Author Topic: DBE question  (Read 8039 times)

Offline DaveS

  • PixInsight Addict
  • ***
  • Posts: 115
    • Dave's Astronomy Pages
DBE question
« on: 2010 September 25 08:14:21 »
Sorry guys, but yet another question.

I have an image open, I open the DBE window, generate a DBE image, adjust it with the HST, until I have pleasing image.

On the desk top, I now have two images, the original with the background 'markers' still in place, and the new DBE enhanced image.

How do I apply the new DBE enhanced to the original image?

I have tried making the original image 'active' and apply the DBE from the DBE window. Also tried dragging the DBE image onto the original, where it then appears as a red mask, but from there on, I can't do anything sensible with it.  ???

Dave

8" LX200ACF
William Optics FLT110
NEQ6 Mount
SXVF-H9
SXVR-16
SX Lodestar
DMK21AU04
Baader LRGB and NB filters
DiY Observatory
http://www.progressiveastroimaging.com/davesastronomy/

Offline Nocturnal

  • PixInsight Jedi Council Member
  • *******
  • Posts: 2727
    • http://www.carpephoton.com
Re: DBE question
« Reply #1 on: 2010 September 25 08:27:46 »
Dave,

open the 'target image correction' tab on the DBE process. Select 'subtract' and 'replace target image'. DBE used to be just what the name implies, *extraction* only. No correction. You were supposed to subtract the background using PixelMath. This still works but Juan luckily agreed that it would be better to integrate subtraction into the process. Modularity is nice but only if it helps.

I encourage you to write more descriptive Subjects for your posts, if you don't mind. Something like 'DBE: how to apply' or 'how do I modify the DBE target image' are much more useful. Keep in mind you're trying to attract people to read you message so they will help -and- you want your post to be found when people go searching for the same topic.
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 DaveS

  • PixInsight Addict
  • ***
  • Posts: 115
    • Dave's Astronomy Pages
Re: DBE question
« Reply #2 on: 2010 September 25 08:54:32 »
Hi Sanders

Ok, that works now.

I was expecting the new image with the DBE suffix, to be intergated (subtracted) from the original image, but in doing so see its effect appear on the original image, as with masking functions. Not the DBE suffixed image being a 'stand alone' replacement.

At the moment I'm still think intuitevely in PS mode. :laugh:

Duly noted about posting more descriptive titles.

Dave
8" LX200ACF
William Optics FLT110
NEQ6 Mount
SXVF-H9
SXVR-16
SX Lodestar
DMK21AU04
Baader LRGB and NB filters
DiY Observatory
http://www.progressiveastroimaging.com/davesastronomy/

Offline Harry page

  • PTeam Member
  • PixInsight Jedi Knight
  • *****
  • Posts: 1458
    • http://www.harrysastroshed.com
Re: DBE question
« Reply #3 on: 2010 September 25 09:18:03 »
Hi Dave

Have a read of how to use dbe here http://www.harrysastroshed.com/pixuser/DBE.html

Please remember there are no layers in PI , you are either creating new images to replace the old ,or using masks to work on restricted areas of the image  :D

Harry
Harry Page

Offline Nocturnal

  • PixInsight Jedi Council Member
  • *******
  • Posts: 2727
    • http://www.carpephoton.com
Re: DBE question
« Reply #4 on: 2010 September 25 09:42:02 »
Glad you got it working Dave :)
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 Niall Saunders

  • PTeam Member
  • PixInsight Jedi Knight
  • *****
  • Posts: 1456
  • We have cookies? Where ?
Re: DBE question
« Reply #5 on: 2010 September 25 15:47:56 »
Hi Dave,

You will also notice that the 'background' image that is 'created' by DBE (the image that will subsequently be 'subtracted' from the original) is usually presented as a 'half-size' image. Don;t worry about this - this is just a 'space-saving' feature of PI - whereby PI knows that it does not need a 'full-resolution' image in order to be able to perform the DBE, so it doesn't bother creating one - and, it also knows how to 'subtract' a half-size image from a full-size image. Clever, eh!!

It can also be very useful to 'hold on' to the 'background image' - I do, even if only briefly. I like to throw a strong STF onto the 'background' just to 'see' what PI is actually subtracting. I like to make sure that there is little (in reality, nothing) in the way of 'real structure', and I also like to see just what the gradients looked like.

Sometimes I will 'double STF' the background image.

