Author Topic: deconvolution on a m81/m82 image  (Read 457 times)

Offline mmnb

  • Member
  • *
  • Posts: 66
    • View Profile
deconvolution on a m81/m82 image
« on: 2019 July 18 10:28:11 »
Took some image with my C14 Hyperstar + EOS 60Da (32x30s IRRC) and trying to take the drizzled and pcc'd linear image through deconvolution/denoising and stretching. I think I am reasonably proficient with star masks now (I do miss a few double stars....seems like a lot of trouble to refine to get those last bits, is this typical?) and tried to create an appropriate deringing support as advised here:
http://www.pixinsight.com/examples/M81M82/index.html

No matter what I do, I can't seem to to improve the image with deconvolution.  I can see "too much" where it looks like it is being stretched around noise or what looks like next to nothing. I can't see the same sorts of visual contrasts that are in the tutorial.  An important issue:

When I create the external PSF, the star profile is not round; a brutally honest reminder that the collimation is off (I've been trying to fix, but having a hard time).  Does the poor collimation affect how deconvolution will perform? E.g. Will it try an model collimation error in the underlying model that is based on atmosphere?

Are there any other image issues that stand out? Note I only correct with bias and flat due to the way the Canon does darks.


I've included masks in the zip file (1G):
https://drive.google.com/open?id=1tA7bES-s9rlv09VReQBoYFz6qVgUcDxv

Edit: Added a screenshot of a blind mask stretch and HDRMT.  Nice to see that there is some comparable detail in the data...I wouldn't know much more than to do pre/post denoising in the dark and a little bit of saturation and brightness adjustment.  Should I be able to do much better with this data? I swear I see a bit the IFN when I compare to other pictures, but it is quite faint and not sure if I can successfully pull it out.
« Last Edit: 2019 July 18 22:44:23 by mmnb »

Offline mmnb

  • Member
  • *
  • Posts: 66
    • View Profile
Re: deconvolution on a m81/m82 image
« Reply #1 on: 2019 July 19 10:34:18 »
Ah one important thing that I missed (and was clearly stated in the article) was that you need to check the deconvolution result with stretching.  After applying MaskedStretch to the deconvoluted image I can see some effects that aren't visible with the default STF.  The enhancement to the image seems relatively small to me in this case nonetheless.

Offline mmnb

  • Member
  • *
  • Posts: 66
    • View Profile
Re: deconvolution on a m81/m82 image
« Reply #2 on: 2019 July 19 23:29:59 »
Best I could do.  Deconvolution (really not sure if it did a whole lot), denoising, masked stretch, HDRMT (seemed to work like magic). Saturation adjustments (tried to really punch red selectively for the red hydrogen filaments. Is this way off in any way?  The weak signal in the background has some shapes that line up with the IFN in better data, tried to be honest in keeping a little contrast but very faint.

Would love to see if someone else can do much better.

Offline John Massey

  • Newcomer
  • Posts: 45
    • View Profile
Re: deconvolution on a m81/m82 image
« Reply #3 on: 2019 July 20 06:38:03 »
Ron Brecher (Astrodoc) had an article in S&T Mag of July 2017 that may help you. http://astrodoc.ca/wp-content/uploads/2017/06/Sky-and-Telescope-July-2017-Deconvolution-article.pdf
He mentions that the differences are subtle in the linear stage. It may be a good read for you.

Offline Bobinius

  • Newcomer
  • Posts: 42
    • View Profile
Re: deconvolution on a m81/m82 image
« Reply #4 on: 2019 August 03 01:55:27 »
Took some image with my C14 Hyperstar + EOS 60Da (32x30s IRRC) and trying to take the drizzled and pcc'd linear image through deconvolution/denoising and stretching. I think I am reasonably proficient with star masks now (I do miss a few double stars....seems like a lot of trouble to refine to get those last bits, is this typical?) and tried to create an appropriate deringing support as advised here:
http://www.pixinsight.com/examples/M81M82/index.html

No matter what I do, I can't seem to to improve the image with deconvolution.  I can see "too much" where it looks like it is being stretched around noise or what looks like next to nothing. I can't see the same sorts of visual contrasts that are in the tutorial.  An important issue:

When I create the external PSF, the star profile is not round; a brutally honest reminder that the collimation is off (I've been trying to fix, but having a hard time).  Does the poor collimation affect how deconvolution will perform? E.g. Will it try an model collimation error in the underlying model that is based on atmosphere?

Are there any other image issues that stand out? Note I only correct with bias and flat due to the way the Canon does darks.


I've included masks in the zip file (1G):
https://drive.google.com/open?id=1tA7bES-s9rlv09VReQBoYFz6qVgUcDxv

Edit: Added a screenshot of a blind mask stretch and HDRMT.  Nice to see that there is some comparable detail in the data...I wouldn't know much more than to do pre/post denoising in the dark and a little bit of saturation and brightness adjustment.  Should I be able to do much better with this data? I swear I see a bit the IFN when I compare to other pictures, but it is quite faint and not sure if I can successfully pull it out.

I I understand correctly, you are expecting some spectacular change but you don't see any difference after deconvolution? The change is rarely spectacular but it can increase the details in the nebulosity. However, in order to see it you need to use a preview with a small nebular selection (maybe you do that already) and some stars; if your field is too wide it is hard to see the difference in the details. I usually use ctrl-shift-z on the preview to rapidly shift between before/after. You did not tell your parameters for deconvolution, how many steps, etc. I usually start with 20 iterations and try to correct progressively for the dark rings artifacts. Eventually increase the iterations after as much as possible in order to keep the convergence (you will have to also increase the global dark ring protection once you increase the iterations).

The PSF not being round is not wrong, it reflects the stars that were entered in the equation. Your whole image was exposed to the same collimation/tracking/flexion problem. Hope this helps.

B