Merging frme sets of different exposure lengths??

[TheMadOne]

Lo all,

I have a question about blending images of different exposure timings.  I have several sets of sub-frames of M42 from my canon 350D taken at different exposure timings and I would like to merge them all together to create one finished image.  I have all of the calibration frames for each of the different sets of varying exposure timings so I can calibrate each of the different timed sets separately, do the cosmetic corrections, then it would be going to registration.  So here are many questions, yes there are more than just one question.

1.)  Once each different set of subframes has been calibrated should the registration be done separately?  Just wondering due to the how the star alignment functions the difference in the star illumination of the different exposure timings would affect star alignment matching up stars.

2.)  If the registration is done separately what would be the best way to align the resulting integrations of each set?  Same issue of varying star illuminations if stack is left as a linear image.

3.)  Now about the actual integration of each set.  I'm thinking pixel math should be used to perform the final stacking of the resulting integrations from each of the different  timing set frame integrations.  My exposure timings are a set of 10 second, 30 second, 60 second,  120 second, & 180 second frames.  Likely do a percentile ratio of merging the frames together?  These would be RGB OSC frames to be merged, any help or suggestions?  I have never performed as series of different timing to one final image before with PIxInsight any help would be greatly appreciated.  On a side note these images are from 2009 I attempted to do this in PS CS4 Extended, but the result was less than satisfactory, it has been my one and only such attempt prior to this.  I was hoping I could achieve better results with PixInsight using PI's better deeper processing capabilities!  Here is the original attempt from those exposures using the PS CS4 Extended.

http://www.thestardeckobservatory.com/Star_Deck_Albums/main.php/v/TheStarDeck/CanonXTAstrophotos/Nebulae/M42_012209_Composite.jpg.html


Thanks,

Mark Jordan

http://www.thestardeckobservatory.com

[pfile]

okay - #1 - no, just register everything to the same reference frame. StarAlignment will figure it all out - there should be no problem with the various exposures
so #2 is not an issue.
for #3, stack each set of exposures, then save the integrations and use the HDRComposition process to merge your exposures together. i think there have been some tutorials posted here regarding that process - try searching. you will probably have to mess with the binarizing threshold, and pay attention to the masks that come out - they describe what pixels were taken from what image.

rob

[TheMadOne]

Thanks Rob,

I give it a whirl & let you know how it goes.  I did the "total" registration routine in one pass on all of the calibrated frames last night, with a couple of minor tweaks it worked perfectly, I decreased the log(sensitivity) slightly and bumped up the Peak response slightly.  I did a regular integration of all of the files at once just to check out the alignment,  the end result showed all of the differently timed sets of sub-frames were aligned perfectly!

Thanks again,

Mark Jordan

http:/www.thestardeckobservatory.com

[TheMadOne]

Ok rob,

Here's the end results of using PixInsight with the HDRComposition!  It went smooth as butter, and got the "best results" with the binarizing threshold set to 75, although there was not a whole lot of noticeable difference between the 75 & the 80 default setting.  I have yet a couple more such multiple timed sets of sub-frames saved back from 2009 & 2010 I never felt like I could process to a satisfactory final image.  I will be reprocessing them in the near future with PixInsight, the more I use it and do with it the more I like PI!

You can get a closer look at the reprocessed version here:

http://www.thestardeckobservatory.com/Star_Deck_Albums/main.php/v/TheStarDeck/CanonXTAstrophotos/Nebulae/M42_012209_Composite.jpg.html

To get to higher resolutions just click on the image in the link.

Mark Jordan

http://www.thestardeckobservatory.com

[pfile]

now that's what i'm talkin about

but how did you get behind the nebula and image it from the other side? 

rob

[TheMadOne]

LOL, I'll never tell, but lets just say I would have never been able to with out PI, I'm loving this application!

Mark Jordan

http://www.thestardeckobservatory.com