I do this by selecting the background image first - the 'title bar' goes 'blue' to confirm this
Then I select the STF window (mine is ALWAYS open, usually on my RHS auxiliary monitor, or resized right along the bottom of the main window)
I then make sure the <Track View> tick mark is 'active' on the STF window (bottom RH corner - it should be 'green')
Now, any changes made in the STF window will automatically 'update' the screen view of the (currently 'active') DBE 'background 'image.
Once I have this image looking the way I want it to look I click on my HistogramTransform window (which, in my case, is 'always open' on my RHS auxiliary monitor)
I make sure that it also has the <Track View> tick mark checked 'green'
Now, I drag the blue triangle <New Instance> icon from the STF window over to the bottom bar of the Histo window
When the floating icon eventually has the 'cross' change to a 'box', I know that I can 'drop' it on the Histo window, and the Histo process 'learns' the slider positions from the STF process
Note that the actual image has still NOT yet been changed permanently - all we are seeing is the 'screen view' transformation as a result of the applied STF
However, I can now click the <Apply (F5)> icon, the little blue square, on the HISTO WINDOW - and the copied STF parameters will be used to CHANGE the actual image data
(Remember, because the Histo process is running in the <Track View> mode, it is FAR EASIER to click the little blue box (or hit the F5 key) rather than dragging the blue triangle all over the screen)
At this point the image will look 'horrible', because it now effectively has the STF transform 'applied twice' - once by the Histo process (where ACTUAL data has changed), and then again by the STF process itself.
I will usually then <Reset> the STF sliders, and then give them a second 'tweak' to help me better visualise the Background Gradients that have been extracted

Obviously the above process is equally as applicable if you just want to quickly set the Histo sliders themselves - for any image. Just set the sliders in STF first, and then 'copy over' the STF setup to the Histo box, most likely then applying final adjustments in the Histo box, before applying the changes to the image.

And, perhaps less intuitive, is the fact that the 'copy over' process is reversible. You can set the sliders in the Histo windo, and 'copy over' the Histo setup to the STF process - which allows the transform to be 'soft' (i.e. only affecting the 'visualisation' of the image data.

Basically, I copy slider positions back and forth between STF and Histo quite a lot as I work through my processing workflow.

And, remember to make good use of the <Track View> check-box, as well as using the 'blue square' or <Apply (F5)> icon to save the hassle of dragging the <New Instance> (blue triangle) to an active image - and these two shortcut suggestions apply to 'most' PixInsight process dialogues.

Hope this helps.
Cheers,
Niall Saunders
Clinterty Observatories
Aberdeen, UK

Altair Astro GSO 10" f/8 Ritchey Chrétien CF OTA on EQ8 mount with homebrew 3D Balance and Pier
Moonfish ED80 APO & Celestron Omni XLT 120
QHY10 CCD & QHY5L-II Colour
9mm TS-OAG and Meade DSI-IIC

Offline Enzo De Bernardini

  • PTeam Member
  • PixInsight Addict
  • ***
  • Posts: 274
  • Resistance is futile.
    • Astronomí­a Sur
Re: DBE question
« Reply #6 on: 2010 September 25 18:09:43 »
I usually check the "Normalize" option for image correction. Is this a recommended procedure? (DSLR images)

Enzo.

Offline Nocturnal

  • PixInsight Jedi Council Member
  • *******
  • Posts: 2727
    • http://www.carpephoton.com
Re: DBE question
« Reply #7 on: 2010 September 25 18:46:50 »
The tooltip indicates what happens when you do. I leave it unchecked as I want the gray background. I can then skip the background neutralization step.
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 Enzo De Bernardini

  • PTeam Member
  • PixInsight Addict
  • ***
  • Posts: 274
  • Resistance is futile.
    • Astronomí­a Sur
Re: DBE question
« Reply #8 on: 2010 September 25 22:02:50 »
Thanks Sander!

Regards,

Enzo.

Offline DaveS

  • PixInsight Addict
  • ***
  • Posts: 115
    • Dave's Astronomy Pages
Re: DBE question
« Reply #9 on: 2010 September 26 04:24:33 »
Hi Niall

That's very helpful information thank you.

I have copied it into my growing PI printed reference folder.

Dave
8" LX200ACF
William Optics FLT110
NEQ6 Mount
SXVF-H9
SXVR-16
SX Lodestar
DMK21AU04
Baader LRGB and NB filters
DiY Observatory
http://www.progressiveastroimaging.com/davesastronomy/

Offline Niall Saunders

  • PTeam Member
  • PixInsight Jedi Knight
  • *****
  • Posts: 1456
  • We have cookies? Where ?
Re: DBE question
« Reply #10 on: 2010 September 26 04:42:17 »
Glad to be of help Dave,

and once you have 'filled' your PI reference folder, perhaps it should be released as the PI User's Manual ::)

Just kidding - I reckon most of us have started that way - I no longer 'print' so much (though sometimes I do), but I do (try to) use the 'bookmark' facility of my browser to capture threads here on the Forum that I suspect I will want to refer back to.

In any case, the more time you spend with PI the 'more natural' the interface becomes. Many of the tasks become almost 'trivial' in the way that you use and apply them, leaving you free to spend more time getting to know the constant list of 'new' processes and features!
Cheers,
Niall Saunders
Clinterty Observatories
Aberdeen, UK

Altair Astro GSO 10" f/8 Ritchey Chrétien CF OTA on EQ8 mount with homebrew 3D Balance and Pier
Moonfish ED80 APO & Celestron Omni XLT 120
QHY10 CCD & QHY5L-II Colour
9mm TS-OAG and Meade DSI-IIC

Offline DaveS

  • PixInsight Addict
  • ***
  • Posts: 115
    • Dave's Astronomy Pages
Re: DBE question
« Reply #11 on: 2010 September 26 06:37:34 »
Niall,

I guess the best way to describe PixIsight is "different", but I am now making good progress, with the help of you guys, video tutorials, and other sources.

