Please help ! I'm halfway there !

[boris sputnik]

I am working on a picture of the pelican nebula taken recently. So far I did the Batch Processing/integration/calibration/Batch debayer and image integration...Histogram Transformation/ACDNR/Colour Calibration.
Now I don't know where to go next !  There is a lot of white pixels all over the picture ( Noise ) ,and the colour is way off...
Note that there are no darks in the picture ,but I did a preprocess in DSS than a Histogram Transformation/Debayer/Histogram Transformation , and it turned out quite nice.
http://i1355.photobucket.com/albums/q713/BorisSputnik/Astrophotography/500x605_Peacutelican_DSS_18mai2013_1600iso_26x5min_200bias_30flats_CLS_stellarvue_RGB_VNGPixinsight_zps82a0b857.jpg

 But now I need to do everything on Pixinsight.
So this is the whole picture where I'm at now
https://www.dropbox.com/s/c46zfz8so8oa94h/integration_clone%20p%C3%A9lican%202.fit?v=0rc-s

Thank you for helping me !

[Philip de Louraille]

I do believe that the ACDNR and color calibration ought to have been done before the histogram stretch.

[astropixel]

Colour calibration can be off without proper calibration. I would take some darks around the same temperature and give that a go, along with bias and flats.

Pepper and salt (light and dark spots/pixels) may also be a calibration issue.

DBE or ABE before CC. Anytime I mention ABE on this forum I get a slap on the wrist. But honestly for trial runs its quick and easy.

EDIT: Boris. You can probably fix the issues with clever processing. But you will be working at it for hours - frustration. I would go back a few steps and get those darks and bias and flats (if you can) and go from there.

Personally, I don't think results from DSS are a good measure for comparison. With PI you really get to see your data and how imporant calibration is to the steps that follow. Your image will be better as a result,

[boris sputnik]

I do believe that the ACDNR and color calibration ought to have been done before the histogram stretch.

You are right ! LOL
I have a hard time learning the processing...but I will get there !
Thanks !
Maxx

[boris sputnik]

Colour calibration can be off without proper calibration. I would take some darks around the same temperature and give that a go, along with bias and flats.

Pepper and salt (light and dark spots/pixels) may also be a calibration issue.

DBE or ABE before CC. Anytime I mention ABE on this forum I get a slap on the wrist. But honestly for trial runs its quick and easy.

EDIT: Boris. You can probably fix the issues with clever processing. But you will be working at it for hours - frustration. I would go back a few steps and get those darks and bias and flats (if you can) and go from there.
I will make those missing darks this week ,and do it all again ! I think it's worth the effort !
Thank you for helping me !
Maxx

Personally, I don't think results from DSS are a good measure for comparison. With PI you really get to see your data and how imporant calibration is to the steps that follow. Your image will be better as a result,

[Harry page]

Hi

Colour calibration is a linear tool , but sorry ACDNR is after stretching 

Harry

[pfile]

yes, i agree - atrous, mmt or tgvdenoise for pre-stretch NR, ACDNR or tgvdenoise for post-stretch. i think greycstoration belongs in the post-stretch camp as well.

[Philip de Louraille]

Harry and Pfile, I'll take your word for it however I have at least two lists of PI tool processing steps where ACDNR is applied before the Histogram stretch. It is applied afterwards as well.

Not to say that the lists are correct of course but I picked them up from PI forums.

Always happier when I learn new stuff!  Thanks,

[Geoff]

Here is an extensive tutorial on ACDNR  http://www.pixinsight.com/tutorials/ACDNR/en.html .  Definitely stretch first then apply ACDNR.
Geoff

[boris sputnik]

Hi

Colour calibration is a linear tool , but sorry ACDNR is after stretching 

Harry

I'm searching for a step by step instructions with a list of all the tools and where they are:linear or not linear...That would help me a lot !!  Have an idea ??
Thanks !
Maxx

[astropixel]

Maxx. Sorry. Looking at the DSS image, it looks worthwhile the extra effort.

[Geoff]

I'm searching for a step by step instructions with a list of all the tools and where they are:linear or not linear...That would help me a lot !!  Have an idea ??
Thanks !
Maxx

Here http://pixinsight.com/forum/index.php?topic=4055.msg28531#msg28531
Geoff

[pfile]

Harry and Pfile, I'll take your word for it however I have at least two lists of PI tool processing steps where ACDNR is applied before the Histogram stretch. It is applied afterwards as well.

Not to say that the lists are correct of course but I picked them up from PI forums.

Always happier when I learn new stuff!  Thanks,

it might work, i don't know. i know that some of the ACDNR tutorials say not to stretch the noise too much or else you create false contrast which the algorithm then tries to preserve. i guess i've never tried it but somewhere along the line i was left with the impression that it's better for nonlinear data. it's definitely next to impossible to create a mask using the built-in ACDNR masking tool when the image is linear, because the sliders are too coarse for that.

rob

[boris sputnik]

I'm searching for a step by step instructions with a list of all the tools and where they are:linear or not linear...That would help me a lot !!  Have an idea ??
Thanks !
Maxx

Here http://pixinsight.com/forum/index.php?topic=4055.msg28531#msg28531
Geoff

Dude That is perfect...Exactly what I was looking for !
Thank you very much !
Maxx