m42 with nasty uneven illumination

[Atreta]

Hi everybody,

yesterday i managed to get about 50 minutes total time of the orion nebula, but i can't process it properly because of vignetting and i guess uneven field illumination.

i'd like if someone could help with tips to get rid of it or reduce it for me so i can process it. i tried gradient exterminator but the results weren'r that good with this one.

this is the autosaved tif just after it was stacked with dss:

Autosave.tif

 

thank you

[wimvb]

Here's my effort

Cropped to remove stacking artefacts

DBE applied to remove vignetting

CanonBandingReduction to remove band at bottom

Stretching and HDRMultiscaleTransform to avoid washed out core

Original (just stretched to show it)

Processed

[wimvb]

BTW, here's the file in 16 bit tif with just the cropping and dbe applied. (95 MB)

Autosave_1__DBE1_DBE.tif

[Atreta]

44 minutes ago, wimvb said:

BTW, here's the file in 16 bit tif with just the cropping and dbe applied. (95 MB)

Autosave_1__DBE1_DBE.tif

thanks a lot for your time, Wim. 

it looks very good the process you did. i'll download right now to give it a go :D.

 

[Atreta]

This is my go after Wim kindly did the background extraction for me.

I'm very happy with it so far, gonna try to add some more exposures this weekend.

Please tell your thoughts about it.

46x70s subs iso 800

no calibration frames.

initial stretch  with DSS and later processed in photoshop.

 

thanks again, Wim for the help.

[cjdawson]

Do some short exposures to bring out the trapezium.

to help with the vignetting try taking a set of flats, it's what they for

[Atreta]

2 minutes ago, cjdawson said:

Do some short exposures to bring out the trapezium.

to help with the vignetting try taking a set of flats, it's what they for

 

thanks, Cjdawson

I was going to take some short ones yesterday but forgot to take it .

Flats i don't know why i never tried to make it, it'll be the next thing i'm going to try as it looks like to help a lot.

 

[wimvb]

6 hours ago, cjdawson said:

Do some short exposures to bring out the trapezium.

to help with the vignetting try taking a set of flats, it's what they for

Not necessary to take new subs for the core as it is not overexposed.

Simply copy the unstretched original. Stretch it to the level you want the core to be. Then blend with the image you have so far. It's all about getting the mask and blending right.

Flats are a must. Either point your rig at a flat, evenly illuminated surface to take them, or put a white cloth over your scope/lens and evenly illuminate that.

[Astrosurf]

That's a lovely shot! Oh boy, for the use of DBE. PI is too out of my price range yet.

Just in case you're not familiar with masks, here's some instruction:

http://www.astropix.com/HTML/J_DIGIT/LAYMASK.HTM

Alexxx

[uhb1966]

I like it a lot

[Guy Wells]

[ollypenrice]

You have to be careful with markers in DBE. I suspect Wim's excellent effort had one or two too many between the Trapezium and the Running Man, creating an overly dark artefact between the two. But for the rest it worked well. Once you've tried an intitial DBE it is vital to look at the extracted map which PI offers you. If, in the map, you see anything that even vaguely resembles the target you should scrap it. It has picked up target when it was looking for gradient. You must keep the target out of the gradient map. This is one of the reasons why I often use DBE with only a handful of hand placed markers. They must also be placed well away from the glow of any bright stars. If they pick up this glow, or any object glow, they will instruct the algorithm to darken that region.

When DBE has picked up object glow it will create a dark region around it. If you have imaged a galaxy beware of markers anywhere near it. Galaxies glow out a long way from their bright and easily visible parts.

Olly

[wimvb]

I agree on this, Olly. Sample placement is the most critical step in DBE. Although this target may seem easy, it is actually extremely difficult to judge where nebulosity ends, and the gradient begins. I feel a revisit coming on ...

[ollypenrice]

M42 is probably the hardest target in the sky. Popularity does not bring simplicity!

Olly

[wimvb]

Sample placement. More samples where the gradient is at its worst. No stars covered, stayed away from any nebula.

The background, typical for vignetting problems, but looks clean

The result of DBE. Note that the band at the bottom was already visible in the original, PI autostretch just enhanced it.

The bottom right corner is also slightly darker. It appears that DBE didn't clean it all up.