I do very much like the DBE function, but then find star reduction easier and more effective in Photoshop. Of course that may well be down to my lack of experience with PI.

I have taken some old unprocessed (other than aligned & combined) image data, and processed them in both PI and PS, and have to say that at this stage the result are little if at all different. Then again I'm am very familiar with PS, which may not make it a fair contest at this time.

On our PAIG forum we are running a PI v PS 'thread', and one of our members has provided (on his website) an unprocessed image for members to download and process, with either PI or PS and 'post' the results on the forum. This is proving very interesting, and very in-keeping with the ethos of the forum.

Dave
8" LX200ACF
William Optics FLT110
NEQ6 Mount
SXVF-H9
SXVR-16
SX Lodestar
DMK21AU04
Baader LRGB and NB filters
DiY Observatory
http://www.progressiveastroimaging.com/davesastronomy/

Offline Niall Saunders

  • PTeam Member
  • PixInsight Jedi Knight
  • *****
  • Posts: 1456
  • We have cookies? Where ?
Re: DBE question
« Reply #12 on: 2010 September 26 08:56:14 »
Hi Dave,

Don't get me wrong - I  know that there are methods available in PS that are not available in PI, and PI may yet develop to provide further methods where these are identified as being essential.

I am no 'master' of either PI or PS, I just feel 'more comfortable' working with PI - for astro image processing. If I need to process images other than astro images, I actually turn to Corel Paint - even though I also have Photoshop available - and that is because I spent more time 'growing up' with Corel.

However, it is ALWAYS interesting to take up the challenge - as you are apparently doing on your other Forum - of trying to use BOTH resources on the same data. Of course, the results are then probably more dependent on the skill of the individual 'user' rather than the capabilities of the software.

Keep us informed of the outcome.
Cheers,
Niall Saunders
Clinterty Observatories
Aberdeen, UK

Altair Astro GSO 10" f/8 Ritchey Chrétien CF OTA on EQ8 mount with homebrew 3D Balance and Pier
Moonfish ED80 APO & Celestron Omni XLT 120
QHY10 CCD & QHY5L-II Colour
9mm TS-OAG and Meade DSI-IIC

Offline DaveS

  • PixInsight Addict
  • ***
  • Posts: 115
    • Dave's Astronomy Pages
Re: DBE question
« Reply #13 on: 2010 September 26 09:36:06 »
Of course, the results are then probably more dependent on the skill of the individual 'user' rather than the capabilities of the software.

That is a signifcant variable Niall, and will undoubtedly have an impact on the results. However, we do have one very competent PI user, and another that is well on the way to becoming so. Competent PS users we have many.

As with most processing algorithms, the settings entered by the user will have a major impact on the result, so the process of comparing images produced by two different software packages, is not exactly 'black and white', as there are some very grey areas (excuse the pun  :laugh:).

However, I'm sure that we will arrive at some reasonable consensus of opinion, and if at this early stage I had to place a bet on it, I think it likely that the result will very close, and perhaps even a 'draw'.

Personally, I'd like to find PI extracted more from my images than PS, as improvement is what we all strive to achieve.

I will indeed keep this forum informed of what we find.

Dave
8" LX200ACF
William Optics FLT110
NEQ6 Mount
SXVF-H9
SXVR-16
SX Lodestar
DMK21AU04
Baader LRGB and NB filters
DiY Observatory
http://www.progressiveastroimaging.com/davesastronomy/

Offline Niall Saunders

  • PTeam Member
  • PixInsight Jedi Knight
  • *****
  • Posts: 1456
  • We have cookies? Where ?
Re: DBE question
« Reply #14 on: 2010 September 26 10:00:57 »
Quote
as improvement is what we all strive to achieve

Agreed there Dave - however, 'improvement' is not just in the final result, but also in the processing required to achieve it.

That was the main reason for my initial move towards PI (the original LE version at the time) - I found that PI gave me FAR more 'control' over my images than ANYTHING else I could find on the market (again, at the time), even though it did seem that everybody and their grandmother was forcing PS to work on the data.

I have heard some well-known PS users say that PI is worth the cost of the package for DBE alone - which was exactly where your thread started ;)
Cheers,
Niall Saunders
Clinterty Observatories
Aberdeen, UK

Altair Astro GSO 10" f/8 Ritchey Chrétien CF OTA on EQ8 mount with homebrew 3D Balance and Pier
Moonfish ED80 APO & Celestron Omni XLT 120
QHY10 CCD & QHY5L-II Colour
9mm TS-OAG and Meade DSI-IIC