Subs appear Useless

carlstronomy (Carl) · #1 10-10-2013, 04:17 PM

Hello to all

I took 3 hours of data on the flame nebula last night. Processed today and found that I have some sort of reflection. I have attached two over stretched images showing the issue.

Image 1: Is all three hours of data.

Image 2: A trial of only 20 minutes of data.

Darks for each, a set of bias and flats (I even took a second lot of flats today to see if it made any difference that was image 2)

Image train.

Celestron C9.25
Meade 6.3 reducer
Bintel T adapter
Canon 450D

I also have a home made due shield which is lined with black material.

This is the same image train used a few nights ago for my M42 which has no issues and they are both in the same area of the sky. I am in a fairly light polluted area but this has not changed in the last week.

Can someone help with what could be causing this issue and how I can either fix these subs or prevent this happening in the future.

Thanks

Carl

lazjen (Chris) · #2 10-10-2013, 04:31 PM

Is the reflection in the same position each time? For example, pick 2 different targets, take sub(s). Check results.

That huge blue star? could be the source of the reflection in your subs in this case, maybe?

rustigsmed (Russell) · #3 10-10-2013, 04:38 PM

I think you can blame the star Alnitak for the glare.
Its a common problem in imaging the Flame. I have no tips i'm afraid, someone else may have some nifty processing tricks to advise you on.

carlstronomy (Carl) · #4 10-10-2013, 04:38 PM

I think the blue star is the issue but why and how do I stop if being an issue. Many people have taken images of this subject and there is no reflection visible in theirs. I have already carried out the process in target changing as the camera has not been removed since my M42 from the other day.

Carl

carlstronomy (Carl) · #5 10-10-2013, 04:41 PM

Russell, thanks I was replying to Chris as you were posting. I assumed this may have been the nasty culprit now I have to see if there is a way around it.

Thanks

multiweb (Marc) · #6 10-10-2013, 05:00 PM

Yes this is a reflection from Alnitak on the left of your field. Couple of things you can do.

1_ offset your field to the right of the flame, then reject the halo with the new subs as it won't be there anymore or in a different position.

2_ Process it out in PS. It's fairly easy to do as it doesn't overlap any nebulosity.

#1 is easier.

carlstronomy (Carl) · #7 10-10-2013, 05:55 PM

Quote:

Originally Posted by multiweb

Yes this is a reflection from Alnitak on the left of your field. Couple of things you can do.

1_ offset your field to the right of the flame, then reject the halo with the new subs as it won't be there anymore or in a different position.

2_ Process it out in PS. It's fairly easy to do as it doesn't overlap any nebulosity.

#1 is easier.

Excellent that should work nicely Marc, can I just asked a couple of questions to ensure I get the process correct.

Step 1. Have the completed image stacked from all my original subs.
Step 2. Take a new set of subs offset to remove Alnitak
Step 3. Stack the one original from step 1 along with the new subs in step 2

I assume then as lets say I have 50 new subs with no halo, and only one with it the halo will be removed?

I think I got it right let me know if I screwed it up

I like option 1 also I hate to PS errors or irregularities from images, I would rather work them out the correct way.

Thanks for your help

Carl

multiweb (Marc) · #8 11-10-2013, 08:48 AM

That's correct. If you shoot the same area offset then the halo won't be in the same spot. So you can reject it by stacking the old stack with the new stack. A little painting in PS for halo removal is not a crime. Here's a quick technique from Jase posted a while ago that might work for you:

Quote:

Click on the lasso tool and set the feathering to approx 20 pixels
Using the lasso tool, select an area that is of same size and is relatively close to the colour flare area you wish to remove. The purpose of this exercise is to match the background hue of the surrounding area so the proximity to the area you're trying to fix is fairly important. At this point, don't worry if the selected area contains stars.
Copy and paste the lasso area so you now have a new layer that contains the small patch of the sky in which you've just copied.
Hold down the control key and move the mouse to shift the small patch over the colour flare area you wish to remove.
Now go to the Filters | Noise | Median
Increase the radius until you see no more stars in the small patch of sky that you've copied. If you don't like the median tool, dust and scratches, or the minimum filters can do a similar job. Median I've found works well in general.
With the patch layer selected, change its blend mode from normal to colour
You'll now notice that your colour flare has disappeared i.e. has been blending into the surrounding background hues. If it hasn't you may need to increase your selection size or reduce the feather.
...but don't stop there...look what you've done to the stars in the area that's been repaired...they've inherited the same background tone you copied. Easily fixed...
Select the original layer and use the colour range tool to highlight the stars. Alter the fuzziness slider until you get a good match.
Expand the selection by 2 or 3 pixels (or to your taste)
Feather by 2 pixels (or to your taste)
Then select the patch layer (important, make sure its selected!!)
Hit the delete key
POW! You've now brought the star colours back in the repaired area...as you've deleted the selections from the patch layer so the colours come back through from the original layer.
If you find that the colour range tool selection wasn't ideal, you can always use the eraser tool to do a similar task, but I would recommend going back a few steps until you get the right selection i.e. manipulate fuzziness.