Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

Manual stacking in Pixinsight quick guide instructions

  • Please log in to reply
9 replies to this topic

#1 Becomart

Becomart

    Viking 1

  • -----
  • topic starter
  • Posts: 967
  • Joined: 03 Jan 2015

Posted 20 April 2019 - 04:59 PM

I’m one of those guys who continually forgets all the settings and switches in Pixinsight. This has proven particularly frustrating when pre-processing as it’s taken hours getting mono images ready for any kind of processing. I found it useful to write my own summary sheet and hopefully this will be useful to others. A credit to light vortex website on which this notes are based. 

 

Pixinsight Quick Guide

Calibration and Stacking

Flats
Image calibration process - add files - tick master bias box and link to image - uncheck master dark (unless you are using one) - set output file and create a folder called flats_lum (or other relevant) - disable ‘optimize’ unless using a dark and ‘calibrate’ - apply global

Image intergration process - add files from calibration process - in image integration choose ‘average’ for combination - ‘multiplicative’ for normalization - ‘don’t care’ for weights - leave evaluate noise enabled - in rejection algorithm ‘percentile clipping’ - ‘equalize fluxes’ - tick ‘clip low/high pixels’ - apply global

 

Light Frames
Blink Process - Add files - click on ‘apply an automatic histogram transformation to all images’ and ‘compute autoSTF’ - look through images and lose any with poor focus or extreme light pollution

 

Image calibration process - add light frames - create output folder called lum_cal (or other relevant) - tick master bias and locate - tick master flat and locate - leave ‘calibrate’ and ‘optimize’ unchecked for all masters - apply global

 

Cosmetic Correction Process - Add files from calibrated folder created above - create output folder called lum_cal_cc (or other relevant) - leave master dark unchecked - click on ‘use auto detect’ - double click one frame and apply a stretch - create a preview of smaller area using alt and ‘n’ - click real time preview - use sliders (default of 3 usually works fine) - apply global

 

Subframe Selector - add files - click ‘measure’ - eccentricity should be around 0.60 for good subs, FWHM values=the lower the better - SNRweight=the higher the better - you can sort by ascending to make it easier to see - open best exposure  and check for artefacts or satellite trials as this will be the reference image

 

Star Alignment Process - Add files from the lum_cal_cc (or other relevant) folder - choose appropriate reference image from subframe selector - set output directory and create folder called ‘lum_staralignment’ (or other) - click ‘generate drizzle data’ - apply global

 

Local Normalisation Process - Add reference image from subframe selector - Add files from star alignment folder - set output files to same folder as star alignment saved files - leave scale set to ‘128’ - apply global

 

Image Integration Process -  Add files from the star alignment folder - Add the drizzle files from the same folder - Add the local normalization files from the same folder - select ‘average’ in combination - select ‘local normalization’ in ‘normalization’, select ‘noise evaluation’ in weights - click ‘evaluate noise’ - click ‘generate drizzle data’ - in rejection algorithm choose ‘linear fit clipping’ (or ‘winsorized sigma clipping for 10-20 frames) - select ‘local normalization’ in normalization - enable clip low/high pixels (uncheck ‘range’ options) - apply global - don’t need to save this if proceeding to the next step

 

Drizzle Integration Process - Add drizzle integration files in star alignment folder- Add the local normalization files from the star alignment process - apply global


  • Ballyhoo and 42itous1 like this

#2 Ballyhoo

Ballyhoo

    Gemini

  • *****
  • Posts: 3491
  • Joined: 22 Jul 2011
  • Loc: San Diego

Posted 20 April 2019 - 06:00 PM

I am still not quite sure about the star align tool.  If one calibrates then integrates, why would one need the star align?  seems like my images were stacked well without.



#3 Becomart

Becomart

    Viking 1

  • -----
  • topic starter
  • Posts: 967
  • Joined: 03 Jan 2015

Posted 20 April 2019 - 06:12 PM

Without star alignment, your images won’t be aligned to each other. Your subs might be very short and therefore you may not see a noticeable difference. 


  • psandelle likes this

#4 fmeschia

fmeschia

    Apollo

  • *****
  • Posts: 1015
  • Joined: 20 May 2016
  • Loc: Mountain View, CA

Posted 20 April 2019 - 06:18 PM

Also dithering is usually a good idea, which means alignment is a necessity.

Francesco


  • psandelle likes this

#5 Ballyhoo

Ballyhoo

    Gemini

  • *****
  • Posts: 3491
  • Joined: 22 Jul 2011
  • Loc: San Diego

Posted 20 April 2019 - 10:56 PM

Without star alignment, your images won’t be aligned to each other. Your subs might be very short and therefore you may not see a noticeable difference. 

Now that I think about this, I had some bizzare alignment issues on some instances. Maybe it was that [lack of]star align. THANKS!



#6 Ballyhoo

Ballyhoo

    Gemini

  • *****
  • Posts: 3491
  • Joined: 22 Jul 2011
  • Loc: San Diego

Posted 20 April 2019 - 10:57 PM

I’m one of those guys who continually forgets all the settings and switches in Pixinsight. This has proven particularly frustrating when pre-processing as it’s taken hours getting mono images ready for any kind of processing. I found it useful to write my own summary sheet and hopefully this will be useful to others. A credit to light vortex website on which this notes are based. 