The quest continues:

CanonBandingReduction (85 % strength) can remove the band, but it creates some weak gradients in the upper half of the image. These are cleaned up with DBE

The final result

Pretty close, I think.

 

BTW, the core is only slightly over exposed. No need to reshoot, in my opinion, unless you're picky. This is the final image, unstretched.

[wimvb]

And here's the tif (188 MB)

Atreta_DBE_2nd.tif

BTW, Frank:

1. flats will really help improve your images

2. the streaks in the background can be avoided by dithering: at least 12 pixels according to the experts. I usually round that up to 15.

3. you might want to investigate the origin of the light band at the bottom of the image. I have no idea where it can come from, but it's unlikely that it's light pollution.

4. there is some uneven coma in your image, more at the right than at the left. I assume that you have a coma corrector for your scope? Since the vignetting also was worse on the right than on the left, you may have some misalignment. (That's just a guess)

[Astrosurf]

Brilliant, Wim!

[wimvb]

Thanks, Alex

 

Btw, now that I see the final image on my mobile phone, it still looks uneven. But processing the image works quite well, despite the uneven background.

[alacant]

On 8/9/2016 at 15:04, Astrosurf said:

for the use of DBE. PI is too out of my price range yet.

Hi. StarTools has it with many other options for a fraction of the price. HTH

[Atreta]

On 9/9/2016 at 18:07, wimvb said:

And here's the tif (188 MB)

Atreta_DBE_2nd.tif

BTW, Frank:

1. flats will really help improve your images

2. the streaks in the background can be avoided by dithering: at least 12 pixels according to the experts. I usually round that up to 15.

3. you might want to investigate the origin of the light band at the bottom of the image. I have no idea where it can come from, but it's unlikely that it's light pollution.

4. there is some uneven coma in your image, more at the right than at the left. I assume that you have a coma corrector for your scope? Since the vignetting also was worse on the right than on the left, you may have some misalignment. (That's just a guess)

Sorry for the time to reply.  

I have a CC but haven't used it yet because i don't have the spacers to get it the distance.  I guess it's certainly misalignment.  

Could that banding be from power lines?  There's one just in front of ny porch and in front of the area where m42 was. 

Thanks for the new file,  I'll try to reprocess it and post in here, gonna try the technique presented in the page that Alex posted earlier. 

[Atreta]

On 9/8/2016 at 10:04, Astrosurf said:

That's a lovely shot! Oh boy, for the use of DBE. PI is too out of my price range yet.

Just in case you're not familiar with masks, here's some instruction:

http://www.astropix.com/HTML/J_DIGIT/LAYMASK.HTM

Alexxx

Thanks,  Alex.  Will try it

On 9/9/2016 at 13:07, uhb1966 said:

I like it a lot

Thanks. 

[Atreta]

On 9/9/2016 at 13:51, Guy Wells said:

Very nice.  Thanks. 

[wimvb]

5 hours ago, Atreta said:

Sorry for the time to reply.  

I have a CC but haven't used it yet because i don't have the spacers to get it the distance.  I guess it's certainly misalignment.  

Could that banding be from power lines?  There's one just in front of ny porch and in front of the area where m42 was. 

Thanks for the new file,  I'll try to reprocess it and post in here, gonna try the technique presented in the page that Alex posted earlier. 

That's very likely. It may be difficult to move your rig, to keep it out of view.

Good luck with reprocessing

[Herzy]

I couldn't seem to remove the banding on the bottom, but I got rid of the gradients pretty well. That is some nice data. This was the first object i successfully imaged, and I only managed 20 minutes. That was when I first got my setup. Now my average integration is 3-4 hours, so I can't wait to see what it will look like once I have some REAL data on it rather then a measly 20 minutes. I would strongly advise you to go for this again (probably when it is higher in the sky later in the year) and to get several hours. You'd be surprised at the amount of faint nebulosity that is there. Good luck!

[Atreta]

On 9/12/2016 at 22:44, Herzy said:

I would strongly advise you to go for this again (probably when it is higher in the sky later in the year) and to get several hours. You'd be surprised at the amount of faint nebulosity that is there. Good luck!

 

thanks, i have some more m42 data, but it was from previous attempts and camera orientation is different, is it worth to try stacking with this one?

nice processing, how did you remove the gradients?