Author Topic: Can't get Luminance mask in ACDNR working right in 1.2  (Read 3450 times)

Offline Nocturnal

  • PixInsight Jedi Council Member
  • *******
  • Posts: 2727
    • http://www.carpephoton.com
Hi,

well I bought a license so it's open season now :) I'm hoping someone can tell me what I'm doing wrong here. I'm trying to reduce noise in my image. To keep the bright areas protected I use a luminance mask. It seems the mask prevents any noise reduction to take place, not just in the black areas. So clearly I didn't quite understand how this works :)

Here's a screen shot of the mask:

http://gallery.tungstentech.com/main.php?g2_itemId=767

Here's a screen shot of the preview with mask enabled:

http://gallery.tungstentech.com/main.php?g2_itemId=773

Notice the lack of noise reduction.

Here's the mask disabled:

http://gallery.tungstentech.com/main.php?g2_itemId=770

Notice noise reduction in full effect.


I've created an image of the preview I used and uploaded it here:

http://cid-a93625fef5ca95fb.skydrive.live.com/self.aspx/PixInsight/SmallSection.7z

Thanks for any tips you can provide,

   Sander
Best,

    Sander
---
Edge HD 1100
QHY-8 for imaging, IMG0H mono for guiding, video cameras for occulations
ASI224, QHY5L-IIc
HyperStar3
WO-M110ED+FR-III/TRF-2008
Takahashi EM-400
PIxInsight, DeepSkyStacker, PHD, Nebulosity

Offline Juan Conejero

  • PTeam Member
  • PixInsight Jedi Grand Master
  • ********
  • Posts: 7111
    • http://pixinsight.com/
Can't get Luminance mask in ACDNR working right in 1.2
« Reply #1 on: 2008 July 23 03:50:25 »
Hi Sander,

Now I have a new user, a terrific bug hunter, and a C++/JS developer, all in a single operation, so what else can I ask for? :) Thank you for your support, and welcome aboard.

There's nothing wrong in the way you're using ACDNR. It's just that you have defined a mask that is too restrictive. I have reproduced your parameters and I get just the same poor noise reduction.

Try with a more permissive mask; for example:

http://forum-images.pixinsight.com/legacy/acdnr-sander/01.jpg

which gives you a reasonable noise reduction:

http://forum-images.pixinsight.com/legacy/acdnr-sander/02.jpg

However, for this image you actually don't need a mask:

http://forum-images.pixinsight.com/legacy/acdnr-sander/03.jpg

As you see, ACDNR performs very well without the help of a mask for images like this one. A mask is required when there are more differences between high-SNR and low-SNR regions; for example, when there are subtle variations over bright areas with little noise. It's a matter of practice.

Note that I've used edge protection overdrive parameters on both sides. This has helped in protecting star profiles. I've used overdrive also in the chrominance, to protect small stars from becoming grayed out:

http://forum-images.pixinsight.com/legacy/acdnr-sander/04.jpg

The overdrive parameters exaggerate edge protection. With overdrive=0, the algorithm tries to achieve the same contrast as in the original image for protected edges. When overdrive > 0, contrast is increased for protected edges, which has a sort of sharpening effect. Overdrive can help, but these parameters must be used with great care. If in doubt, better don't use them.

Hope this helps.
Juan Conejero
PixInsight Development Team
http://pixinsight.com/