 

Pixinsight Quick Guide

Calibration and Stacking

Flats
Image calibration process - add files - tick master bias box and link to image - uncheck master dark (unless you are using one) - set output file and create a folder called flats_lum (or other relevant) - disable ‘optimize’ unless using a dark and ‘calibrate’ - apply global

Image intergration process - add files from calibration process - in image integration choose ‘average’ for combination - ‘multiplicative’ for normalization - ‘don’t care’ for weights - leave evaluate noise enabled - in rejection algorithm ‘percentile clipping’ - ‘equalize fluxes’ - tick ‘clip low/high pixels’ - apply global

 

Light Frames
Blink Process - Add files - click on ‘apply an automatic histogram transformation to all images’ and ‘compute autoSTF’ - look through images and lose any with poor focus or extreme light pollution

 

Image calibration process - add light frames - create output folder called lum_cal (or other relevant) - tick master bias and locate - tick master flat and locate - leave ‘calibrate’ and ‘optimize’ unchecked for all masters - apply global

 

Cosmetic Correction Process - Add files from calibrated folder created above - create output folder called lum_cal_cc (or other relevant) - leave master dark unchecked - click on ‘use auto detect’ - double click one frame and apply a stretch - create a preview of smaller area using alt and ‘n’ - click real time preview - use sliders (default of 3 usually works fine) - apply global

 

Subframe Selector - add files - click ‘measure’ - eccentricity should be around 0.60 for good subs, FWHM values=the lower the better - SNRweight=the higher the better - you can sort by ascending to make it easier to see - open best exposure  and check for artefacts or satellite trials as this will be the reference image

 

Star Alignment Process - Add files from the lum_cal_cc (or other relevant) folder - choose appropriate reference image from subframe selector - set output directory and create folder called ‘lum_staralignment’ (or other) - click ‘generate drizzle data’ - apply global

 

Local Normalisation Process - Add reference image from subframe selector - Add files from star alignment folder - set output files to same folder as star alignment saved files - leave scale set to ‘128’ - apply global

 

Image Integration Process -  Add files from the star alignment folder - Add the drizzle files from the same folder - Add the local normalization files from the same folder - select ‘average’ in combination - select ‘local normalization’ in ‘normalization’, select ‘noise evaluation’ in weights - click ‘evaluate noise’ - click ‘generate drizzle data’ - in rejection algorithm choose ‘linear fit clipping’ (or ‘winsorized sigma clipping for 10-20 frames) - select ‘local normalization’ in normalization - enable clip low/high pixels (uncheck ‘range’ options) - apply global - don’t need to save this if proceeding to the next step

 

Drizzle Integration Process - Add drizzle integration files in star alignment folder- Add the local normalization files from the star alignment process - apply global

I think this post may be worthy of pinning.



#7 Stelios

Stelios

    Fly Me to the Moon

  • *****
  • Moderators
  • Posts: 7346
  • Joined: 04 Oct 2003
  • Loc: West Hills, CA

Posted 21 April 2019 - 01:32 AM

In addition to dithering and drift from imperfect PA, star align is also crucial to match images from different sessions. I tend to like to image my targets at high elevation, and image for at least 5-6 hours per target (more for NB). This means that I often have to image over multiple nights. For the Sculptor galaxy which rarely gets above 28 degrees here, I had to do four sessions. 

 

DSS incidentally will automatically star align, which is probably why you think you don't need it. 


  • psandelle likes this

#8 Stelios

Stelios

    Fly Me to the Moon

  • *****
  • Moderators
  • Posts: 7346
  • Joined: 04 Oct 2003
  • Loc: West Hills, CA

Posted 21 April 2019 - 01:41 AM

Incidentally, as always with PI, there's more than one way of doing things. Mine is very different. For example, I don't use Blink--SS will catch anything I would've noticed in Blink, and I take enough images that I don't need to worry about excluding images with satellites and airplanes, the software will take what it needs. I also don't drizzle as my image scale doesn't lend itself to it. 

 

My sequence is:

Original directory => Subframe Selector => "Selected" directory

 

"Selected" directory + Flats/Bias/Dark masters => Image calibration => "Calibrated" directory

 

"Calibrated" directory => Cosmetic Correction => "Corrected" directory

 

"Corrected" directory => Star align => "Aligned" Directory

 

"Aligned" Directory => Image Integration => Masters by filter

 

Masters by Filter => Dynamic Crop => Cropped masters

 

... rest follows (ABE/DBE, Deconvolution, Denoise, Delinearization, etc, etc, etc.)



#9 Ballyhoo

Ballyhoo

    Gemini

  • *****
  • Posts: 3491
  • Joined: 22 Jul 2011
  • Loc: San Diego

Posted 21 April 2019 - 02:36 AM

In addition to dithering and drift from imperfect PA, star align is also crucial to match images from different sessions. I tend to like to image my targets at high elevation, and image for at least 5-6 hours per target (more for NB). This means that I often have to image over multiple nights. For the Sculptor galaxy which rarely gets above 28 degrees here, I had to do four sessions. 

 

DSS incidentally will automatically star align, which is probably why you think you don't need it. 

However I have processed images from PI where I went from, I believe it was the script, batch-preprocessing, then to the process, image integration, and the images all appear stacked. But I tried on other occasions and my final image was outtta whack. I want to get all my processing done within PI and not use DSS, just for uniformity sake. So I better integrate the star align tool. 



#10 Stelios

Stelios

    Fly Me to the Moon

  • *****
  • Moderators
  • Posts: 7346
  • Joined: 04 Oct 2003
  • Loc: West Hills, CA

Posted 21 April 2019 - 12:55 PM

If you use BPP, I believe it does SA. I don't use BPP. 


  • 42itous1 likes this


CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics