PI DBE question

[edarter]

Hi,

Should applying DBE also result in quite a considerable amount of noise also being generated? Below is a before and after image of a Whirlpool stack I'm working on. While DBE has killed the background gradients etc, it has also introduced a lot of noise. I saw the same with the M106 image I did recently as well but don't recall seeing it when I first started with PI..... though that may have been because there was so much going on with my learning curve I just didn't spot it lol.

Thanks
Ed

Edited November 9, 2022 by edarter
spelling correction

[Ouroboros]

@edarter  I’m happy to be corrected by more experienced Pixinsighters but I don’t believe DBE can introduce noise. I wonder whether DBE just makes it more noticeable because it’s removed the background that previously was hiding it.   Have you applied a new STF stretch to the data after DBE? That may bring out different features in the corrected image, including noise perhaps.

Edited November 9, 2022 by Ouroboros

[scotty38]

I "think" it's to do with how STF computes the stretch after the background removal so the image isn't actually noisier.

[wimvb]

Both answers are correct. Because DBE subtracts uneven sky background, you end up with less captured light (less background light, not subject light), but the associated noise is still there. The statistics of the image have changed, and the new STF is a bit more aggressive, revealing the noise that was always there.

You can adjust the permanent stretch so as to avoid the prominent noise shown by the stf. In other words, avoid using the new stf as a permanent stretch.

[Ouroboros]

The last two posts are saying what I was thinking but didn’t articulate very well.   In the normal course of processing it should be possible to reduce that noise before applying a permanent  stretch. 

[edarter]

8 hours ago, Ouroboros said:

The last two posts are saying what I was thinking but didn’t articulate very well.   In the normal course of processing it should be possible to reduce that noise before applying a permanent  stretch. 

Indeed - I've taken to using this method for denoise pre stretch for my last 2-3 images and it seems to work well with my clunky ham-fisted approach
https://jonrista.com/the-astrophotographers-guide/pixinsights/effective-noise-reduction-part-2/

[Ouroboros]

8 hours ago, edarter said:

Indeed - I've taken to using this method for denoise pre stretch for my last 2-3 images and it seems to work well with my clunky ham-fisted approach
https://jonrista.com/the-astrophotographers-guide/pixinsights/effective-noise-reduction-part-2/

I recently started using EZ Denoise in the free EZ Processing suite. EZ Denoise gives better results than my attempts at reducing noise. It’s worth taking a look at. 

[edarter]

22 minutes ago, Ouroboros said:

I recently started using EZ Denoise in the free EZ Processing suite. EZ Denoise gives better results than my attempts at reducing noise. It’s worth taking a look at. 

Thanks, I had completely forgotten about that option!

[scotty38]

EZDenoise here too although on my aging machine it does take a while....

[Stuart1971]

As well as the EZ denoise, try the EZ soft stretch to as your way of moving to non linear stage, it’s better than the STF

personnaly I now use the Noise eXterminator…it’s superb, but not free..