Blotchy background

[Eddy Timmermans]

Hi all,
I recently applied Kayron's tutorial to some pics I had made of M52.
I was very careful not to overdo stretching and nise reduction.
But apparently I cannot succeed in making the background smooth.
There are always blotches.
What do I need to look for in order to avoid that ?
I used masks and Atrouswavelets and MMT and then ACDNR, but it doesn't seem to work.
Other people who use nebulosity and PS, seem to succeed.
Any advice ?
Eddy

[Alejandro Tombolini]

Hi Eddy, if I understand correctly your problem This could be useful.
Saludos, Alejandro

[Eddy Timmermans]

Hi Alejandro,

No, there is not really color blotching.
I included a crop.
I'm working onthe Luminance extracted from an DSLR shot.

I tried not overdoing the stretching of the histogram. But apparently that didn't work.

Any advice would be welcome
Eddy

[Eddy Timmermans]

I tried again but this time a much less agressive histrogram adjustment.
Kayron warns in his tutorial, not to adjust the shadow side of the histogram so much as to make the background blotchy.
This time it seems to work
Eddy

[tgeib]

Have you tried pixel math?  iif($T<med($T),med($T),&T)

Wish I could remember who to give credit to for this.  It's good for removing very dark areas from a background.

Clear skies,

Tim Geib

[Geoff]

Interesting.  I've never seen that before.
BTW it should read iif($T<med($T),med($T),$T).
Geoff

[lucchett]

Hi,
it seems a way too aggressive noise reduction.
what happens if you jump this step?

Also, your low clipping point could be too precise.
I have found that several tools slightly change the clipping point.
I prefer to leave some space and adjust the backgrund at the very end of the process.

PS: i think it depends of SNR of your frames , but the standard values for the STF are not working well for me. they always force me to clip data in the processing later. I prefer to increase a bit the target median and decrease the clipping point bfrom -2.8 to about -3.5 (but every image is different in this regard). if you copy the STF stats to the HT you may want to check this.

Andrea

[Eddy Timmermans]

Andrea,

Are you referring to the procedure or the Pixel Math formula ?
Because the procedure I followed uses masks and insists on the fact that it is noise reduction and not noise eliminatiion that is the goal.
I noticed however that the darkening of the background, using Histogram Transformation, needs to be done very carefully in order to avoid a blotchy background.
So I did it a second time, and this time the background remained quite even.
I'm using a stack of way to few frames, so the noise is quite strong. I need to control the urge to try and remove as much as possible.
I need more frames for that.
I just wanted to try out the procedure, and it seems to work pretty well.
I also tried his manual stacking procedure instead of the BatchPreProcessing tool. A lot of work but apparently a better result.
I find this trial and error very enriching and not tediuous at all.
Eddy

[lucchett]

Hi Eddy,
I was not referring to the pixel math formulas.
I was trying to point out possible root causes for the blackening , but I think you came to the same conclusions.

also wanted to ensure that you know you can tweak the STF PARAMETERS so that you can have a weaker stretch, but based on statistical properties and repeatable.

Ciao
Andrea

[jkmorse]

Eddy,

Check out this post: http://pixinsight.com/forum/index.php?topic=8618.msg55913#msg55913

In it Mike Smythers has some great scripts and one includes a background smoothing routine that is fantastic.  It uses the Morphological Transformation tool to great effect.  I like it so much I have it saved in my Process Container and use it regularly.  It helped me solve problems on a number of images where I was having the same blotchy background problems. 

Best,

Jim