Author Topic: Help! trouble with dbe  (Read 2647 times)

Offline bcstein12

  • Newcomer
  • Posts: 22
    • View Profile
Help! trouble with dbe
« on: 2014 February 11 06:35:18 »
Cant seen to get the gradiant in the lower lt corner  out of this 15sec ISO 400 frame I was planning to use in the hdr app.  linkg to file if anyone wants to give it a try.  Thanks! 
https://www.dropbox.com/s/cmrjkf1mv73loi7/15sec_iso400.fit

Offline edd

  • Newcomer
  • Posts: 32
    • View Profile
Re: Help! trouble with dbe
« Reply #1 on: 2014 February 11 08:02:19 »
Have you tried ABE instead (or in addition)? Seems to help on it for me.

Offline bcstein12

  • Newcomer
  • Posts: 22
    • View Profile
Re: Help! trouble with dbe
« Reply #2 on: 2014 February 11 11:01:46 »
no i have not.  I'll give it a try.  how does the symmetry settings work.  could I use a point on the opposite side of the gradient.  When i try to place a point on the gradient  now it come up with a "red" marker. Tried to increase the tolerance but the end result is extremely noisy background as compared to the original pic

Offline pfile

  • PTeam Member
  • PixInsight Jedi Grand Master
  • ********
  • Posts: 4685
    • View Profile
Re: Help! trouble with dbe
« Reply #3 on: 2014 February 11 11:09:08 »
the red marker is the problem. increase the tolerance a bit and then click "resize all". that will cause the dbe sample markers to be re-rendered and you can then see if the samples have changed from red to green.

the idea of the symmetry settings is to pick a point in the gradient that's equal to where you want the sample to lie, but where the background is exposed. so if there is a top-to-bottom gradient in your image, but on the left hand side the DSO is in the way of the background, you can tell DBE to use a point to the right of the DSO and it will be applied as though it was sampled from the left side. but it only makes sense if your gradient is relatively linear/smooth. you should not have to mess with the symmetry options for this image.

rob

Offline georg.viehoever

  • PTeam Member
  • PixInsight Jedi Master
  • ******
  • Posts: 2132
    • View Profile
Re: Help! trouble with dbe
« Reply #4 on: 2014 February 11 11:22:25 »
DBE is not increasing the noise in your image. It is just that without gradient, STF can more agressively enhance the background values, giving the visual impression that there is more noise. DBE just subtracts the background model, and you you inspect it, it is very smooth.

There are two problems with your image.
1. Due to its high dynamic range, you need to increase DBE tolerance
2. There is not much background that you can use as a reference - just loads of nebulosity.

I nevertheless tried by placing a couple of samples in locations that I believed should have similar brightness values. The settings and the result can be seen in the screenshot.

Georg
Georg (6 inch Newton, unmodified Canon EOS40D+80D, unguided EQ5 mount)

Offline bcstein12

  • Newcomer
  • Posts: 22
    • View Profile
Re: Help! trouble with dbe
« Reply #5 on: 2014 February 11 11:58:43 »
I got the same looking result as you.  I was worried about the "grainy noisy" result.  SO that really not a problem ?  May 180sec image turned out great and I was planning to combine this 15sec using the HDR tool.  THanks you your help :)

Offline Josh Lake

  • PixInsight Old Hand
  • ****
  • Posts: 424
    • View Profile
Re: Help! trouble with dbe
« Reply #6 on: 2014 February 11 14:44:47 »
In that case, don't even worry about it, HDR will drop the 15 second core right in there and you won't have to worry about the noise transferring to your nice long exposure work.

Offline bcstein12

  • Newcomer
  • Posts: 22
    • View Profile
Re: Help! trouble with dbe
« Reply #7 on: 2014 February 17 21:25:59 »
well im embarrassed to post my attempt at HDR after seeing rod771's post of orion :tongue:  My hdr image didnt seem to change much.  I used an ISO 800 300sec image and an ISO 400 15sec image but it seems that the core is still over saturated.  I tried the hdr tool with the binaring setting at 0.05 and 0.8 and it didn't sem to affect the final result except for the mask.  Do I need to add another shorter exposure?