling the number of frames make the problem worse.

I never thought dithering was THAT important if one used enough Dark-corrected subs…

Thank you much for your help!